51
|
Jiang B, Zhu G, Xie Y, Heit JJ, Chen H, Li Y, Ding V, Eskandari A, Michel P, Zaharchuk G, Wintermark M. Prediction of Clinical Outcome in Patients with Large-Vessel Acute Ischemic Stroke: Performance of Machine Learning versus SPAN-100. AJNR Am J Neuroradiol 2021; 42:240-246. [PMID: 33414230 DOI: 10.3174/ajnr.a6918] [Citation(s) in RCA: 12] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/24/2020] [Accepted: 09/12/2020] [Indexed: 01/09/2023]
Abstract
BACKGROUND AND PURPOSE Traditional statistical models and pretreatment scoring systems have been used to predict the outcome for acute ischemic stroke patients (AIS). Our aim was to select the most relevant features in terms of outcome prediction on the basis of machine learning algorithms for patients with acute ischemic stroke and to compare the performance between multiple models and the Stroke Prognostication Using Age and National Institutes of Health Stroke Scale (SPAN-100) index model. MATERIALS AND METHODS A retrospective multicenter cohort of 1431 patients with acute ischemic stroke was subdivided into recanalized and nonrecanalized patients. Extreme Gradient Boosting machine learning models were built to predict the mRS score at 90 days using clinical, imaging, combined, and best-performing features. Feature selection was performed using the relative weight and frequency of occurrence in the models. The model with the best performance was compared with the SPAN-100 index model using area under the receiver operating curve analysis. RESULTS In 3 groups of patients, the baseline NIHSS was the most significant predictor of outcome among all the parameters, with relative weights of 0.36∼0.69; ischemic core volume on CTP ranked as the most important imaging biomarker with relative weights of 0.29∼0.47. The model with the best-performing features had a better performance than the other machine learning models. The area under the curve of the model with the best-performing features was higher than SPAN-100 model and reached statistical significance for the total (P < .05) and the nonrecanalized patients (P < .001). CONCLUSIONS Machine learning-based feature selection can identify parameters with higher performance in outcome prediction. Machine learning models with the best-performing features, especially advanced CTP data, had superior performance of the recovery outcome prediction for patients with stroke at admission in comparison with SPAN-100.
Collapse
Affiliation(s)
- B Jiang
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - G Zhu
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - Y Xie
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - J J Heit
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - H Chen
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - Y Li
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - V Ding
- Department of Medicine (V.D.), Quantitative Sciences Unit, Stanford University, Stanford, California
| | - A Eskandari
- Neurology Service (A.E., P.M.), Centre Hospitalier Universitaire Vaudois and Lausanne University, Lausanne, Switzerland
| | - P Michel
- Neurology Service (A.E., P.M.), Centre Hospitalier Universitaire Vaudois and Lausanne University, Lausanne, Switzerland
| | - G Zaharchuk
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| | - M Wintermark
- From the Department of Radiology, Neuroradiology Section (B.J., G.Z., Y.X., J.J.H., H.C., Y.L., G.Z., M.W.), Stanford University School of Medicine, Palo Alto, California
| |
Collapse
|
52
|
Le A, Salifu MO, McFarlane IM. Artificial Intelligence in Colorectal Polyp Detection and Characterization. INTERNATIONAL JOURNAL OF CLINICAL RESEARCH & TRIALS 2021; 6:157. [PMID: 33884326 PMCID: PMC8057724 DOI: 10.15344/2456-8007/2021/157] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Indexed: 01/16/2023]
Abstract
BACKGROUND Over the past 20 years, the advancement of artificial intelligence (AI) and deep learning (DL) has allowed for fast sorting and analysis of large sets of data. In the field of gastroenterology, colorectal screening procedures produces an abundance of data through video and imaging. With AI and DL, this information can be used to create systems where automatic polyp detection and characterization is possible. Convoluted Neural Networks (CNNs) have proven to be an effective way to increase polyp detection and ultimately adenoma detection rates. Different methods of polyp characterization of being hyperplastic vs. adenomatous or non-neoplastic vs. neoplastic has also been investigated showing promising results. FINDINGS The rate of missed polyps on colonoscopy can be as high as 25%. At the beginning of the 2000s, hand-crafted machine learning (ML) algorithms were created and trained retrospectively on colonoscopy images and videos, achieving high sensitivity, specificity, and accuracy of over 90% in many of the studies. Over time, the advancement of DL and CNNs has allowed algorithms to be trained on non-medical images and applied retrospectively to colonoscopy videos and images with similar results. Within the past few years, these algorithms have been applied in real-time colonoscopies and has shown mixed results, one showing no difference while others showing increased polyp detection.Various methods of polyp characterization have also been investigated. Through AI, DL, and CNNs polyps can be identified has hyperplastic/adenomatous or non-neoplastic/neoplastic with high sensitivity, specificity, and accuracy. One of the research areas in polyp characterization is how to capture the polyp image. This paper looks at different modalities of characterizing polyps such as magnifying narrow band imaging (NBI), endocytoscopy, laser-induced florescent spectroscopy, auto-florescent endoscopy, and white-light endoscopy. CONCLUSIONS Overall, much progress has been made in automatic detection and characterization of polyps in real time. Barring ethical or mass adoption setbacks, it is inevitable that AI will be involved in the field of GI, especially in colorectal polyp detection and identification.
Collapse
Affiliation(s)
| | | | - Isabel M. McFarlane
- Corresponding Author: Dr. Isabel M. McFarlane, Clinical Assistant Professor of Medicine, Director, Third Year Internal Medicine Clerkship, Department of Internal Medicine, Brooklyn, NY 11203, USA Tel: 718-270-2390, Fax: 718-270-1324;
| |
Collapse
|
53
|
Hamann J, Herzog L, Wehrli C, Dobrocky T, Bink A, Piccirelli M, Panos L, Kaesmacher J, Fischer U, Stippich C, Luft AR, Gralla J, Arnold M, Wiest R, Sick B, Wegener S. Machine-learning-based outcome prediction in stroke patients with middle cerebral artery-M1 occlusions and early thrombectomy. Eur J Neurol 2020; 28:1234-1243. [PMID: 33220140 PMCID: PMC7986098 DOI: 10.1111/ene.14651] [Citation(s) in RCA: 19] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/30/2020] [Accepted: 11/16/2020] [Indexed: 01/01/2023]
Abstract
Background and purpose Clinical outcomes vary substantially among individuals with large vessel occlusion (LVO) stroke. A small infarct core and large imaging mismatch were found to be associated with good recovery. The aim of this study was to investigate whether those imaging variables would improve individual prediction of functional outcome after early (<6 h) endovascular treatment (EVT) in LVO stroke. Methods We included 222 patients with acute ischemic stroke due to middle cerebral artery (MCA)‐M1 occlusion who received EVT. As predictors, we used clinical variables and region of interest (ROI)‐based magnetic resonance imaging features. We developed different machine‐learning models and quantified their prediction performance according to the area under the receiver‐operating characteristic curves and the Brier score. Results The rate of successful recanalization was 78%, with 54% patients having a favorable outcome (modified Rankin scale score 0–2). Small infarct core was associated with favorable functional outcome. Outcome prediction improved only slightly when imaging was added to patient variables. Age was the driving factor, with a sharp decrease in likelihood of favorable functional outcome above the age of 78 years. Conclusions In patients with MCA‐M1 occlusion strokes referred to EVT within 6 h of symptom onset, infarct core volume was associated with outcome. However, ROI‐based imaging variables led to no significant improvement in outcome prediction at an individual patient level when added to a set of clinical predictors. Our study is in concordance with current practice, where imaging mismatch or collateral readouts are not recommended as factors for excluding patients with MCA‐M1 occlusion for early EVT.
Collapse
Affiliation(s)
- Janne Hamann
- Department of Neurology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| | - Lisa Herzog
- Department of Neurology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland.,Epidemiology, Biostatistics and Prevention Institute (EBPI), University of Zurich, Zurich, Switzerland.,Institute of Data Analysis and Process Design, ZHAW Winterthur, Winterthur, Switzerland
| | - Carina Wehrli
- Department of Neurology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland.,Department of Neuroradiology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| | - Tomas Dobrocky
- Diagnostic and Interventional Neuroradiology, University Hospital of Berne, Berne, Switzerland
| | - Andrea Bink
- Department of Neuroradiology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| | - Marco Piccirelli
- Department of Neuroradiology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| | - Leonidas Panos
- Department of Neurology, University Hospital of Berne, Berne, Switzerland
| | - Johannes Kaesmacher
- Diagnostic and Interventional Neuroradiology, University Hospital of Berne, Berne, Switzerland.,Department of Diagnostic, Interventional and Pediatric Radiology, University Hospital of Berne, Berne, Switzerland
| | - Urs Fischer
- Department of Neurology, University Hospital of Berne, Berne, Switzerland
| | - Christoph Stippich
- Department of Neuroradiology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| | - Andreas R Luft
- Department of Neurology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| | - Jan Gralla
- Diagnostic and Interventional Neuroradiology, University Hospital of Berne, Berne, Switzerland
| | - Marcel Arnold
- Department of Neurology, University Hospital of Berne, Berne, Switzerland
| | - Roland Wiest
- Diagnostic and Interventional Neuroradiology, University Hospital of Berne, Berne, Switzerland
| | - Beate Sick
- Epidemiology, Biostatistics and Prevention Institute (EBPI), University of Zurich, Zurich, Switzerland.,Institute of Data Analysis and Process Design, ZHAW Winterthur, Winterthur, Switzerland
| | - Susanne Wegener
- Department of Neurology and Clinical Neuroscience Center, University Hospital Zurich and University of Zurich, Zurich, Switzerland
| |
Collapse
|
54
|
Aguiar de Sousa D, Katan M. Promising Use of Automated Electronic Phenotyping: Turning Big Data Into Big Value in Stroke Research. Stroke 2020; 52:190-192. [PMID: 33297867 DOI: 10.1161/strokeaha.120.033061] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Affiliation(s)
- Diana Aguiar de Sousa
- Department of Neurosciences and Mental Health (Neurology), Hospital de Santa Maria-Centro Hospitalar Universitário Lisboa Norte (CHULN), Lisbon, Portugal (D.A.d.S.).,Faculdade de Medicina, Universidade de Lisboa, Lisbon, Portugal (D.A.d.S.)
| | - Mira Katan
- Department of Neurology, University Hospital of Zurich, Switzerland (M.K.).,Neuroscience Center of Zurich, University of Zurich, Switzerland (M.K.)
| |
Collapse
|
55
|
Brugnara G, Neuberger U, Mahmutoglu MA, Foltyn M, Herweh C, Nagel S, Schönenberger S, Heiland S, Ulfert C, Ringleb PA, Bendszus M, Möhlenbruch MA, Pfaff JA, Vollmuth P. Multimodal Predictive Modeling of Endovascular Treatment Outcome for Acute Ischemic Stroke Using Machine-Learning. Stroke 2020; 51:3541-3551. [DOI: 10.1161/strokeaha.120.030287] [Citation(s) in RCA: 42] [Impact Index Per Article: 10.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/26/2022]
Abstract
Background and Purpose:
This study assessed the predictive performance and relative importance of clinical, multimodal imaging, and angiographic characteristics for predicting the clinical outcome of endovascular treatment for acute ischemic stroke.
Methods:
A consecutive series of 246 patients with acute ischemic stroke and large vessel occlusion in the anterior circulation who underwent endovascular treatment between April 2014 and January 2018 was analyzed. Clinical, conventional imaging (electronic Alberta Stroke Program Early CT Score, acute ischemic volume, site of vessel occlusion, and collateral score), and advanced imaging characteristics (CT-perfusion with quantification of ischemic penumbra and infarct core volumes) before treatment as well as angiographic (interval groin puncture-recanalization, modified Thrombolysis in Cerebral Infarction score) and postinterventional clinical (National Institutes of Health Stroke Scale score after 24 hours) and imaging characteristics (electronic Alberta Stroke Program Early CT Score, final infarction volume after 18–36 hours) were assessed. The modified Rankin Scale (mRS) score at 90 days (mRS-90) was used to measure patient outcome (favorable outcome: mRS-90 ≤2 versus unfavorable outcome: mRS-90 >2). Machine-learning with gradient boosting classifiers was used to assess the performance and relative importance of the extracted characteristics for predicting mRS-90.
Results:
Baseline clinical and conventional imaging characteristics predicted mRS-90 with an area under the receiver operating characteristics curve of 0.740 (95% CI, 0.733–0.747) and an accuracy of 0.711 (95% CI, 0.705–0.717). Advanced imaging with CT-perfusion did not improved the predictive performance (area under the receiver operating characteristics curve, 0.747 [95% CI, 0.740–0.755]; accuracy, 0.720 [95% CI, 0.714–0.727];
P
=0.150). Further inclusion of angiographic and postinterventional characteristics significantly improved the predictive performance (area under the receiver operating characteristics curve, 0.856 [95% CI, 0.850–0.861]; accuracy, 0.804 [95% CI, 0.799–0.810];
P
<0.001). The most important parameters for predicting mRS 90 were National Institutes of Health Stroke Scale score after 24 hours (importance =100%), premorbid mRS score (importance =44%) and final infarction volume on postinterventional CT after 18 to 36 hours (importance =32%).
Conclusions:
Integrative assessment of clinical, multimodal imaging, and angiographic characteristics with machine-learning allowed to accurately predict the clinical outcome following endovascular treatment for acute ischemic stroke. Thereby, premorbid mRS was the most important clinical predictor for mRS-90, and the final infarction volume was the most important imaging predictor, while the extent of hemodynamic impairment on CT-perfusion before treatment had limited importance.
