Artificial neural networks in medical diagnosis

Perceptions of undergraduate medical students on artificial intelligence in medicine: mixed-methods survey study from Palestine

BACKGROUND

The current applications of artificial intelligence (AI) in medicine continue to attract the attention of medical students. This study aimed to identify undergraduate medical students' attitudes toward AI in medicine, explore present AI-related training opportunities, investigate the need for AI inclusion in medical curricula, and determine preferred methods for teaching AI curricula.

METHODS

This study uses a mixed-method cross-sectional design, including a quantitative study and a qualitative study, targeting Palestinian undergraduate medical students in the academic year 2022-2023. In the quantitative part, we recruited a convenience sample of undergraduate medical students from universities in Palestine from June 15, 2022, to May 30, 2023. We collected data by using an online, well-structured, and self-administered questionnaire with 49 items. In the qualitative part, 15 undergraduate medical students were interviewed by trained researchers. Descriptive statistics and an inductive content analysis approach were used to analyze quantitative and qualitative data, respectively.

RESULTS

From a total of 371 invitations sent, 362 responses were received (response rate = 97.5%), and 349 were included in the analysis. The mean age of participants was 20.38 ± 1.97, with 40.11% (140) in their second year of medical school. Most participants (268, 76.79%) did not receive formal education on AI before or during medical study. About two-thirds of students strongly agreed or agreed that AI would become common in the future (67.9%, 237) and would revolutionize medical fields (68.7%, 240). Participants stated that they had not previously acquired training in the use of AI in medicine during formal medical education (260, 74.5%), confirming a dire need to include AI training in medical curricula (247, 70.8%). Most participants (264, 75.7%) think that learning opportunities for AI in medicine have not been adequate; therefore, it is very important to study more about employing AI in medicine (228, 65.3%). Male students (3.15 ± 0.87) had higher perception scores than female students (2.81 ± 0.86) (p < 0.001). The main themes that resulted from the qualitative analysis of the interview questions were an absence of AI learning opportunities, the necessity of including AI in medical curricula, optimism towards the future of AI in medicine, and expected challenges related to AI in medical fields.

CONCLUSION

Medical students lack access to educational opportunities for AI in medicine; therefore, AI should be included in formal medical curricula in Palestine.

Auto-BCS: A Hybrid System for Real-Time Breast Cancer Screening from Pathological Images

Breast cancer is recognized as a prominent cause of cancer-related mortality among women globally, emphasizing the critical need for early diagnosis resulting improvement in survival rates. Current breast cancer diagnostic procedures depend on manual assessments of pathological images by medical professionals. However, in remote or underserved regions, the scarcity of expert healthcare resources often compromised the diagnostic accuracy. Machine learning holds great promise for early detection, yet existing breast cancer screening algorithms are frequently characterized by significant computational demands, rendering them unsuitable for deployment on low-processing-power mobile devices. In this paper, a real-time automated system "Auto-BCS" is introduced that significantly enhances the efficiency of early breast cancer screening. The system is structured into three distinct phases. In the initial phase, images undergo a pre-processing stage aimed at noise reduction. Subsequently, feature extraction is carried out using a lightweight and optimized deep learning model followed by extreme gradient boosting classifier, strategically employed to optimize the overall performance and prevent overfitting in the deep learning model. The system's performance is gauged through essential metrics, including accuracy, precision, recall, F1 score, and inference time. Comparative evaluations against state-of-the-art algorithms affirm that Auto-BCS outperforms existing models, excelling in both efficiency and processing speed. Computational efficiency is prioritized by Auto-BCS, making it particularly adaptable to low-processing-power mobile devices. Comparative assessments confirm the superior performance of Auto-BCS, signifying its potential to advance breast cancer screening technology.

Hemochromatosis: Ferroptosis, ROS, Gut Microbiome, and Clinical Challenges with Alcohol as Confounding Variable