Collapse
Affiliation(s)
- Gianluca Brugnara
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Ulf Neuberger
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Mustafa A. Mahmutoglu
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Martha Foltyn
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Christian Herweh
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Simon Nagel
- Neurology Clinic (S.N., S.S., P.A.R.), Heidelberg University Hospital, Germany
| | | | - Sabine Heiland
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Christian Ulfert
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | | | - Martin Bendszus
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Markus A. Möhlenbruch
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Johannes A.R. Pfaff
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| | - Philipp Vollmuth
- Department of Neuroradiology (G.B., U.N., M.A.M., M.F., Ch.H., S.H., C.U., M.B., M.A.M., J.A.R.P., P.V.), Heidelberg University Hospital, Germany
| |
Collapse
|
56
|
Muhammad LJ, Algehyne EA, Usman SS, Ahmad A, Chakraborty C, Mohammed IA. Supervised Machine Learning Models for Prediction of COVID-19 Infection using Epidemiology Dataset. ACTA ACUST UNITED AC 2020; 2:11. [PMID: 33263111 PMCID: PMC7694891 DOI: 10.1007/s42979-020-00394-7] [Citation(s) in RCA: 72] [Impact Index Per Article: 18.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/26/2020] [Accepted: 11/05/2020] [Indexed: 12/15/2022]
Abstract
COVID-19 or 2019-nCoV is no longer pandemic but rather endemic, with more than 651,247 people around world having lost their lives after contracting the disease. Currently, there is no specific treatment or cure for COVID-19, and thus living with the disease and its symptoms is inevitable. This reality has placed a massive burden on limited healthcare systems worldwide especially in the developing nations. Although neither an effective, clinically proven antiviral agents' strategy nor an approved vaccine exist to eradicate the COVID-19 pandemic, there are alternatives that may reduce the huge burden on not only limited healthcare systems but also the economic sector; the most promising include harnessing non-clinical techniques such as machine learning, data mining, deep learning and other artificial intelligence. These alternatives would facilitate diagnosis and prognosis for 2019-nCoV pandemic patients. Supervised machine learning models for COVID-19 infection were developed in this work with learning algorithms which include logistic regression, decision tree, support vector machine, naive Bayes, and artificial neutral network using epidemiology labeled dataset for positive and negative COVID-19 cases of Mexico. The correlation coefficient analysis between various dependent and independent features was carried out to determine a strength relationship between each dependent feature and independent feature of the dataset prior to developing the models. The 80% of the training dataset were used for training the models while the remaining 20% were used for testing the models. The result of the performance evaluation of the models showed that decision tree model has the highest accuracy of 94.99% while the Support Vector Machine Model has the highest sensitivity of 93.34% and Naïve Bayes Model has the highest specificity of 94.30%.
Collapse
Affiliation(s)
- L J Muhammad
- Department of Mathematics and Computer Science, Faculty of Science, Federal University of Kashere, P.M.B. 0182, Gombe, Nigeria
| | - Ebrahem A Algehyne
- Department of Mathematics, University of Tabuk, Tabuk, 71491 Saudi Arabia
| | - Sani Sharif Usman
- Department of Biological Sciences, Faculty of Science, Federal University of Kashere, P.M.B. 0182, Gombe, Nigeria
| | - Abdulkadir Ahmad
- Department of Computer Science, Kano University of Science and Technology, Wudil, Kano Nigeria
| | - Chinmay Chakraborty
- Department of Electronics and Communication Engineering, Birla Institute of Technology, Ranchi, Jharkhand India
| | - I A Mohammed
- Computer Science Department, Yobe StateUniversity, Damaturu, Yobe State Nigeria
| |
Collapse
|
57
|
Ali F, Hamid U, Zaidat O, Bhatti D, Kalia JS. Role of Artificial Intelligence in TeleStroke: An Overview. Front Neurol 2020; 11:559322. [PMID: 33117259 PMCID: PMC7576935 DOI: 10.3389/fneur.2020.559322] [Citation(s) in RCA: 12] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/05/2020] [Accepted: 08/20/2020] [Indexed: 01/01/2023] Open
Abstract
Teleneurology has provided access to neurological expertise and state-of-the-art stroke care where previously they have been inaccessible. The use of Artificial Intelligence with machine learning to assist telestroke care can be revolutionary. This includes more rapid and more reliable diagnosis through imaging analysis as well as prediction of hospital course and 3-month prognosis. Intelligent Electronic Medical Records can search free text and provide decision assistance by analyzing patient charts. Speech recognition has advanced enough to be reliable and highly convenient. Smart contextually aware communication and alert programs can enhance efficiency of patient flow and improve outcomes. Automated data collection and analysis can make quality improvement and research projects quicker and much less burdensome. Despite current challenges, these synergistic technologies hold immense promise in enhancing the clinician experience, helping to reduce physician burnout while improving patient health outcomes at a lower cost. This brief overview discusses the multifaceted potential of AI use in telestroke.
Collapse
Affiliation(s)
- Faryal Ali
- Department of Medicine, Dow University of Health Sciences, Karachi, Pakistan
| | - Umair Hamid
- Department of Neurology, University of Illinois, College of Medicine, Peoria, IL, United States
| | - Osama Zaidat
- Departments of Endovascular Neurosurgery and Stroke, St. Vincent Mercy Medical Center, Toledo, OH, United States
| | - Danish Bhatti
- Department of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE, United States
| | - Junaid Siddiq Kalia
- AINeuroCare, Dallas, TX, United States.,Clinical Strategy, VeeMed Inc., Roseville, CA, United States
| |
Collapse
|
58
|
Osama S, Zafar K, Sadiq MU. Predicting Clinical Outcome in Acute Ischemic Stroke Using Parallel Multi-parametric Feature Embedded Siamese Network. Diagnostics (Basel) 2020; 10:E858. [PMID: 33105609 PMCID: PMC7690444 DOI: 10.3390/diagnostics10110858] [Citation(s) in RCA: 9] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/05/2020] [Revised: 10/16/2020] [Accepted: 10/19/2020] [Indexed: 11/16/2022] Open
Abstract
Stroke is the second leading cause of death and disability worldwide, with ischemic stroke as the most common type. The preferred diagnostic procedure at the acute stage is the acquisition of multi-parametric magnetic resonance imaging (MRI). This type of imaging not only detects and locates the stroke lesion, but also provides the blood flow dynamics that helps clinicians in assessing the risks and benefits of reperfusion therapies. However, evaluating the outcome of these risky therapies beforehand is a complicated task due to the variability of lesion location, size, shape, and cerebral hemodynamics involved. Though the fully automated model for predicting treatment outcomes using multi-parametric imaging would be highly valuable in clinical settings, MRI datasets acquired at the acute stage are mostly scarce and suffer high class imbalance. In this paper, parallel multi-parametric feature embedded siamese network (PMFE-SN) is proposed that can learn with few samples and can handle skewness in multi-parametric MRI data. Moreover, five suitable evaluation metrics that are insensitive to imbalance are defined for this problem. The results show that PMFE-SN not only outperforms other state-of-the-art techniques in all these metrics but also can predict the class with a small number of samples, as well as the class with high number of samples. An accuracy of 0.67 on leave one cross out testing has been achieved with only two samples (minority class) for training and accuracy of 0.61 with the highest number of samples (majority class). In comparison, state-of-the-art using hand crafted features has 0 accuracy for minority class and 0.33 accuracy for majority class.
Collapse
Affiliation(s)
| | - Kashif Zafar
- Department of Computer Science, National University of Computing and Emerging Sciences, 852-B Milaad St, Block B Faisal Town, Lahore 54000, Pakistan; (S.O.); (M.U.S.)
| | | |
Collapse
|
59
|
Song Z, Guo D, Tang Z, Liu H, Li X, Luo S, Yao X, Song W, Song J, Zhou Z. Noncontrast Computed Tomography-Based Radiomics Analysis in Discriminating Early Hematoma Expansion after Spontaneous Intracerebral Hemorrhage. Korean J Radiol 2020; 22:415-424. [PMID: 33169546 PMCID: PMC7909850 DOI: 10.3348/kjr.2020.0254] [Citation(s) in RCA: 32] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/10/2020] [Revised: 05/26/2020] [Accepted: 07/02/2020] [Indexed: 01/05/2023] Open
Abstract
Objective To determine whether noncontrast computed tomography (NCCT) models based on multivariable, radiomics features, and machine learning (ML) algorithms could further improve the discrimination of early hematoma expansion (HE) in patients with spontaneous intracerebral hemorrhage (sICH). Materials and Methods We retrospectively reviewed 261 patients with sICH who underwent initial NCCT within 6 hours of ictus and follow-up CT within 24 hours after initial NCCT, between April 2011 and March 2019. The clinical characteristics, imaging signs and radiomics features extracted from the initial NCCT images were used to construct models to discriminate early HE. A clinical-radiologic model was constructed using a multivariate logistic regression (LR) analysis. Radiomics models, a radiomics-radiologic model, and a combined model were constructed in the training cohort (n = 182) and independently verified in the validation cohort (n = 79). Receiver operating characteristic analysis and the area under the curve (AUC) were used to evaluate the discriminative power. Results The AUC of the clinical-radiologic model for discriminating early HE was 0.766. The AUCs of the radiomics model for discriminating early HE built using the LR algorithm in the training and validation cohorts were 0.926 and 0.850, respectively. The AUCs of the radiomics-radiologic model in the training and validation cohorts were 0.946 and 0.867, respectively. The AUCs of the combined model in the training and validation cohorts were 0.960 and 0.867, respectively. Conclusion NCCT models based on multivariable, radiomics features and ML algorithm could improve the discrimination of early HE. The combined model was the best recommended model to identify sICH patients at risk of early HE.
Collapse
Affiliation(s)
- Zuhua Song
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Dajing Guo
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Zhuoyue Tang
- Department of Radiology, Chongqing General Hospital, Chongqing, China
| | | | - Xin Li
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Sha Luo
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Xueying Yao
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Wenlong Song
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Junjie Song
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Zhiming Zhou
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China.
| |
Collapse
|
60
|
Ramos LA, Kappelhof M, van Os HJA, Chalos V, Van Kranendonk K, Kruyt ND, Roos YBWEM, van der Lugt A, van Zwam WH, van der Schaaf IC, Zwinderman AH, Strijkers GJ, van Walderveen MAA, Wermer MJH, Olabarriaga SD, Majoie CBLM, Marquering HA. Predicting Poor Outcome Before Endovascular Treatment in Patients With Acute Ischemic Stroke. Front Neurol 2020; 11:580957. [PMID: 33178123 PMCID: PMC7593486 DOI: 10.3389/fneur.2020.580957] [Citation(s) in RCA: 19] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/07/2020] [Accepted: 09/07/2020] [Indexed: 12/31/2022] Open
Abstract
Background: Although endovascular treatment (EVT) has greatly improved outcomes in acute ischemic stroke, still one third of patients die or remain severely disabled after stroke. If we could select patients with poor clinical outcome despite EVT, we could prevent futile treatment, avoid treatment complications, and further improve stroke care. We aimed to determine the accuracy of poor functional outcome prediction, defined as 90-day modified Rankin Scale (mRS) score ≥5, despite EVT treatment. Methods: We included 1,526 patients from the MR CLEAN Registry, a prospective, observational, multicenter registry of ischemic stroke patients treated with EVT. We developed machine learning prediction models using all variables available at baseline before treatment. We optimized the models for both maximizing the area under the curve (AUC), reducing the number of false positives. Results: From 1,526 patients included, 480 (31%) of patients showed poor outcome. The highest AUC was 0.81 for random forest. The highest area under the precision recall curve was 0.69 for the support vector machine. The highest achieved specificity was 95% with a sensitivity of 34% for neural networks, indicating that all models contained false positives in their predictions. From 921 mRS 0–4 patients, 27–61 (3–6%) were incorrectly classified as poor outcome. From 480 poor outcome patients in the registry, 99–163 (21–34%) were correctly identified by the models. Conclusions: All prediction models showed a high AUC. The best-performing models correctly identified 34% of the poor outcome patients at a cost of misclassifying 4% of non-poor outcome patients. Further studies are necessary to determine whether these accuracies are reproducible before implementation in clinical practice.
Collapse
Affiliation(s)
- Lucas A Ramos
- Department of Biomedical Engineering and Physics, University of Amsterdam, Amsterdam, Netherlands.,Department of Clinical Epidemiology and Biostatistics, University of Amsterdam, Amsterdam, Netherlands
| | - Manon Kappelhof
- Department of Radiology and Nuclear Medicine, University of Amsterdam, Amsterdam, Netherlands
| | | | - Vicky Chalos
- Department of Neurology, Erasmus MC - University Medical Center, Rotterdam, Netherlands.,Department of Public Health, Erasmus MC - University Medical Center, Rotterdam, Netherlands.,Department of Radiology and Nuclear Medicine, Erasmus MC - University Medical Center, Rotterdam, Netherlands
| | - Katinka Van Kranendonk
- Department of Radiology and Nuclear Medicine, University of Amsterdam, Amsterdam, Netherlands
| | - Nyika D Kruyt
- Department of Neurology, Leiden University Medical Center, Leiden, Netherlands
| | - Yvo B W E M Roos
- Department of Neurology, Amsterdam UMC, University of Amsterdam, Amsterdam, Netherlands
| | - Aad van der Lugt
- Department of Radiology and Nuclear Medicine, Erasmus MC - University Medical Center, Rotterdam, Netherlands
| | - Wim H van Zwam
- Department of Radiology, Cardiovascular Research Institute Maastricht, Maastricht University Medical Center, Maastricht, Netherlands
| | | | - Aeilko H Zwinderman
- Department of Clinical Epidemiology and Biostatistics, University of Amsterdam, Amsterdam, Netherlands
| | - Gustav J Strijkers
- Department of Biomedical Engineering and Physics, University of Amsterdam, Amsterdam, Netherlands.,Department of Radiology and Nuclear Medicine, University of Amsterdam, Amsterdam, Netherlands
| | | | - Mariekke J H Wermer
- Department of Neurology, Leiden University Medical Center, Leiden, Netherlands
| | - Silvia D Olabarriaga
- Department of Clinical Epidemiology and Biostatistics, University of Amsterdam, Amsterdam, Netherlands
| | - Charles B L M Majoie
- Department of Radiology and Nuclear Medicine, University of Amsterdam, Amsterdam, Netherlands
| | - Henk A Marquering
- Department of Biomedical Engineering and Physics, University of Amsterdam, Amsterdam, Netherlands.,Department of Radiology and Nuclear Medicine, University of Amsterdam, Amsterdam, Netherlands
| |
Collapse
|
61
|
Yedavalli VS, Tong E, Martin D, Yeom KW, Forkert ND. Artificial intelligence in stroke imaging: Current and future perspectives. Clin Imaging 2020; 69:246-254. [PMID: 32980785 DOI: 10.1016/j.clinimag.2020.09.005] [Citation(s) in RCA: 28] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/22/2020] [Revised: 07/08/2020] [Accepted: 09/11/2020] [Indexed: 12/12/2022]
Abstract
Artificial intelligence (AI) is a fast-growing research area in computer science that aims to mimic cognitive processes through a number of techniques. Supervised machine learning, a subfield of AI, includes methods that can identify patterns in high-dimensional data using labeled 'ground truth' data and apply these learnt patterns to analyze, interpret, or make predictions on new datasets. Supervised machine learning has become a significant area of interest within the medical community. Radiology and neuroradiology in particular are especially well suited for application of machine learning due to the vast amount of data that is generated. One devastating disease for which neuroimaging plays a significant role in the clinical management is stroke. Within this context, AI techniques can play pivotal roles for image-based diagnosis and management of stroke. This overview focuses on the recent advances of artificial intelligence methods - particularly supervised machine learning and deep learning - with respect to workflow, image acquisition and reconstruction, and image interpretation in patients with acute stroke, while also discussing potential pitfalls and future applications.
Collapse
Affiliation(s)
- Vivek S Yedavalli
- Stanford University, Department of Radiology, Division of Neuroradiology and Neurointervention, 300 Pasteur Drive, Room S047, Stanford, CA 94305, United States of America; Johns Hopkins Hospital, Department of Radiological Sciences, 600 N. Wolfe St. B 112-D, Baltimore, MD 21287, United States of America.
| | - Elizabeth Tong
- Stanford University, Department of Radiology, Division of Neuroradiology and Neurointervention, 300 Pasteur Drive, Room S031, Stanford, CA 94305, United States of America.
| | - Dann Martin
- Stanford University, Department of Radiology, Division of Neuroradiology and Neurointervention, 300 Pasteur Drive, Room S047, Stanford, CA 94305, United States of America.
| | - Kristen W Yeom
- Stanford University, Department of Radiology, Divisions of Neuroradiology and Pediatric Neuroradiology, 725 Welch Rd. MC 5654, Stanford, CA 94304, United States of America.
| | - Nils D Forkert
- Department of Radiology, Alberta Children's Hospital Research Institute, Hotchkiss Brain Institute Cumming School of Medicine, University of Calgary, HSC Building, Room 2913, 3330 Hospital Drive NW, Calgary, AB T2N 4N1, Canada; Department Clinical Neurosciences, Alberta Children's Hospital Research Institute, Hotchkiss Brain Institute Cumming School of Medicine, University of Calgary, HSC Building, Room 2913, 3330 Hospital Drive NW, Calgary, AB T2N 4N1, Canada.
| |
Collapse
|
62
|
Bhat A, Podstawczyk D, Walther BK, Aggas JR, Machado-Aranda D, Ward KR, Guiseppi-Elie A. Toward a hemorrhagic trauma severity score: fusing five physiological biomarkers. J Transl Med 2020; 18:348. [PMID: 32928219 PMCID: PMC7490913 DOI: 10.1186/s12967-020-02516-4] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2020] [Accepted: 09/04/2020] [Indexed: 12/30/2022] Open
Abstract
BACKGROUND To introduce the Hemorrhage Intensive Severity and Survivability (HISS) score, based on the fusion of multi-biomarker data; glucose, lactate, pH, potassium, and oxygen tension, to serve as a patient-specific attribute in hemorrhagic trauma. MATERIALS AND METHODS One hundred instances of Sensible Fictitious Rationalized Patient (SFRP) data were synthetically generated and the HISS score assigned by five clinically active physician experts (100 [5]). The HISS score stratifies the criticality of the trauma patient as; low(0), guarded(1), elevated(2), high(3) and severe(4). Standard classifier algorithms; linear support vector machine (SVM-L), multi-class ensemble bagged decision tree (EBDT), artificial neural network with bayesian regularization (ANN:BR) and possibility rule-based using function approximation (PRBF) were evaluated for their potential to similarly classify and predict a HISS score. RESULTS SVM-L, EBDT, ANN:BR and PRBF generated score predictions with testing accuracies (majority vote) corresponding to 0.91 ± 0.06, 0.93 ± 0.04, 0.92 ± 0.07, and 0.92 ± 0.03, respectively, with no statistically significant difference (p > 0.05). Targeted accuracies of 0.99 and 0.999 could be achieved with SFRP data size and clinical expert scores of 147[7](0.99) and 154[9](0.999), respectively. CONCLUSIONS The predictions of the data-driven model in conjunction with an adjunct multi-analyte biosensor intended for point-of-care continual monitoring of trauma patients, can aid in patient stratification and triage decision-making.
Collapse
Affiliation(s)
- Ankita Bhat
- Center for Bioelectronics, Biosensors and Biochips (C3B®), Department of Biomedical Engineering, Texas A&M University, College Station, TX 77843 USA
| | - Daria Podstawczyk
- Department of Process Engineering and Technology of Polymer and Carbon Materials, Wroclaw University of Science and Technology, Norwida 4/6, 50-373 Wroclaw, Poland
| | - Brandon K. Walther
- Center for Bioelectronics, Biosensors and Biochips (C3B®), Department of Biomedical Engineering, Texas A&M University, College Station, TX 77843 USA
- Department of Cardiovascular Sciences, Houston Methodist Institute for Academic Medicine and Houston Methodist Research Institute, 6670 Bertner Ave, Houston, TX 77030 USA
| | - John R. Aggas
- Center for Bioelectronics, Biosensors and Biochips (C3B®), Department of Biomedical Engineering, Texas A&M University, College Station, TX 77843 USA
| | - David Machado-Aranda
- Departments of Emergency Medicine and Biomedical Engineering, Michigan Center for Integrative Research in Critical Care, University of Michigan, Ann Arbor, MI 48109 USA
- Department of Surgery, Division of Acute Care Surgery, University of Michigan, Ann Arbor, MI 48109 USA
| | - Kevin R. Ward
- Department of Surgery, Division of Acute Care Surgery, University of Michigan, Ann Arbor, MI 48109 USA
| | - Anthony Guiseppi-Elie
- Center for Bioelectronics, Biosensors and Biochips (C3B®), Department of Biomedical Engineering, Texas A&M University, College Station, TX 77843 USA
- Department of Cardiovascular Sciences, Houston Methodist Institute for Academic Medicine and Houston Methodist Research Institute, 6670 Bertner Ave, Houston, TX 77030 USA
- Department of Electrical and Computer Engineering, Texas A&M University, College Station, TX 77843 USA
- ABTECH Scientific, Inc, Biotechnology Research Park, 800 East Leigh Street, Richmond, VA 23219 USA
| |
Collapse
|
63
|
Wang H, Lin J, Zheng L, Zhao J, Song B, Dai Y. Texture analysis based on ADC maps and T2-FLAIR images for the assessment of the severity and prognosis of ischaemic stroke. Clin Imaging 2020; 67:152-159. [PMID: 32739735 DOI: 10.1016/j.clinimag.2020.06.013] [Citation(s) in RCA: 14] [Impact Index Per Article: 3.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/13/2020] [Revised: 05/12/2020] [Accepted: 06/07/2020] [Indexed: 10/24/2022]
Abstract
OBJECTIVES To explore the feasibility of texture analysis based on T2-weighted fluid-attenuated inversion recovery (T2-FLAIR) images and apparent diffusion coefficient (ADC) maps in the assessment of the severity and prognosis of ischaemic stroke using the National Institutes of Health Stroke Scale (NIHSS) and modified Rankin scale (mRS) scores, respectively. METHODS Overall, 116 patients diagnosed with subacute ischaemic stroke were included in this retrospective study. Based on T2-FLAIR images and ADC maps, 15 texture features were extracted from the ROIs of each patient using grey-level co-occurrence matrix (GLCM) and local binary pattern histogram Fourier (LBP-HF) methods. The correlations of NIHSS score on admission (NIHSSbaseline), NIHSS score 24 h after stroke onset (NIHSS24h) and mRS score with the texture features were evaluated using Spearman's partial correlations. The receiver operating characteristic (ROC) curve was used to compare the performance of the selected texture features in the evaluation of stroke severity and prognosis. RESULTS Texture features derived from the T2-FLAIR images and ADC maps were correlated with NIHSS score and mRS score. EntropyADC and 0.75QuantileT2-FLAIR showed the best diagnostic performance for assessing stroke severity. The combination of EntropyADC and 0.75QuantileT2-FLAIR achieved a better performance in the evaluation of stroke severity (AUC = 0.7, p = 0.01) than either feature alone. Only 0.05QuantileT2-FLAIR was found to be correlated with mRS score, and none of the texture features were predictive of mRS score. CONCLUSION Texture features derived from T2-FLAIR images and ADC maps might serve as biomarkers to evaluate stroke severity, but were insufficient to predict stroke prognosis.
Collapse
Affiliation(s)
- Hao Wang
- Department of Radiology, Minhang Hospital, Fudan University, Shanghai, China
| | - Jixian Lin
- Department of Neurology, Minhang Hospital, Fudan University, Shanghai, China; Department of Electronic Engineering, Fudan University, Shanghai, China
| | - Liyun Zheng
- Central Research Institute, United Imaging Healthcare, Shanghai, China
| | - Jing Zhao
- Department of Neurology, Minhang Hospital, Fudan University, Shanghai, China
| | - Bin Song
- Department of Radiology, Minhang Hospital, Fudan University, Shanghai, China.
| | - Yongming Dai
- Central Research Institute, United Imaging Healthcare, Shanghai, China
| |
Collapse
|
64
|
Muhammad LJ, Algehyne EA, Usman SS. Predictive Supervised Machine Learning Models for Diabetes Mellitus. ACTA ACUST UNITED AC 2020; 1:240. [PMID: 33063051 PMCID: PMC7372976 DOI: 10.1007/s42979-020-00250-8] [Citation(s) in RCA: 12] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/03/2020] [Accepted: 07/10/2020] [Indexed: 02/08/2023]
Abstract
Diabetes mellitus (DM) is one of the deadliest diseases in the world, especially in developed nations. In recent years, it has become rampant in the developing nations such as Nigeria, posing more threats to individuals in the latter than those in the former. More than 415 million people were reported to suffer from DM worldwide as of 2015, with type 2 of the disease accounting for approximately 90% of the cases. The number of people with DM is expected to rise to 592 million by the year 2035. Therefore, DM is one of the growing public health concerns in Nigeria. In this study, the diagnostic dataset of DM type 2 was collected from the Murtala Mohammed Specialist Hospital, Kano, and used to develop predictive supervised machine learning models based on logistic regression, support vector machine, K-nearest neighbor, random forest, naive Bayes and gradient booting algorithms. The random forest predictive learning-based model appeared to be one of the best developed models with 88.76% in terms of accuracy; however, in terms of receiver operating characteristic curve, random forest and gradient booting predictive learning-based models were found to be the best predictive learning models with 86.28% predictive ability, respectively.
Collapse
Affiliation(s)
- L J Muhammad
- Department of Mathematics and Computer Science, Faculty of Science, Federal University of Kashere, P.M.B. 0182, Gombe, Nigeria
| | - Ebrahem A Algehyne
- Department of Mathematics, University of Tabuk, Tabuk, 71491 Saudi Arabia
| | - Sani Sharif Usman
- Department of Biological Sciences, Faculty of Science, Federal University of Kashere, P.M.B. 0182, Gombe, Nigeria
| |
Collapse
|
65
|
Lin CH, Hsu KC, Johnson KR, Fann YC, Tsai CH, Sun Y, Lien LM, Chang WL, Chen PL, Lin CL, Hsu CY. Evaluation of machine learning methods to stroke outcome prediction using a nationwide disease registry. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2020; 190:105381. [PMID: 32044620 PMCID: PMC7245557 DOI: 10.1016/j.cmpb.2020.105381] [Citation(s) in RCA: 28] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 08/27/2019] [Revised: 12/31/2019] [Accepted: 01/31/2020] [Indexed: 06/10/2023]
Abstract
INTRODUCTION Being able to predict functional outcomes after a stroke is highly desirable for clinicians. This allows clinicians to set reasonable goals with patients and relatives, and to reach shared after-care decisions for recovery or rehabilitation. The aim of this study was to apply various machine learning (ML) methods for 90-day stroke outcome predictions, using a nationwide disease registry. METHODS This study used the Taiwan Stroke Registry (TSR) which has prospectively collected data from stroke patients since 2006. Three known ML models (support vector machine, random forest, and artificial neural network), and a hybrid artificial neural network were implemented and evaluated by 10-time repeated hold-out with 10-fold cross-validation. RESULTS ML techniques present over 0.94 AUC in both ischemic and hemorrhagic stroke using preadmission and inpatient data. By adding follow-up data, the prediction ability improved to 0.97 AUC. We screened 206 clinical variables to identify 17 important features from the ischemic stroke dataset and 22 features from the hemorrhagic stroke dataset without losing much performance. Error analysis revealed that most prediction errors come from more severe stroke patients. CONCLUSION The study showed that ML techniques trained from large, cross-reginal registry datasets were able to predict functional outcome after stroke with high accuracy. The follow-up data is important which can further improve the predictive models' performance. With similar performances among different ML techniques, the algorithm's characteristics and performance on severe stroke patients will be the primary focus when we further develop inference models and artificial intelligence tools for potential medical.