Hemochromatosis represents clinically one of the most important genetic storage diseases of the liver caused by iron overload, which is to be differentiated from hepatic iron overload due to excessive iron release from erythrocytes in patients with genetic hemolytic disorders. This disorder is under recent mechanistic discussion regarding ferroptosis, reactive oxygen species (ROS), the gut microbiome, and alcohol abuse as a risk factor, which are all topics of this review article. Triggered by released intracellular free iron from ferritin via the autophagic process of ferritinophagy, ferroptosis is involved in hemochromatosis as a specific form of iron-dependent regulated cell death. This develops in the course of mitochondrial injury associated with additional iron accumulation, followed by excessive production of ROS and lipid peroxidation. A low fecal iron content during therapeutic iron depletion reduces colonic inflammation and oxidative stress. In clinical terms, iron is an essential trace element required for human health. Humans cannot synthesize iron and must take it up from iron-containing foods and beverages. Under physiological conditions, healthy individuals allow for iron homeostasis by restricting the extent of intestinal iron depending on realistic demand, avoiding uptake of iron in excess. For this condition, the human body has no chance to adequately compensate through removal. In patients with hemochromatosis, the molecular finetuning of intestinal iron uptake is set off due to mutations in the high-FE2+ (HFE) genes that lead to a lack of hepcidin or resistance on the part of ferroportin to hepcidin binding. This is the major mechanism for the increased iron stores in the body. Hepcidin is a liver-derived peptide, which impairs the release of iron from enterocytes and macrophages by interacting with ferroportin. As a result, iron accumulates in various organs including the liver, which is severely injured and causes the clinically important hemochromatosis. This diagnosis is difficult to establish due to uncharacteristic features. Among these are asthenia, joint pain, arthritis, chondrocalcinosis, diabetes mellitus, hypopituitarism, hypogonadotropic hypogonadism, and cardiopathy. Diagnosis is initially suspected by increased serum levels of ferritin, a non-specific parameter also elevated in inflammatory diseases that must be excluded to be on the safer diagnostic side. Diagnosis is facilitated if ferritin is combined with elevated fasting transferrin saturation, genetic testing, and family screening. Various diagnostic attempts were published as algorithms. However, none of these were based on evidence or quantitative results derived from scored key features as opposed to other known complex diseases. Among these are autoimmune hepatitis (AIH) or drug-induced liver injury (DILI). For both diseases, the scored diagnostic algorithms are used in line with artificial intelligence (AI) principles to ascertain the diagnosis. The first-line therapy of hemochromatosis involves regular and life-long phlebotomy to remove iron from the blood, which improves the prognosis and may prevent the development of end-stage liver disease such as cirrhosis and hepatocellular carcinoma. Liver transplantation is rarely performed, confined to acute liver failure. In conclusion, ferroptosis, ROS, the gut microbiome, and concomitant alcohol abuse play a major contributing role in the development and clinical course of genetic hemochromatosis, which requires early diagnosis and therapy initiation through phlebotomy as a first-line treatment.

Intelligent systems for sitting posture monitoring and anomaly detection: an overview

The number of people who need to use wheelchair for proper mobility is increasing. The integration of technology into these devices enables the simultaneous and objective assessment of posture, while also facilitating the concurrent monitoring of the functional status of wheelchair users. In this way, both the health personnel and the user can be provided with relevant information for the recovery process. This information can be used to carry out an early adaptation of the rehabilitation of patients, thus allowing to prevent further musculoskeletal problems, as well as risk situations such as ulcers or falls. Thus, a higher quality of life is promoted in affected individuals. As a result, this paper presents an orderly and organized analysis of the existing postural diagnosis systems for detecting sitting anomalies in the literature. This analysis can be divided into two parts that compose such postural diagnosis: on the one hand, the monitoring devices necessary for the collection of postural data and, on the other hand, the techniques used for anomaly detection. These anomaly detection techniques will be explained under two different approaches: the traditional generalized approach followed to date by most works, where anomalies are treated as incorrect postures, and a new individualized approach treating anomalies as changes with respect to the normal sitting pattern. In this way, the advantages, limitations and opportunities of the different techniques are analyzed. The main contribution of this overview paper is to synthesize and organize information, identify trends, and provide a comprehensive understanding of sitting posture diagnosis systems, offering researchers an accessible resource for navigating the current state of knowledge of this particular field.

Penile Erection Morphometry: The Need for a Novel Approach

For many males, sexual function holds significant value in determining their quality of life. Despite the importance of male erectile function, no quantitative method to measure it accurately is currently available. Standardized assessment methods such as RigiScan™, International Index of Erectile Function (IIEF-5), and the stamp test are used to evaluate sexual function, but those methods cannot repetitively and quantitatively measure erectile function. Only direct measurement can quantitatively assess the shape of an erect penis. This paper presents the essential requirements for developing an ideal measurement method for penile erection. It also introduces current approaches for diagnosing male sexual function and reviews ongoing research to quantitatively measure erectile function. The paper further summarizes and analyzes the advantages and disadvantages of each method with respect to the essential requirements. Finally, the paper discusses the future direction toward the development of Penile Erection Morphometry.