Collapse
Affiliation(s)
- Ching-Heng Lin
- Center for Information Technology, National Institutes of Health, Bethesda, MD, United States; Bioinformatics Section, National Institute of Neurological Disorder and Stroke, National Institutes of Health, Bethesda, MD, United States
| | - Kai-Cheng Hsu
- Bioinformatics Section, National Institute of Neurological Disorder and Stroke, National Institutes of Health, Bethesda, MD, United States; Department of Neurology, National Taiwan University Hospital, Taipei, Taiwan
| | - Kory R Johnson
- Bioinformatics Section, National Institute of Neurological Disorder and Stroke, National Institutes of Health, Bethesda, MD, United States
| | - Yang C Fann
- Bioinformatics Section, National Institute of Neurological Disorder and Stroke, National Institutes of Health, Bethesda, MD, United States.
| | - Chon-Haw Tsai
- Division of Nephrology, China Medical University Hospital, Taichung, Taiwan
| | - Yu Sun
- Neurology, En Chu Kong Hospital, New Taipei City, Taiwan
| | - Li-Ming Lien
- Department of Neurology, Shin Kong Wu-Ho-Su Memorial Hospital, Taipei, Taiwan; Department of Neurology, College of Medicine, Taipei Medical University, Taipei, Taiwan
| | - Wei-Lun Chang
- Department of Neurology, Show Chwan Memorial Hospital, Changhua County, Taiwan
| | - Po-Lin Chen
- Neurological Institute, Taichung Veterans General Hospital, Taichung, Taiwan; Department of Neurology, School of Medicine, National Yang-Ming University, Taipei, Taiwan
| | - Cheng-Li Lin
- Management Office for Health Data, China Medical University Hospital, Taichung, Taiwan
| | - Chung Y Hsu
- Graduate Institute of Biomedical Sciences, China Medical University, Taichung, Taiwan
| |
Collapse
|
66
|
Wang W, Kiik M, Peek N, Curcin V, Marshall IJ, Rudd AG, Wang Y, Douiri A, Wolfe CD, Bray B. A systematic review of machine learning models for predicting outcomes of stroke with structured data. PLoS One 2020; 15:e0234722. [PMID: 32530947 PMCID: PMC7292406 DOI: 10.1371/journal.pone.0234722] [Citation(s) in RCA: 65] [Impact Index Per Article: 16.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/19/2020] [Accepted: 06/01/2020] [Indexed: 12/11/2022] Open
Abstract
Background and purpose Machine learning (ML) has attracted much attention with the hope that it could make use of large, routinely collected datasets and deliver accurate personalised prognosis. The aim of this systematic review is to identify and critically appraise the reporting and developing of ML models for predicting outcomes after stroke. Methods We searched PubMed and Web of Science from 1990 to March 2019, using previously published search filters for stroke, ML, and prediction models. We focused on structured clinical data, excluding image and text analysis. This review was registered with PROSPERO (CRD42019127154). Results Eighteen studies were eligible for inclusion. Most studies reported less than half of the terms in the reporting quality checklist. The most frequently predicted stroke outcomes were mortality (7 studies) and functional outcome (5 studies). The most commonly used ML methods were random forests (9 studies), support vector machines (8 studies), decision trees (6 studies), and neural networks (6 studies). The median sample size was 475 (range 70–3184), with a median of 22 predictors (range 4–152) considered. All studies evaluated discrimination with thirteen using area under the ROC curve whilst calibration was assessed in three. Two studies performed external validation. None described the final model sufficiently well to reproduce it. Conclusions The use of ML for predicting stroke outcomes is increasing. However, few met basic reporting standards for clinical prediction tools and none made their models available in a way which could be used or evaluated. Major improvements in ML study conduct and reporting are needed before it can meaningfully be considered for practice.
Collapse
Affiliation(s)
- Wenjuan Wang
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
- * E-mail:
| | - Martin Kiik
- School of Medical Education, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
| | - Niels Peek
- Division of Informatics, Imaging and Data Science, School of Health Sciences, University of Manchester, Manchester, United Kingdom
- NIHR Manchester Biomedical Research Centre, Manchester Academic Health Science Centre, University of Manchester, Manchester, United Kingdom
| | - Vasa Curcin
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
- NIHR Biomedical Research Centre, Guy's and St Thomas' NHS Foundation Trust and King's College London, London, United Kingdom
- NIHR Applied Research Collaboration (ARC) South London, London, United Kingdom
| | - Iain J. Marshall
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
| | - Anthony G. Rudd
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
| | - Yanzhong Wang
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
- NIHR Biomedical Research Centre, Guy's and St Thomas' NHS Foundation Trust and King's College London, London, United Kingdom
- NIHR Applied Research Collaboration (ARC) South London, London, United Kingdom
| | - Abdel Douiri
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
- NIHR Biomedical Research Centre, Guy's and St Thomas' NHS Foundation Trust and King's College London, London, United Kingdom
- NIHR Applied Research Collaboration (ARC) South London, London, United Kingdom
| | - Charles D. Wolfe
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
- NIHR Biomedical Research Centre, Guy's and St Thomas' NHS Foundation Trust and King's College London, London, United Kingdom
- NIHR Applied Research Collaboration (ARC) South London, London, United Kingdom
| | - Benjamin Bray
- School of Population Health & Environmental Sciences, Faculty of Life Science and Medicine, King’s College London, London, United Kingdom
| |
Collapse
|
67
|
Zihni E, Madai VI, Livne M, Galinovic I, Khalil AA, Fiebach JB, Frey D. Opening the black box of artificial intelligence for clinical decision support: A study predicting stroke outcome. PLoS One 2020; 15:e0231166. [PMID: 32251471 PMCID: PMC7135268 DOI: 10.1371/journal.pone.0231166] [Citation(s) in RCA: 52] [Impact Index Per Article: 13.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/08/2019] [Accepted: 03/17/2020] [Indexed: 01/02/2023] Open
Abstract
State-of-the-art machine learning (ML) artificial intelligence methods are increasingly leveraged in clinical predictive modeling to provide clinical decision support systems to physicians. Modern ML approaches such as artificial neural networks (ANNs) and tree boosting often perform better than more traditional methods like logistic regression. On the other hand, these modern methods yield a limited understanding of the resulting predictions. However, in the medical domain, understanding of applied models is essential, in particular, when informing clinical decision support. Thus, in recent years, interpretability methods for modern ML methods have emerged to potentially allow explainable predictions paired with high performance. To our knowledge, we present in this work the first explainability comparison of two modern ML methods, tree boosting and multilayer perceptrons (MLPs), to traditional logistic regression methods using a stroke outcome prediction paradigm. Here, we used clinical features to predict a dichotomized 90 days post-stroke modified Rankin Scale (mRS) score. For interpretability, we evaluated clinical features' importance with regard to predictions using deep Taylor decomposition for MLP, Shapley values for tree boosting and model coefficients for logistic regression. With regard to performance as measured by Area under the Curve (AUC) values on the test dataset, all models performed comparably: Logistic regression AUCs were 0.83, 0.83, 0.81 for three different regularization schemes; tree boosting AUC was 0.81; MLP AUC was 0.83. Importantly, the interpretability analysis demonstrated consistent results across models by rating age and stroke severity consecutively amongst the most important predictive features. For less important features, some differences were observed between the methods. Our analysis suggests that modern machine learning methods can provide explainability which is compatible with domain knowledge interpretation and traditional method rankings. Future work should focus on replication of these findings in other datasets and further testing of different explainability methods.
Collapse
Affiliation(s)
- Esra Zihni
- Charité Lab for Artificial Intelligence in Medicine—CLAIM, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Vince Istvan Madai
- Charité Lab for Artificial Intelligence in Medicine—CLAIM, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Michelle Livne
- Charité Lab for Artificial Intelligence in Medicine—CLAIM, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Ivana Galinovic
- Centre for Stroke Research Berlin, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Ahmed A. Khalil
- Centre for Stroke Research Berlin, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Jochen B. Fiebach
- Centre for Stroke Research Berlin, Charité - Universitätsmedizin Berlin, Berlin, Germany
| | - Dietmar Frey
- Charité Lab for Artificial Intelligence in Medicine—CLAIM, Charité - Universitätsmedizin Berlin, Berlin, Germany
| |
Collapse
|
68
|
Khodadadi M, Shayanfar H, Maghooli K, Hooshang Mazinan A. Fuzzy cognitive map based approach for determining the risk of ischemic stroke. IET Syst Biol 2020; 13:297-304. [PMID: 31778126 DOI: 10.1049/iet-syb.2018.5128] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/18/2023] Open
Abstract
Stroke is the third major cause of mortality in the world. The diagnosis of stroke is a very complex issue considering controllable and uncontrollable factors. These factors include age, sex, blood pressure, diabetes, obesity, heart disease, smoking, and so on, having a considerable influence on the diagnosis of stroke. Hence, designing an intelligent system leading to immediate and effective treatment is essential. In this study, the soft computing method known as fuzzy cognitive mapping was proposed for diagnosis of the risk of ischemic stroke. Non-linear Hebbian learning method was used for fuzzy cognitive maps training. In the proposed method, the risk rate for each person was determined based on the opinions of the neurologists. The accuracy of the proposed model was tested using 10-fold cross-validation, for 110 real cases, and the results were compared with those of support vector machine and K-nearest neighbours. The proposed system showed a superior performance with a total accuracy of (93.6 ± 4.5)%. The data used in this study is available by emailing the first author for academic and non-commercial purposes.
Collapse
Affiliation(s)
- Mahsa Khodadadi
- Department of Control Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran
| | - Heidarali Shayanfar
- Center of Excellence for Power Automation and Operation, College of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran.
| | - Keivan Maghooli
- Department of Biomedical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
| | - Amir Hooshang Mazinan
- Department of Control Engineering, South Tehran Branch, Islamic Azad University, Tehran, Iran
| |
Collapse
|
69
|
Waddle SL, Juttukonda MR, Lants SK, Davis LT, Chitale R, Fusco MR, Jordan LC, Donahue MJ. Classifying intracranial stenosis disease severity from functional MRI data using machine learning. J Cereb Blood Flow Metab 2020; 40:705-719. [PMID: 31068081 PMCID: PMC7168799 DOI: 10.1177/0271678x19848098] [Citation(s) in RCA: 17] [Impact Index Per Article: 4.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/05/2023]
Abstract
Translation of many non-invasive hemodynamic MRI methods to cerebrovascular disease patients has been hampered by well-known artifacts associated with delayed blood arrival times and reduced microvascular compliance. Using machine learning and support vector machine (SVM) algorithms, we investigated whether arrival time-related artifacts in these methods could be exploited as novel contrast sources to discriminate angiographically confirmed stenotic flow territories. Intracranial steno-occlusive moyamoya patients (n = 53; age = 45 ± 14.2 years; sex = 43 F) underwent (i) catheter angiography, (ii) anatomical MRI, (iii) cerebral blood flow (CBF)-weighted arterial spin labeling, and (iv) cerebrovascular reactivity (CVR)-weighted hypercapnic blood-oxygenation-level-dependent MRI. Mean, standard deviation (std), and 99th percentile of CBF, CVR, CVRDelay, and CVRMax were calculated in major anterior and posterior flow territories perfused by vessels with vs. without stenosis (≥70%) confirmed by catheter angiography. These and demographic variables were input into SVMs to evaluate discriminatory capacity for stenotic flow territories using k-fold cross-validation and receiver-operating-characteristic-area-under-the-curve to quantify variable combination relevance. Anterior circulation CBF-std, attributable to heterogeneous endovascular signal and prolonged arterial transit times, was the best performing single variable and CVRDelay-mean and CBF-std, both reflective of delayed vascular compliance, were a high-performing two-variable combination (specificity = 0.67; sensitivity = 0.75). Findings highlight the relevance of hemodynamic imaging and machine learning for identifying cerebrovascular impairment.
Collapse
Affiliation(s)
- Spencer L Waddle
- Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Meher R Juttukonda
- Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Sarah K Lants
- Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Larry T Davis
- Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Rohan Chitale
- Department of Neurosurgery, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Matthew R Fusco
- Department of Neurosurgery, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Lori C Jordan
- Department of Pediatrics, Division of Pediatric Neurology, Vanderbilt University Medical Center, Nashville, TN, USA
| | - Manus J Donahue
- Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA.,Department of Neurology, Vanderbilt University Medical Center, Nashville, TN, USA.,Department of Psychiatry, Vanderbilt University Medical Center, Nashville, TN, USA.,Department of Physics and Astronomy, Vanderbilt University, Nashville, TN, USA
| |
Collapse
|
70
|
Tozlu C, Edwards D, Boes A, Labar D, Tsagaris KZ, Silverstein J, Pepper Lane H, Sabuncu MR, Liu C, Kuceyeski A. Machine Learning Methods Predict Individual Upper-Limb Motor Impairment Following Therapy in Chronic Stroke. Neurorehabil Neural Repair 2020; 34:428-439. [PMID: 32193984 DOI: 10.1177/1545968320909796] [Citation(s) in RCA: 26] [Impact Index Per Article: 6.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/25/2022]
Abstract
Background. Accurate prediction of clinical impairment in upper-extremity motor function following therapy in chronic stroke patients is a difficult task for clinicians but is key in prescribing appropriate therapeutic strategies. Machine learning is a highly promising avenue with which to improve prediction accuracy in clinical practice. Objectives. The objective was to evaluate the performance of 5 machine learning methods in predicting postintervention upper-extremity motor impairment in chronic stroke patients using demographic, clinical, neurophysiological, and imaging input variables. Methods. A total of 102 patients (female: 31%, age 61 ± 11 years) were included. The upper-extremity Fugl-Meyer Assessment (UE-FMA) was used to assess motor impairment of the upper limb before and after intervention. Elastic net (EN), support vector machines, artificial neural networks, classification and regression trees, and random forest were used to predict postintervention UE-FMA. The performances of methods were compared using cross-validated R2. Results. EN performed significantly better than other methods in predicting postintervention UE-FMA using demographic and baseline clinical data (median REN2=0.91,RRF2=0.88,RANN2=0.83,RSVM2=0.79,RCART2=0.70; P < .05). Preintervention UE-FMA and the difference in motor threshold (MT) between the affected and unaffected hemispheres were the strongest predictors. The difference in MT had greater importance than the absence or presence of a motor-evoked potential (MEP) in the affected hemisphere. Conclusion. Machine learning methods may enable clinicians to accurately predict a chronic stroke patient's postintervention UE-FMA. Interhemispheric difference in the MT is an important predictor of chronic stroke patients' response to therapy and, therefore, could be included in prospective studies.
Collapse
Affiliation(s)
- Ceren Tozlu
- Department of Radiology, Weill Cornell Medicine, New York, NY, USA.,Brain and Mind Research Institute, Weill Cornell Medicine, New York, NY, USA
| | - Dylan Edwards
- Moss Rehabilitation Research Institute, Elkins Park, PA, USA.,Edith Cowan University, Joondalup, Australia.,Burke Neurological Institute, White Plains, NY, USA
| | - Aaron Boes
- Departments of Pediatrics, Neurology & Psychiatry, Iowa Neuroimaging and Noninvasive Brain Stimulation Laboratory, University of Iowa Hospitals and Clinics, Iowa City, IA, USA
| | - Douglas Labar
- Department of Neurology, Weill Cornell Medical College, New York, NY, USA
| | | | | | | | - Mert R Sabuncu
- School of Electrical and Computer Engineering and Meinig School of Biomedical Engineering, Cornell University, Ithaca, NY, USA
| | - Charles Liu
- USC Neurorestoration Center, Los Angeles, CA.,Rancho Los Amigos National Rehabilitation Center, Downey, CA, USA
| | - Amy Kuceyeski
- Department of Radiology, Weill Cornell Medicine, New York, NY, USA.,Brain and Mind Research Institute, Weill Cornell Medicine, New York, NY, USA
| |
Collapse
|
71
|
Ma Y, Zhang P, Tang Y, Pan C, Li G, Liu N, Hu Y, Tang Z. Artificial intelligence: The dawn of a new era for cutting-edge technology based diagnosis and treatment for stroke. BRAIN HEMORRHAGES 2020. [DOI: 10.1016/j.hest.2020.01.006] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022] Open
|
72
|
Esposti F, Banfi G. Fighting healthcare rocketing costs with value-based medicine: the case of stroke management. BMC Health Serv Res 2020; 20:75. [PMID: 32007089 PMCID: PMC6995121 DOI: 10.1186/s12913-020-4925-0] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/27/2019] [Accepted: 01/22/2020] [Indexed: 12/21/2022] Open
Abstract
Value-Based Medicine (VBM) is imposing itself as 'a new paradigm in healthcare management and medical practice.In this perspective paper, we discuss the role of VBM in dealing with the large productivity issue of the healthcare industry and examine some of the worldwide industrial and technological trends linked with VBM introduction. To clarify the points, we discuss examples of VBM management of stroke patients.In our conclusions, we support the idea of VBM as a strategic aid to manage rising costs in healthcare, and we explore the idea that VBM, by establishing value-generating networks among different healthcare stakeholders, can serve as the long sought-after redistributive mechanism that compensate patients for the industrial exploitation of their personal medical records.
Collapse
Affiliation(s)
- Federico Esposti
- Università Vita-Salute San Raffaele, via Olgettina 58, Milan, Italy
- IRCCS Ospedale San Raffaele, via Olgettina 60, Milan, Italy
| | - Giuseppe Banfi
- Università Vita-Salute San Raffaele, via Olgettina 58, Milan, Italy
- IRCCS Istituto Ortopedico Galeazzi, via Galeazzi 4, Milan, Italy
| |
Collapse
|
73
|
Lee H, Lee EJ, Ham S, Lee HB, Lee JS, Kwon SU, Kim JS, Kim N, Kang DW. Machine Learning Approach to Identify Stroke Within 4.5 Hours. Stroke 2020; 51:860-866. [PMID: 31987014 DOI: 10.1161/strokeaha.119.027611] [Citation(s) in RCA: 70] [Impact Index Per Article: 17.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Background and Purpose- We aimed to investigate the ability of machine learning (ML) techniques analyzing diffusion-weighted imaging (DWI) and fluid-attenuated inversion recovery (FLAIR) magnetic resonance imaging to identify patients within the recommended time window for thrombolysis. Methods- We analyzed DWI and FLAIR images of consecutive patients with acute ischemic stroke within 24 hours of clear symptom onset by applying automatic image processing approaches. These processes included infarct segmentation, DWI, and FLAIR imaging registration and image feature extraction. A total of 89 vector features from each image sequence were captured and used in the ML. Three ML models were developed to estimate stroke onset time for binary classification (≤4.5 hours): logistic regression, support vector machine, and random forest. To evaluate the performance of ML models, the sensitivity and specificity for identifying patients within 4.5 hours were compared with the sensitivity and specificity of human readings of DWI-FLAIR mismatch. Results- Data from a total of 355 patients were analyzed. DWI-FLAIR mismatch from human readings identified patients within 4.5 hours of symptom onset with 48.5% sensitivity and 91.3% specificity. ML algorithms had significantly greater sensitivities than human readers (75.8% for logistic regression, P=0.020; 72.7% for support vector machine, P=0.033; 75.8% for random forest, P=0.013) in detecting patients within 4.5 hours, but their specificities were comparable (82.6% for logistic regression, P=0.157; 82.6% for support vector machine, P=0.157; 82.6% for random forest, P=0.157). Conclusions- ML algorithms using multiple magnetic resonance imaging features were feasible even more sensitive than human readings in identifying patients with stroke within the time window for acute thrombolysis.
Collapse
Affiliation(s)
- Hyunna Lee
- From the Health Innovation Big Data Center, Asan Institute for Life Science, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (H.L.)
| | - Eun-Jae Lee
- Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (E.-J.L., H.-B.L., S.U.K., J.S.K., D.-W.K.)
| | - Sungwon Ham
- Asan Institute for Life Sciences, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (S.H.)
| | - Han-Bin Lee
- Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (E.-J.L., H.-B.L., S.U.K., J.S.K., D.-W.K.)
| | - Ji Sung Lee
- Clinical Research Center, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (J.S.L.)
| | - Sun U Kwon
- Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (E.-J.L., H.-B.L., S.U.K., J.S.K., D.-W.K.)
| | - Jong S Kim
- Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (E.-J.L., H.-B.L., S.U.K., J.S.K., D.-W.K.)
| | - Namkug Kim
- Department of Convergence Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (N.K.).,Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (N.K.)
| | - Dong-Wha Kang
- Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, South Korea. (E.-J.L., H.-B.L., S.U.K., J.S.K., D.-W.K.)
| |
Collapse
|
74
|
Turova V, Sidorenko I, Eckardt L, Rieger-Fackeldey E, Felderhoff-Müser U, Alves-Pinto A, Lampe R. Machine learning models for identifying preterm infants at risk of cerebral hemorrhage. PLoS One 2020; 15:e0227419. [PMID: 31940391 PMCID: PMC6961932 DOI: 10.1371/journal.pone.0227419] [Citation(s) in RCA: 13] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/26/2019] [Accepted: 12/18/2019] [Indexed: 11/18/2022] Open
Abstract
Intracerebral hemorrhage in preterm infants is a major cause of brain damage and cerebral palsy. The pathogenesis of cerebral hemorrhage is multifactorial. Among the risk factors are impaired cerebral autoregulation, infections, and coagulation disorders. Machine learning methods allow the identification of combinations of clinical factors to best differentiate preterm infants with intra-cerebral bleeding and the development of models for patients at risk of cerebral hemorrhage. In the current study, a Random Forest approach is applied to develop such models for extremely and very preterm infants (23-30 weeks gestation) based on data collected from a cohort of 229 individuals. The constructed models exhibit good prediction accuracy and might be used in clinical practice to reduce the risk of cerebral bleeding in prematurity.
Collapse
Affiliation(s)
- Varvara Turova
- Research Unit for Pediatric Neuroorthopedics and Cerebral Palsy of the Buhl-Strohmaier Foundation, Orthopedic Department, Klinikum Rechts der Isar, Technical University of Munich, München, Germany
- * E-mail:
| | - Irina Sidorenko
- Chair of Mathematical Modelling, Mathematical Faculty, Technical University of Munich, Garching bei München, Germany
| | - Laura Eckardt
- Departments of Pediatrics and Neonatology, University Hospital Essen, University of Duisburg‐Essen, Essen, Germany
| | - Esther Rieger-Fackeldey
- Department of Pediatrics, Neonatology, Klinikum Rechts der Isar, Technical University of Munich, München, Germany
| | - Ursula Felderhoff-Müser
- Departments of Pediatrics and Neonatology, University Hospital Essen, University of Duisburg‐Essen, Essen, Germany
| | - Ana Alves-Pinto
- Research Unit for Pediatric Neuroorthopedics and Cerebral Palsy of the Buhl-Strohmaier Foundation, Orthopedic Department, Klinikum Rechts der Isar, Technical University of Munich, München, Germany
| | - Renée Lampe
- Research Unit for Pediatric Neuroorthopedics and Cerebral Palsy of the Buhl-Strohmaier Foundation, Orthopedic Department, Klinikum Rechts der Isar, Technical University of Munich, München, Germany
| |
Collapse
|
75
|
Heo J, Yoon JG, Park H, Kim YD, Nam HS, Heo JH. Machine Learning-Based Model for Prediction of Outcomes in Acute Stroke. Stroke 2020; 50:1263-1265. [PMID: 30890116 DOI: 10.1161/strokeaha.118.024293] [Citation(s) in RCA: 257] [Impact Index Per Article: 64.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/19/2022]
Abstract
Background and Purpose- The prediction of long-term outcomes in ischemic stroke patients may be useful in treatment decisions. Machine learning techniques are being increasingly adapted for use in the medical field because of their high accuracy. This study investigated the applicability of machine learning techniques to predict long-term outcomes in ischemic stroke patients. Methods- This was a retrospective study using a prospective cohort that enrolled patients with acute ischemic stroke. Favorable outcome was defined as modified Rankin Scale score 0, 1, or 2 at 3 months. We developed 3 machine learning models (deep neural network, random forest, and logistic regression) and compared their predictability. To evaluate the accuracy of the machine learning models, we also compared them to the Acute Stroke Registry and Analysis of Lausanne (ASTRAL) score. Results- A total of 2604 patients were included in this study, and 2043 (78%) of them had favorable outcomes. The area under the curve for the deep neural network model was significantly higher than that of the ASTRAL score (0.888 versus 0.839; P<0.001), while the areas under the curves of the random forest (0.857; P=0.136) and logistic regression (0.849; P=0.413) models were not significantly higher than that of the ASTRAL score. Using only the 6 variables that are used for the ASTRAL score, the performance of the machine learning models did not significantly differ from that of the ASTRAL score. Conclusions- Machine learning algorithms, particularly the deep neural network, can improve the prediction of long-term outcomes in ischemic stroke patients.
Collapse
Affiliation(s)
- JoonNyung Heo
- From the Department of Neurology (J.H., H.P., Y.D.K., H.S.N., J.H.H.), Yonsei University College of Medicine, Seoul, Korea
| | - Jihoon G Yoon
- Department of Laboratory Medicine (J.G.Y.), Yonsei University College of Medicine, Seoul, Korea
| | - Hyungjong Park
- From the Department of Neurology (J.H., H.P., Y.D.K., H.S.N., J.H.H.), Yonsei University College of Medicine, Seoul, Korea
| | - Young Dae Kim
- From the Department of Neurology (J.H., H.P., Y.D.K., H.S.N., J.H.H.), Yonsei University College of Medicine, Seoul, Korea
| | - Hyo Suk Nam
- From the Department of Neurology (J.H., H.P., Y.D.K., H.S.N., J.H.H.), Yonsei University College of Medicine, Seoul, Korea
| | - Ji Hoe Heo
- From the Department of Neurology (J.H., H.P., Y.D.K., H.S.N., J.H.H.), Yonsei University College of Medicine, Seoul, Korea
| |
Collapse
|
76
|
Abstract
Artificial intelligence (AI), a discipline encompassed by data science, has seen recent rapid growth in its application to healthcare and beyond, and is now an integral part of daily life. Uses of AI in gastroenterology include the automated detection of disease and differentiation of pathology subtypes and disease severity. Although a majority of AI research in gastroenterology focuses on adult applications, there are a number of pediatric pathologies that could benefit from more research. As new and improved diagnostic tools become available and more information is retrieved from them, AI could provide physicians a method to distill enormous amounts of data into enhanced decision-making and cost saving for children with digestive disorders. This review provides a broad overview of AI and examples of its possible applications in pediatric gastroenterology.
Collapse
|
77
|
Alberts M, Chen YW, Lin JH, Kogan E, Twyman K, Milentijevic D. Risks of Stroke and Mortality in Atrial Fibrillation Patients Treated With Rivaroxaban and Warfarin. Stroke 2019; 51:549-555. [PMID: 31888412 PMCID: PMC7004448 DOI: 10.1161/strokeaha.119.025554] [Citation(s) in RCA: 31] [Impact Index Per Article: 6.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/22/2023]
Abstract
Background and Purpose- Oral anticoagulation therapy is standard of care for patients with nonvalvular atrial fibrillation to prevent stroke. This study compared rivaroxaban and warfarin for stroke and all-cause mortality risk reduction in a real-world setting. Methods- This retrospective cohort study (2011-2017) included de-identified patients from the Optum Clinformatics Database who started treatment with rivaroxaban or warfarin within 30 days following initial diagnosis of nonvalvular atrial fibrillation. Before nonvalvular atrial fibrillation diagnosis, patients had 6 months of continuous health plan enrollment and CHA2DS2-VASc score ≥2. Stroke severity was determined by the National Institutes of Health Stroke Scale, imputed based on machine learning algorithms. Stroke and all-cause mortality risks were compared by treatment using Cox proportional hazard regression, with inverse probability of treatment weighting to balance cohorts for baseline risk factors. Stratified analysis by treatment duration was also performed. Results- During a mean follow-up of 27 months, 175 (1.33/100 patient-years [PY]) rivaroxaban-treated and 536 (1.66/100 PY) warfarin-treated patients developed stroke. The inverse probability of treatment weighting model showed that rivaroxaban reduced stroke risk by 19% (hazard ratio [HR], 0.81 [95% CI, 0.73-0.91]). Analysis by stroke severity revealed risk reductions by rivaroxaban of 48% for severe stroke (National Institutes of Health Stroke Scale score, 16-42; HR, 0.52 [95% CI, 0.33-0.82]) and 19% for minor stroke (National Institutes of Health Stroke Scale score, 1 to <5; HR, 0.81 [95% CI, 0.68-0.96]), but no difference for moderate stroke (National Institutes of Health Stroke Scale score, 5 to <16; HR, 0.93 [95% CI, 0.78-1.10]). A total of 41 (0.31/100 PY) rivaroxaban-treated and 147 (0.44/100 PY) warfarin-treated patients died poststroke, 12 (0.09/100 PY) and 67 (0.20/100 PY) of whom died within 30 days, representing mortality risk reductions by rivaroxaban of 24% (HR, 0.76 [95% CI, 0.61-0.95]) poststroke and 59% (HR, 0.41 [95% CI, 0.28-0.60]) within 30 days. Conclusions- After the initial diagnosis of atrial fibrillation, patients treated with rivaroxaban versus warfarin had significant risk reduction for stroke, especially severe stroke, and all-cause mortality after a stroke. Findings from this observational study may help inform anticoagulant choice for stroke prevention in patients with nonvalvular atrial fibrillation.
Collapse
Affiliation(s)
| | - Yen-Wen Chen
- Janssen Scientific Affairs, LLC, Titusville, NJ (Y.-W.C., J.H.L., D.M.)
| | - Jennifer H Lin
- Janssen Scientific Affairs, LLC, Titusville, NJ (Y.-W.C., J.H.L., D.M.)
| | - Emily Kogan
- Janssen Research & Development, LLC, Raritan, NJ (E.K., K.T.)
| | - Kathryn Twyman
- Janssen Research & Development, LLC, Raritan, NJ (E.K., K.T.).,Mount Sinai Hospital, New York (K.T.)
| | | |
Collapse
|
78
|
Wang K, Shou Q, Ma SJ, Liebeskind D, Qiao XJ, Saver J, Salamon N, Kim H, Yu Y, Xie Y, Zaharchuk G, Scalzo F, Wang DJJ. Deep Learning Detection of Penumbral Tissue on Arterial Spin Labeling in Stroke. Stroke 2019; 51:489-497. [PMID: 31884904 DOI: 10.1161/strokeaha.119.027457] [Citation(s) in RCA: 20] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/28/2022]
Abstract
Background and Purpose- Selection of patients with acute ischemic stroke for endovascular treatment generally relies on dynamic susceptibility contrast magnetic resonance imaging or computed tomography perfusion. Dynamic susceptibility contrast magnetic resonance imaging requires injection of contrast, whereas computed tomography perfusion requires high doses of ionizing radiation. The purpose of this work was to develop and evaluate a deep learning (DL)-based algorithm for assisting the selection of suitable patients with acute ischemic stroke for endovascular treatment based on 3-dimensional pseudo-continuous arterial spin labeling (pCASL). Methods- A total of 167 image sets of 3-dimensional pCASL data from 137 patients with acute ischemic stroke scanned on 1.5T and 3.0T Siemens MR systems were included for neural network training. The concurrently acquired dynamic susceptibility contrast magnetic resonance imaging was used to produce labels of hypoperfused brain regions, analyzed using commercial software. The DL and 6 machine learning (ML) algorithms were trained with 10-fold cross-validation. The eligibility for endovascular treatment was determined retrospectively based on the criteria of perfusion/diffusion mismatch in the DEFUSE 3 trial (Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke). The trained DL algorithm was further applied on twelve 3-dimensional pCASL data sets acquired on 1.5T and 3T General Electric MR systems, without fine-tuning of parameters. Results- The DL algorithm can predict the dynamic susceptibility contrast-defined hypoperfusion region in pCASL with a voxel-wise area under the curve of 0.958, while the 6 ML algorithms ranged from 0.897 to 0.933. For retrospective determination for subject-level endovascular treatment eligibility, the DL algorithm achieved an accuracy of 92%, with a sensitivity of 0.89 and specificity of 0.95. When applied to the GE pCASL data, the DL algorithm achieved a voxel-wise area under the curve of 0.94 and a subject-level accuracy of 92% for endovascular treatment eligibility. Conclusions- pCASL perfusion magnetic resonance imaging in conjunction with the DL algorithm provides a promising approach for assisting decision-making for endovascular treatment in patients with acute ischemic stroke.
Collapse
Affiliation(s)
- Kai Wang
- From the Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, Los Angeles (K.W., Q.S., S.J.M., H.K., D.J.J.W.)
| | - Qinyang Shou
- From the Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, Los Angeles (K.W., Q.S., S.J.M., H.K., D.J.J.W.)
| | - Samantha J Ma
- From the Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, Los Angeles (K.W., Q.S., S.J.M., H.K., D.J.J.W.)
| | - David Liebeskind
- Department of Neurology (D.L., J.S., F.S.), University of California, Los Angeles
| | - Xin J Qiao
- Department of Radiology (X.J.Q., N.S.), University of California, Los Angeles
| | - Jeffrey Saver
- Department of Neurology (D.L., J.S., F.S.), University of California, Los Angeles
| | - Noriko Salamon
- Department of Radiology (X.J.Q., N.S.), University of California, Los Angeles
| | - Hosung Kim
- From the Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, Los Angeles (K.W., Q.S., S.J.M., H.K., D.J.J.W.)
| | - Yannan Yu
- Department of Radiology, Stanford University, Palo Alto, CA (Y.Y., Y.X., G.Z.)
| | - Yuan Xie
- Department of Radiology, Stanford University, Palo Alto, CA (Y.Y., Y.X., G.Z.)
| | - Greg Zaharchuk
- Department of Radiology, Stanford University, Palo Alto, CA (Y.Y., Y.X., G.Z.)
| | - Fabien Scalzo
- Department of Neurology (D.L., J.S., F.S.), University of California, Los Angeles
| | - Danny J J Wang
- From the Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, Los Angeles (K.W., Q.S., S.J.M., H.K., D.J.J.W.)
| |
Collapse
|
79
|
Li X, Bian D, Yu J, Li M, Zhao D. Using machine learning models to improve stroke risk level classification methods of China national stroke screening. BMC Med Inform Decis Mak 2019; 19:261. [PMID: 31822270 PMCID: PMC6902572 DOI: 10.1186/s12911-019-0998-2] [Citation(s) in RCA: 19] [Impact Index Per Article: 3.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/14/2019] [Accepted: 11/25/2019] [Indexed: 12/20/2022] Open
Abstract
Background With the character of high incidence, high prevalence and high mortality, stroke has brought a heavy burden to families and society in China. In 2009, the Ministry of Health of China launched the China national stroke screening and intervention program, which screens stroke and its risk factors and conducts high-risk population interventions for people aged above 40 years old all over China. In this program, stroke risk factors include hypertension, diabetes, dyslipidemia, smoking, lack of exercise, apparently overweight and family history of stroke. People with more than two risk factors or history of stroke or transient ischemic attack (TIA) are considered as high-risk. However, it is impossible for this criterion to classify stroke risk levels for people with unknown values in fields of risk factors. The missing of stroke risk levels results in reduced efficiency of stroke interventions and inaccuracies in statistical results at the national level. In this paper, we use 2017 national stroke screening data to develop stroke risk classification models based on machine learning algorithms to improve the classification efficiency. Method Firstly, we construct training set and test sets and process the imbalance training set based on oversampling and undersampling method. Then, we develop logistic regression model, Naïve Bayesian model, Bayesian network model, decision tree model, neural network model, random forest model, bagged decision tree model, voting model and boosting model with decision trees to classify stroke risk levels. Result The recall of the boosting model with decision trees is the highest (99.94%), and the precision of the model based on the random forest is highest (97.33%). Using the random forest model (recall: 98.44%), the recall will be increased by about 2.8% compared with the method currently used, and several thousands more people with high risk of stroke can be identified each year. Conclusion Models developed in this paper can improve the current screening method in the way that it can avoid the impact of unknown values, and avoid unnecessary rescreening and intervention expenditures. The national stroke screening program can choose classification models according to the practice need.
Collapse
Affiliation(s)
- Xuemeng Li
- Information Center, Academy of Military Medical Sciences, Beijing, China
| | - Di Bian
- School of Electrical and Control Engineering, Xi'an University of Science and Technology
- , Xi'an, China
| | - Jinghui Yu
- Information Center, Academy of Military Medical Sciences, Beijing, China
| | - Mei Li
- China Stroke Data Center, Beijing, China
| | - Dongsheng Zhao
- Information Center, Academy of Military Medical Sciences, Beijing, China.
| |
Collapse
|
80
|
Artificial Intelligence in Interventional Radiology: A Literature Review and Future Perspectives. JOURNAL OF ONCOLOGY 2019; 2019:6153041. [PMID: 31781215 PMCID: PMC6874978 DOI: 10.1155/2019/6153041] [Citation(s) in RCA: 15] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Subscribe] [Scholar Register] [Received: 05/18/2019] [Revised: 09/22/2019] [Accepted: 10/01/2019] [Indexed: 01/17/2023]
Abstract
The term “artificial intelligence” (AI) includes computational algorithms that can perform tasks considered typical of human intelligence, with partial to complete autonomy, to produce new beneficial outputs from specific inputs. The development of AI is largely based on the introduction of artificial neural networks (ANN) that allowed the introduction of the concepts of “computational learning models,” machine learning (ML) and deep learning (DL). AI applications appear promising for radiology scenarios potentially improving lesion detection, segmentation, and interpretation with a recent application also for interventional radiology (IR) practice, including the ability of AI to offer prognostic information to both patients and physicians about interventional oncology procedures. This article integrates evidence-reported literature and experience-based perceptions to assist not only residents and fellows who are training in interventional radiology but also practicing colleagues who are approaching to locoregional mini-invasive treatments.
Collapse
|
81
|
Murray NM, Unberath M, Hager GD, Hui FK. Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review. J Neurointerv Surg 2019; 12:156-164. [DOI: 10.1136/neurintsurg-2019-015135] [Citation(s) in RCA: 107] [Impact Index Per Article: 21.4] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/20/2019] [Revised: 07/29/2019] [Accepted: 07/29/2019] [Indexed: 11/04/2022]
Abstract
Background and purposeAcute stroke caused by large vessel occlusions (LVOs) requires emergent detection and treatment by endovascular thrombectomy. However, radiologic LVO detection and treatment is subject to variable delays and human expertise, resulting in morbidity. Imaging software using artificial intelligence (AI) and machine learning (ML), a branch of AI, may improve rapid frontline detection of LVO strokes. This report is a systematic review of AI in acute LVO stroke identification and triage, and characterizes LVO detection software.MethodsA systematic review of acute stroke diagnostic-focused AI studies from January 2014 to February 2019 in PubMed, Medline, and Embase using terms: ‘artificial intelligence’ or ‘machine learning or deep learning’ and ‘ischemic stroke’ or ‘large vessel occlusion’ was performed.ResultsVariations of AI, including ML methods of random forest learning (RFL) and convolutional neural networks (CNNs), are used to detect LVO strokes. Twenty studies were identified that use ML. Alberta Stroke Program Early CT Score (ASPECTS) commonly used RFL, while LVO detection typically used CNNs. Image feature detection had greater sensitivity with CNN than with RFL, 85% versus 68%. However, AI algorithm performance metrics use different standards, precluding ideal objective comparison. Four current software platforms incorporate ML: Brainomix (greatest validation of AI for ASPECTS, uses CNNs to automatically detect LVOs), General Electric, iSchemaView (largest number of perfusion study validations for thrombectomy), and Viz.ai (uses CNNs to automatically detect LVOs, then automatically activates emergency stroke treatment systems).ConclusionsAI may improve LVO stroke detection and rapid triage necessary for expedited treatment. Standardization of performance assessment is needed in future studies.
Collapse
|
82
|
Allen M, Pearn K, Monks T, Bray BD, Everson R, Salmon A, James M, Stein K. Can clinical audits be enhanced by pathway simulation and machine learning? An example from the acute stroke pathway. BMJ Open 2019; 9:e028296. [PMID: 31530590 PMCID: PMC6756466 DOI: 10.1136/bmjopen-2018-028296] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/30/2018] [Revised: 08/05/2019] [Accepted: 08/21/2019] [Indexed: 01/02/2023] Open
Abstract
OBJECTIVE To evaluate the application of clinical pathway simulation in machine learning, using clinical audit data, in order to identify key drivers for improving use and speed of thrombolysis at individual hospitals. DESIGN Computer simulation modelling and machine learning. SETTING Seven acute stroke units. PARTICIPANTS Anonymised clinical audit data for 7864 patients. RESULTS Three factors were pivotal in governing thrombolysis use: (1) the proportion of patients with a known stroke onset time (range 44%-73%), (2) pathway speed (for patients arriving within 4 hours of onset: per-hospital median arrival-to-scan ranged from 11 to 56 min; median scan-to-thrombolysis ranged from 21 to 44 min) and (3) predisposition to use thrombolysis (thrombolysis use ranged from 31% to 52% for patients with stroke scanned with 30 min left to administer thrombolysis). A pathway simulation model could predict the potential benefit of improving individual stages of the clinical pathway speed, whereas a machine learning model could predict the benefit of 'exporting' clinical decision making from one hospital to another, while allowing for differences in patient population between hospitals. By applying pathway simulation and machine learning together, we found a realistic ceiling of 15%-25% use of thrombolysis across different hospitals and, in the seven hospitals studied, a realistic opportunity to double the number of patients with no significant disability that may be attributed to thrombolysis. CONCLUSIONS National clinical audit may be enhanced by a combination of pathway simulation and machine learning, which best allows for an understanding of key levers for improvement in hyperacute stroke pathways, allowing for differences between local patient populations. These models, based on standard clinical audit data, may be applied at scale while providing results at individual hospital level. The models facilitate understanding of variation and levers for improvement in stroke pathways, and help set realistic targets tailored to local populations.
Collapse
Affiliation(s)
| | - Kerry Pearn
- Medical School, University of Exeter, Exeter, UK
| | | | | | | | | | - Martin James
- Stroke Consultant, Royal Devon & Exeter NHS Trust, Exeter, UK
| | - Ken Stein
- Medical School, University of Exeter, Exeter, UK
| |
Collapse
|
83
|
Technical considerations of multi-parametric tissue outcome prediction methods in acute ischemic stroke patients. Sci Rep 2019; 9:13208. [PMID: 31519923 PMCID: PMC6744509 DOI: 10.1038/s41598-019-49460-y] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2018] [Accepted: 08/23/2019] [Indexed: 12/31/2022] Open
Abstract
Decisions regarding acute stroke treatment rely heavily on imaging, but interpretation can be difficult for physicians. Machine learning methods can assist clinicians by providing tissue outcome predictions for different treatment approaches based on acute multi-parametric imaging. To produce such clinically viable machine learning models, factors such as classifier choice, data normalization, and data balancing must be considered. This study gives comprehensive consideration to these factors by comparing the agreement of voxel-based tissue outcome predictions using acute imaging and clinical parameters with manual lesion segmentations derived from follow-up imaging. This study considers random decision forest, generalized linear model, and k-nearest-neighbor machine learning classifiers in conjunction with three data normalization approaches (non-normalized, relative to contralateral hemisphere, and relative to contralateral VOI), and two data balancing strategies (full dataset and stratified subsampling). These classifier settings were evaluated based on 90 MRI datasets from acute ischemic stroke patients. Distinction was made between patients recanalized using intraarterial and intravenous methods, as well as those without successful recanalization. For primary quantitative comparison, the Dice metric was computed for each voxel-based tissue outcome prediction and its corresponding follow-up lesion segmentation. It was found that the random forest classifier outperformed the generalized linear model and the k-nearest-neighbor classifier, that normalization did not improve the Dice score of the lesion outcome predictions, and that the models generated lesion outcome predictions with higher Dice scores when trained with balanced datasets. No significant difference was found between the treatment groups (intraarterial vs intravenous) regarding the Dice score of the tissue outcome predictions.
Collapse
|
84
|
Nishi H, Oishi N, Ishii A, Ono I, Ogura T, Sunohara T, Chihara H, Fukumitsu R, Okawa M, Yamana N, Imamura H, Sadamasa N, Hatano T, Nakahara I, Sakai N, Miyamoto S. Predicting Clinical Outcomes of Large Vessel Occlusion Before Mechanical Thrombectomy Using Machine Learning. Stroke 2019; 50:2379-2388. [PMID: 31409267 DOI: 10.1161/strokeaha.119.025411] [Citation(s) in RCA: 67] [Impact Index Per Article: 13.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Background and Purpose- The clinical course of acute ischemic stroke with large vessel occlusion (LVO) is a multifactorial process with various prognostic factors. We aimed to model this process with machine learning and predict the long-term clinical outcome of LVO before endovascular treatment and to compare our method with previously developed pretreatment scoring methods. Methods- The derivation cohort included 387 LVO patients, and the external validation cohort included 115 LVO patients with anterior circulation who were treated with mechanical thrombectomy. The statistical model with logistic regression without regularization and machine learning algorithms, such as regularized logistic regression, linear support vector machine, and random forest, were used to predict good clinical outcome (modified Rankin Scale score of 0-2 at 90 days) with standard and multiple pretreatment clinical variables. Five previously reported pretreatment scoring methods (the Pittsburgh Response to Endovascular Therapy score, the Stroke Prognostication Using Age and National Institutes of Health Stroke Scale index, the Totaled Health Risks in Vascular Events score, the Houston Intra-Arterial Therapy score, and the Houston Intra-Arterial Therapy 2 score) were compared with these models for the area under the receiver operating characteristic curve. Results- The area under the receiver operating characteristic curve of random forest, which was the worst among the machine learning algorithms, was significantly higher than those of the standard statistical model and the best model among the previously reported pretreatment scoring methods in the derivation (the area under the receiver operating characteristic curve were 0.85±0.07 for random forest, 0.78±0.08 for logistic regression without regularization, and 0.77±0.09 for Stroke Prognostication using Age and National Institutes of Health Stroke Scale) and validation cohorts (the area under the receiver operating characteristic curve were 0.87±0.01 for random forest, 0.56±0.07 for logistic regression without regularization, and 0.83±0.00 for Pittsburgh Response to Endovascular Therapy). Conclusions- Machine learning methods with multiple pretreatment clinical variables can predict clinical outcomes of patients with anterior circulation LVO who undergo mechanical thrombectomy more accurately than previously developed pretreatment scoring methods.
Collapse
Affiliation(s)
- Hidehisa Nishi
- Form the Department of Neurosurgery (H.N., I.O., M.O., S.M.), Kyoto University Graduate School of Medicine, Kyoto, Japan
| | - Naoya Oishi
- Medical Innovation Center (N.O.), Kyoto University Graduate School of Medicine, Kyoto, Japan
| | - Akira Ishii
- Department of Neurology (A.I.), Kyoto University Graduate School of Medicine, Kyoto, Japan
| | - Isao Ono
- Form the Department of Neurosurgery (H.N., I.O., M.O., S.M.), Kyoto University Graduate School of Medicine, Kyoto, Japan
| | - Takenori Ogura
- Department of Neurosurgery, Kokura Memorial Hospital, Kitakyushu, Japan (T.O., H.C., T.H.)
| | - Tadashi Sunohara
- Department of Neurosurgery, Kobe City Medical Center General Hospital, Japan (T.S., R.F., H.I., N. Sakai)
| | - Hideo Chihara
- Department of Neurosurgery, Kokura Memorial Hospital, Kitakyushu, Japan (T.O., H.C., T.H.)
| | - Ryu Fukumitsu
- Department of Neurosurgery, Kobe City Medical Center General Hospital, Japan (T.S., R.F., H.I., N. Sakai)
| | - Masakazu Okawa
- Form the Department of Neurosurgery (H.N., I.O., M.O., S.M.), Kyoto University Graduate School of Medicine, Kyoto, Japan
| | - Norikazu Yamana
- Department of Neurosurgery, Koseikai Takeda Hospital, Kyoto, Japan (N.Y., N. Sadamasa)
| | - Hirotoshi Imamura
- Department of Neurosurgery, Kobe City Medical Center General Hospital, Japan (T.S., R.F., H.I., N. Sakai)
| | - Nobutake Sadamasa
- Department of Neurosurgery, Koseikai Takeda Hospital, Kyoto, Japan (N.Y., N. Sadamasa)
| | - Taketo Hatano
- Department of Neurosurgery, Kokura Memorial Hospital, Kitakyushu, Japan (T.O., H.C., T.H.)
| | - Ichiro Nakahara
- Department of Comprehensive Strokology, Fujita Health University School of Medicine, Toyoake, Japan (I.N.)
| | - Nobuyuki Sakai
- Department of Neurosurgery, Kobe City Medical Center General Hospital, Japan (T.S., R.F., H.I., N. Sakai)
| | - Susumu Miyamoto
- Form the Department of Neurosurgery (H.N., I.O., M.O., S.M.), Kyoto University Graduate School of Medicine, Kyoto, Japan
| |
Collapse
|
85
|
Zhou RQ, Ji HC, Liu Q, Zhu CY, Liu R. Leveraging machine learning techniques for predicting pancreatic neuroendocrine tumor grades using biochemical and tumor markers. World J Clin Cases 2019; 7:1611-1622. [PMID: 31367620 PMCID: PMC6658377 DOI: 10.12998/wjcc.v7.i13.1611] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/27/2019] [Revised: 05/15/2019] [Accepted: 05/16/2019] [Indexed: 02/05/2023] Open
Abstract
BACKGROUND The incidence of pancreatic neuroendocrine tumors (PNETs) is now increasing rapidly. The tumor grade of PNETs significantly affects the treatment strategy and prognosis. However, there is still no effective way to non-invasively classify PNET grades. Machine learning (ML) algorithms have shown potential in improving the prediction accuracy using comprehensive data.
AIM To provide a ML approach to predict PNET tumor grade using clinical data.
METHODS The clinical data of histologically confirmed PNET cases between 2012 and 2018 were collected. A method of minimum P for the Chi-square test was used to divide the continuous variables into binary variables. The continuous variables were transformed into binary variables according to the cutoff value, while the P value was minimum. Four classical supervised ML models, including logistic regression, support vector machine (SVM), linear discriminant analysis (LDA) and multi-layer perceptron (MLP) were trained by clinical data, and the models were labeled with the pathological tumor grade of each PNET patient. The performance of each model, including the weight of the different parameters, were evaluated.
RESULTS In total, 91 PNET cases were included in this study, in which 32 were G1, 48 were G2 and 11 were G3. The results showed that there were significant differences among the clinical parameters of patients with different grades. Patients with higher grades tended to have higher values of total bilirubin, alpha fetoprotein, carcinoembryonic antigen, carbohydrate antigen 19-9 and carbohydrate antigen 72-4. Among the models we used, LDA performed best in predicting the PNET tumor grade. Meanwhile, MLP had the highest recall rate for G3 cases. All of the models stabilized when the sample size was over 70 percent of the total, except for SVM. Different parameters varied in affecting the outcomes of the models. Overall, alanine transaminase, total bilirubin, carcinoembryonic antigen, carbohydrate antigen 19-9 and carbohydrate antigen 72-4 affected the outcome greater than other parameters.
CONCLUSION ML could be a simple and effective method in non-invasively predicting PNET grades by using the routine data obtained from the results of biochemical and tumor markers.
Collapse
Affiliation(s)
- Rui-Quan Zhou
- School of Medicine, Nankai University, Tianjin 300071, China
| | - Hong-Chen Ji
- The Second Department of Hepatobiliary Surgery, Chinese PLA General Hospital, Beijing 100853, China
| | - Qu Liu
- The Second Department of Hepatobiliary Surgery, Chinese PLA General Hospital, Beijing 100853, China
| | - Chun-Yu Zhu
- The Second Department of Hepatobiliary Surgery, Chinese PLA General Hospital, Beijing 100853, China
| | - Rong Liu
- School of Medicine, Nankai University, Tianjin 300071, China
- The Second Department of Hepatobiliary Surgery, Chinese PLA General Hospital, Beijing 100853, China
| |
Collapse
|
86
|
Bang OY, Li W. Applications of diffusion-weighted imaging in diagnosis, evaluation, and treatment of acute ischemic stroke. PRECISION AND FUTURE MEDICINE 2019. [DOI: 10.23838/pfm.2019.00037] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/02/2023] Open
|
87
|
Diprose WK, Wang MTM, McFetridge A, Sutcliffe J, Barber PA. Glycated hemoglobin (HbA1c) and outcome following endovascular thrombectomy for ischemic stroke. J Neurointerv Surg 2019; 12:30-32. [PMID: 31147437 DOI: 10.1136/neurintsurg-2019-015023] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/15/2019] [Revised: 05/06/2019] [Accepted: 05/13/2019] [Indexed: 01/04/2023]
Abstract
BACKGROUND In ischemic stroke, increased glycated hemoglobin (HbA1c) and glucose levels are associated with worse outcome following thrombolysis, and possibly, endovascular thrombectomy. OBJECTIVE To evaluate the association between admission HbA1c and glucose levels and outcome following endovascular thrombectomy. METHODS Consecutive patients treated with endovascular thrombectomy with admission HbA1c and glucose levels were included. The primary outcome was functional independence, defined as a modified Rankin Scale score of 0-2 at 3 months. Secondary outcomes included successful reperfusion (modified Thrombolysis in Cerebral Infarction 2b-3), early neurological improvement (reduction in National Institutes of Health Stroke Scale (NIHSS) score ≥8 points, or NIHSS score of 0-1 at 24 hours), symptomatic intracerebral hemorrhage (sICH), and mortality at 3 months. RESULTS 223 patients (136 (61%) men; mean±SD age 64.5±14.6) were included. The median (IQR) HbA1c and glucose were 39 (36-45) mmol/mol and 6.9 (5.8-8.4) mmol/L, respectively. Multiple logistic regression analysis demonstrated that increasing HbA1c levels (per 10 mmol/mol) were associated with reduced functional independence (OR=0.76; 95% CI 0.60-0.96; p=0.02), increased sICH (OR=1.33; 95% CI 1.03 to 1.71; p=0.03), and increased mortality (OR=1.26; 95% CI 1.01 to 1.57; p=0.04). There were no significant associations between glucose levels and outcome measures (all p>0.05). CONCLUSIONS HbA1c levels are an independent predictor of worse outcome following endovascular thrombectomy. The addition of HbA1c to decision-support tools for endovascular thrombectomy should be evaluated in future studies.
Collapse
Affiliation(s)
- William K Diprose
- Department of Medicine, University of Auckland, Auckland, New Zealand.,Department of Neurology, Auckland City Hospital, Auckland, New Zealand
| | - Michael T M Wang
- Department of Medicine, University of Auckland, Auckland, New Zealand
| | - Andrew McFetridge
- Department of Radiology, Auckland City Hospital, Auckland, New Zealand
| | - James Sutcliffe
- Department of Radiology, Auckland City Hospital, Auckland, New Zealand
| | - P Alan Barber
- Department of Medicine, University of Auckland, Auckland, New Zealand.,Department of Neurology, Auckland City Hospital, Auckland, New Zealand
| |
Collapse
|
88
|
Abidi SSR, Abidi SR. Intelligent health data analytics: A convergence of artificial intelligence and big data. Healthc Manage Forum 2019; 32:178-182. [PMID: 31117831 DOI: 10.1177/0840470419846134] [Citation(s) in RCA: 15] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
Healthcare is a living system that generates a significant volume of heterogeneous data. As healthcare systems are pivoting to value-based systems, intelligent and interactive analysis of health data is gaining significance for health system management, especially for resource optimization whilst improving care quality and health outcomes. Health data analytics is being influenced by new concepts and intelligent methods emanating from artificial intelligence and big data. In this article, we contextualize health data and health data analytics in terms of the emerging trends of artificial intelligence and big data. We examine the nature of health data using the big data criterion to understand "how big" is health data. Next, we explain the working of artificial intelligence-based data analytics methods and discuss "what insights" can be derived from a broad spectrum of health data analytics methods to improve health system management, health outcomes, knowledge discovery, and healthcare innovation.
Collapse
Affiliation(s)
- Syed Sibte Raza Abidi
- 1 NICHE Research Group, Faculty of Computer Science, Dalhousie University, Halifax, Nova Scotia, Canada
| | - Samina Raza Abidi
- 2 NICHE Research Group, Department of Community Health and Epidemiology, Faculty of Medicine, Dalhousie University, Halifax, Nova Scotia, Canada
| |
Collapse
|
89
|
Barrett LA, Payrovnaziri SN, Bian J, He Z. Building Computational Models to Predict One-Year Mortality in ICU Patients with Acute Myocardial Infarction and Post Myocardial Infarction Syndrome. AMIA JOINT SUMMITS ON TRANSLATIONAL SCIENCE PROCEEDINGS. AMIA JOINT SUMMITS ON TRANSLATIONAL SCIENCE 2019; 2019:407-416. [PMID: 31258994 PMCID: PMC6568079] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Subscribe] [Scholar Register] [Indexed: 06/09/2023]
Abstract
Heart disease remains the leading cause of death in the United States. Compared with risk assessment guidelines that require manual calculation of scores, machine learning-based prediction for disease outcomes such as mortality can be utilized to save time and improve prediction accuracy. This study built and evaluated various machine learning models to predict one-year mortality in patients diagnosed with acute myocardial infarction or post myocardial infarction syndrome in the MIMIC-III database. The results of the best performing shallow prediction models were compared to a deep feedforward neural network (Deep FNN) with back propagation. We included a cohort of 5436 admissions. Six datasets were developed and compared. The models applying Logistic Model Trees (LMT) and Simple Logistic algorithms to the combined dataset resulted in the highest prediction accuracy at 85.12% and the highest AUC at .901. In addition, other factors were observed to have an impact on outcomes as well.
Collapse
Affiliation(s)
- Laura A Barrett
- School of Information, Florida State University, Tallahassee, Florida, USA
| | | | - Jiang Bian
- Department of Health Outcomes and Biomedical Informatics, University of Florida, Gainesville, Florida, USA
| | - Zhe He
- School of Information, Florida State University, Tallahassee, Florida, USA
| |
Collapse
|
90
|
Saber H, Somai M, Rajah GB, Scalzo F, Liebeskind DS. Predictive analytics and machine learning in stroke and neurovascular medicine. Neurol Res 2019; 41:681-690. [PMID: 31038007 DOI: 10.1080/01616412.2019.1609159] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/26/2022]
Abstract
Advances in predictive analytics and machine learning supported by an ever-increasing wealth of data and processing power are transforming almost every industry. Accuracy and precision of predictive analytics have significantly increased over the past few years and are evolving at an exponential pace. There have been significant breakthroughs in using Predictive Analytics in healthcare where it is held as the foundation of precision medicine. Yet, although the research in the field is expanding with the profuse volume of papers applying machine learning algorithms to medical data, very few have contributed meaningfully to clinical care. This lack of impact stands in stark contrast to the enormous relevance of machine learning to many other industries. Regardless of the status of its current contribution, the field of predictive analytics is expected to fundamentally change the way we diagnose and treat diseases, as well as the conduct of biomedical science research. In this review, we describe the main tools and techniques in predictive analytics and will analyze the trends in application of these techniques over the recent years. We will also provide examples of its application in medicine and more specifically in stroke and neurovascular research and outline current limitations.
Collapse
Affiliation(s)
- Hamidreza Saber
- a Wayne State Department of Neurology, Wayne State University , Detroit , MI , USA
| | - Melek Somai
- b Neuro-Epidemiology and Ageing Research Unit, School of Public Health, Imperial College London , London , UK
| | - Gary B Rajah
- c Wayne State Department of Neurosurgery, Wayne State University , Detroit , MI , USA
| | - Fabien Scalzo
- d Departement of Neurology, University of California Los Angeles , Los Angeles , CA , USA
| | - David S Liebeskind
- d Departement of Neurology, University of California Los Angeles , Los Angeles , CA , USA
| |
Collapse
|
91
|
Abstract
Clinical judgment to reach final diagnosis has remained a challenge since time immemorial. The present times are witness to artificial intelligence (AI) and machine learning programs competing to outperform the seasoned physician in arriving at a differential diagnosis. We discuss here the possible roles of AI in neurology.
Collapse
Affiliation(s)
- Venugopalan Y Vishnu
- Department of Neurology, All India Institute of Medical Sciences, New Delhi, India
| | | |
Collapse
|
92
|
JOURNAL CLUB: Use of Gradient Boosting Machine Learning to Predict Patient Outcome in Acute Ischemic Stroke on the Basis of Imaging, Demographic, and Clinical Information. AJR Am J Roentgenol 2019; 212:44-51. [DOI: 10.2214/ajr.18.20260] [Citation(s) in RCA: 46] [Impact Index Per Article: 9.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/18/2022]
|
93
|
Bang OY. Silent brain infarction: a quiet predictor of future stroke. PRECISION AND FUTURE MEDICINE 2018. [DOI: 10.23838/pfm.2018.00086] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2022] Open
|
94
|
Sakai K, Yamada K. Machine learning studies on major brain diseases: 5-year trends of 2014–2018. Jpn J Radiol 2018; 37:34-72. [DOI: 10.1007/s11604-018-0794-4] [Citation(s) in RCA: 73] [Impact Index Per Article: 12.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/21/2018] [Accepted: 11/14/2018] [Indexed: 12/17/2022]
|
95
|
Ramos LA, van der Steen WE, Sales Barros R, Majoie CBLM, van den Berg R, Verbaan D, Vandertop WP, Zijlstra IJAJ, Zwinderman AH, Strijkers GJ, Olabarriaga SD, Marquering HA. Machine learning improves prediction of delayed cerebral ischemia in patients with subarachnoid hemorrhage. J Neurointerv Surg 2018; 11:497-502. [DOI: 10.1136/neurintsurg-2018-014258] [Citation(s) in RCA: 25] [Impact Index Per Article: 4.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/16/2018] [Revised: 10/09/2018] [Accepted: 10/11/2018] [Indexed: 01/04/2023]
Abstract
Background and purposeDelayed cerebral ischemia (DCI) is a severe complication in patients with aneurysmal subarachnoid hemorrhage. Several associated predictors have been previously identified. However, their predictive value is generally low. We hypothesize that Machine Learning (ML) algorithms for the prediction of DCI using a combination of clinical and image data lead to higher predictive accuracy than previously applied logistic regressions.Materials and methodsClinical and baseline CT image data from 317 patients with aneurysmal subarachnoid hemorrhage were included. Three types of analysis were performed to predict DCI. First, the prognostic value of known predictors was assessed with logistic regression models. Second, ML models were created using all clinical variables. Third, image features were extracted from the CT images using an auto-encoder and combined with clinical data to create ML models. Accuracy was evaluated based on the area under the curve (AUC), sensitivity and specificity with 95% CI.ResultsThe best AUC of the logistic regression models for known predictors was 0.63 (95% CI 0.62 to 0.63). For the ML algorithms with clinical data there was a small but statistically significant improvement in the AUC to 0.68 (95% CI 0.65 to 0.69). Notably, aneurysm width and height were included in many of the ML models. The AUC was highest for ML models that also included image features: 0.74 (95% CI 0.72 to 0.75).ConclusionML algorithms significantly improve the prediction of DCI in patients with aneurysmal subarachnoid hemorrhage, particularly when image features are also included. Our experiments suggest that aneurysm characteristics are also associated with the development of DCI.
Collapse
|
96
|
Kamal H, Lopez V, Sheth SA. Machine Learning in Acute Ischemic Stroke Neuroimaging. Front Neurol 2018; 9:945. [PMID: 30467491 PMCID: PMC6236025 DOI: 10.3389/fneur.2018.00945] [Citation(s) in RCA: 47] [Impact Index Per Article: 7.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/26/2018] [Accepted: 10/22/2018] [Indexed: 01/14/2023] Open
Abstract
Machine Learning (ML) through pattern recognition algorithms is currently becoming an essential aid for the diagnosis, treatment, and prediction of complications and patient outcomes in a number of neurological diseases. The evaluation and treatment of Acute Ischemic Stroke (AIS) have experienced a significant advancement over the past few years, increasingly requiring the use of neuroimaging for decision-making. In this review, we offer an insight into the recent developments and applications of ML in neuroimaging focusing on acute ischemic stroke.
Collapse
Affiliation(s)
- Haris Kamal
- Department of Neurology, University of Texas at Houston Health Science Center, Houston, TX, United States
| | | | | |
Collapse
|
97
|
Vargas J, Spiotta A, Chatterjee AR. Initial Experiences with Artificial Neural Networks in the Detection of Computed Tomography Perfusion Deficits. World Neurosurg 2018; 124:S1878-8750(18)32382-9. [PMID: 30366140 DOI: 10.1016/j.wneu.2018.10.084] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/31/2018] [Revised: 10/12/2018] [Accepted: 10/15/2018] [Indexed: 10/28/2022]
Abstract
BACKGROUND Head computed tomography (CT) with perfusion imaging has become crucial in the selection of patients for mechanical thrombectomy. In recent years, machine learning has rapidly evolved and found applications in a wide variety of health care tasks. We report our initial experiences with training a neural network to predict the presence and laterality of a perfusion deficit in patients with acute ischemic stroke. METHODS CT perfusion images of patients with suspicion for acute ischemic stroke were obtained. The data were split into training and validation sets. A long-term, recurrent convolutional network was constructed, which consisted of a convolutional neural network stacked on top of a long short-term memory layer. RESULTS Of the 396 patients, 139 (35.1%) had a right-sided perfusion deficit, 199 (50.3%) had a left-sided deficit, and 58 (14.6%) had no evidence of a deficit. The best model was able to achieve an accuracy of 85.8% on validation data. Receiver operating characteristic curves were generated for each class, and an area under the curve (AUC) was calculated for each class. For right-sided deficits, the AUC was 0.90, for left-sided deficits, the AUC was 0.96, and for no deficit, the AUC was 0.93. CONCLUSIONS The field of machine learning, powered by convolutional neural networks for the task of image recognition and processing, has quickly developed in recent years. We constructed an artificial neural network that can identify and classify the presence and laterality of a perfusion deficit on CT perfusion imaging.
Collapse
Affiliation(s)
- Jan Vargas
- Greenville Health System Division of Neuroendovascular Surgery, Greenville, South Carolina, USA.
| | - Alejandro Spiotta
- Department of Neurosurgery, Medical University of South Carolina, Charleston, South Carolina, USA
| | | |
Collapse
|
98
|
Peering Into the Black Box of Artificial Intelligence: Evaluation Metrics of Machine Learning Methods. AJR Am J Roentgenol 2018; 212:38-43. [PMID: 30332290 DOI: 10.2214/ajr.18.20224] [Citation(s) in RCA: 98] [Impact Index Per Article: 16.3] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/12/2022]
Abstract
OBJECTIVE Machine learning (ML) and artificial intelligence (AI) are rapidly becoming the most talked about and controversial topics in radiology and medicine. Over the past few years, the numbers of ML- or AI-focused studies in the literature have increased almost exponentially, and ML has become a hot topic at academic and industry conferences. However, despite the increased awareness of ML as a tool, many medical professionals have a poor understanding of how ML works and how to critically appraise studies and tools that are presented to us. Thus, we present a brief overview of ML, explain the metrics used in ML and how to interpret them, and explain some of the technical jargon associated with the field so that readers with a medical background and basic knowledge of statistics can feel more comfortable when examining ML applications. CONCLUSION Attention to sample size, overfitting, underfitting, cross validation, as well as a broad knowledge of the metrics of machine learning, can help those with little or no technical knowledge begin to assess machine learning studies. However, transparency in methods and sharing of algorithms is vital to allow clinicians to assess these tools themselves.
Collapse
|
99
|
Park E, Chang HJ, Nam HS. A Bayesian Network Model for Predicting Post-stroke Outcomes With Available Risk Factors. Front Neurol 2018; 9:699. [PMID: 30245663 PMCID: PMC6137617 DOI: 10.3389/fneur.2018.00699] [Citation(s) in RCA: 22] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/30/2018] [Accepted: 08/02/2018] [Indexed: 11/13/2022] Open
Abstract
Bayesian network is an increasingly popular method in modeling uncertain and complex problems, because its interpretability is often more useful than plain prediction. To satisfy the core requirement in medical research to obtain interpretable prediction with high accuracy, we constructed an inference engine for post-stroke outcomes based on Bayesian network classifiers. The prediction system that was trained on data of 3,605 patients with acute stroke forecasts the functional independence at 3 months and the mortality 1 year after stroke. Feature selection methods were applied to eliminate less relevant and redundant features from 76 risk variables. The Bayesian network classifiers were trained with a hill-climbing searching for the qualified network structure and parameters measured by maximum description length. We evaluated and optimized the proposed system to increase the area under the receiver operating characteristic curve (AUC) while ensuring acceptable sensitivity for the class-imbalanced data. The performance evaluation demonstrated that the Bayesian network with selected features by wrapper-type feature selection can predict 3-month functional independence with an AUC of 0.889 using only 19 risk variables and 1-year mortality with an AUC of 0.893 using 24 variables. The Bayesian network with 50 features filtered by information gain can predict 3-month functional independence with an AUC of 0.875 and 1-year mortality with an AUC of 0.895. We also built an online prediction service, Yonsei Stroke Outcome Inference System, to substantialize the proposed solution for patients with stroke.
Collapse
Affiliation(s)
- Eunjeong Park
- Cardiovascular Research Institute, College of Medicine, Yonsei University, Seoul, South Korea
| | - Hyuk-Jae Chang
- Department of Cardiology, College of Medicine, Yonsei University, Seoul, South Korea
| | - Hyo Suk Nam
- Department of Neurology, College of Medicine, Yonsei University, Seoul, South Korea
| |
Collapse
|
100
|
Liakos KG, Busato P, Moshou D, Pearson S, Bochtis D. Machine Learning in Agriculture: A Review. SENSORS (BASEL, SWITZERLAND) 2018; 18:E2674. [PMID: 30110960 PMCID: PMC6111295 DOI: 10.3390/s18082674] [Citation(s) in RCA: 362] [Impact Index Per Article: 60.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 06/27/2018] [Revised: 07/31/2018] [Accepted: 08/07/2018] [Indexed: 11/16/2022]
Abstract
Machine learning has emerged with big data technologies and high-performance computing to create new opportunities for data intensive science in the multi-disciplinary agri-technologies domain. In this paper, we present a comprehensive review of research dedicated to applications of machine learning in agricultural production systems. The works analyzed were categorized in (a) crop management, including applications on yield prediction, disease detection, weed detection crop quality, and species recognition; (b) livestock management, including applications on animal welfare and livestock production; (c) water management; and (d) soil management. The filtering and classification of the presented articles demonstrate how agriculture will benefit from machine learning technologies. By applying machine learning to sensor data, farm management systems are evolving into real time artificial intelligence enabled programs that provide rich recommendations and insights for farmer decision support and action.
Collapse
Affiliation(s)
- Konstantinos G Liakos
- Institute for Bio-Economy and Agri-Technology (IBO), Centre of Research and Technology-Hellas (CERTH), 6th km Charilaou-Thermi Rd, GR 57001 Thessaloniki, Greece.
| | - Patrizia Busato
- Department of Agriculture, Forestry and Food Sciences (DISAFA), Faculty of Agriculture, University of Turin, Largo Braccini 2, 10095 Grugliasco, Italy.
| | - Dimitrios Moshou
- Institute for Bio-Economy and Agri-Technology (IBO), Centre of Research and Technology-Hellas (CERTH), 6th km Charilaou-Thermi Rd, GR 57001 Thessaloniki, Greece.
- Agricultural Engineering Laboratory, Faculty of Agriculture, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greece.
| | - Simon Pearson
- Lincoln Institute for Agri-food Technology (LIAT), University of Lincoln, Brayford Way, Brayford Pool, Lincoln LN6 7TS, UK, .
| | - Dionysis Bochtis
- Institute for Bio-Economy and Agri-Technology (IBO), Centre of Research and Technology-Hellas (CERTH), 6th km Charilaou-Thermi Rd, GR 57001 Thessaloniki, Greece.
| |
Collapse
|