Predicting mortality in brain stroke patients using neural networks: outcomes analysis in a longitudinal study

In this study, Neural Networks (NN) modelling has emerged as a promising tool for predicting outcomes in patients with Brain Stroke (BS) by identifying key risk factors. In this longitudinal study, we enrolled 332 patients form Imam hospital in Ardabil, Iran, with mean age: 77.4 (SD 10.4) years, and 50.6% were male. Diagnosis of BS was confirmed using both computerized tomography scan and magnetic resonance imaging, and risk factor and outcome data were collected from the hospital's BS registry, and by telephone follow-up over a period of 10 years, respectively. Using a multilayer perceptron NN approach, we analysed the impact of various risk factors on time to mortality and mortality from BS. A total of 100 NN classification algorithm were trained utilizing STATISTICA 13 software, and the optimal model was selected for further analysis based on their diagnostic performance. We also calculated Kaplan-Meier survival probabilities and conducted Log-rank tests. The five selected NN models exhibited impressive accuracy ranges of 81-85%. However, the optimal model stood out for its superior diagnostic indices. Mortality rate in the training and the validation data set was 7.9 (95% CI 5.7-11.0) per 1000 and 8.2 (7.1-9.6) per 1000, respectively (P = 0.925). The optimal model highlighted significant risk factors for BS mortality, including smoking, lower education, advanced age, lack of physical activity, a history of diabetes, all carrying substantial importance weights. Our study provides compelling evidence that the NN approach is highly effective in predicting mortality in patients with BS based on key risk factors, and has the potential to significantly enhance the accuracy of prediction. Moreover, our findings could inform more effective prevention strategies for BS, ultimately leading to better patient outcomes.

Using Neural Networks Algorithm in Ischemic Stroke Diagnosis: A Systematic Review

Objective

To evaluate the evidence of artificial neural network (NNs) techniques in diagnosing ischemic stroke (IS) in adults.

Methods

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was utilized as a guideline for this review. PubMed, MEDLINE, Web of Science, and CINAHL Plus Full Text were searched to identify studies published between 2018 and 2022, reporting using NNs in IS diagnosis. The Critical Appraisal Checklist for Diagnostic Test Accuracy Studies was adopted to evaluate the included studies.

Results

Nine studies were included in this systematic review. Non-contrast computed tomography (NCCT) (n = 4 studies, 26.67%) and computed tomography angiography (CTA) (n = 4 studies, 26.67%) are among the most common features. Five algorithms were used in the included studies. Deep Convolutional Neural Networks (DCNNs) were commonly used for IS diagnosis (n = 3 studies, 33.33%). Other algorithms including three-dimensional convolutional neural networks (3D-CNNs) (n = 2 studies, 22.22%), two-stage deep convolutional neural networks (Two-stage DCNNs) (n = 2 studies, 22.22%), the local higher-order singular value decomposition denoising algorithm (GL-HOSVD) (n = 1 study, 11.11%), and a new deconvolution network model based on deep learning (AD-CNNnet) (n = 1 study, 11.11%) were also utilized for the diagnosis of IS.

Conclusion

The number of studies ensuring the effectiveness of NNs algorithms in IS diagnosis has increased. Still, more feasibility and cost-effectiveness evaluations are needed to support the implementation of NNs in IS diagnosis in clinical settings.

Biomedical Ethical Aspects Towards the Implementation of Artificial Intelligence in Medical Education

The increasing use of artificial intelligence (AI) in medicine is associated with new ethical challenges and responsibilities. However, special considerations and concerns should be addressed when integrating AI applications into medical education, where healthcare, AI, and education ethics collide. This commentary explores the biomedical ethical responsibilities of medical institutions in incorporating AI applications into medical education by identifying potential concerns and limitations, with the goal of implementing applicable recommendations. The recommendations presented are intended to assist in developing institutional guidelines for the ethical use of AI for medical educators and students.

Artificial intelligence modelling to assess the risk of cardiovascular disease in oncology patients

Aims

There are no comprehensive machine learning (ML) tools used by oncologists to assist with risk identification and referrals to cardio-oncology. This study applies ML algorithms to identify oncology patients at risk for cardiovascular disease for referrals to cardio-oncology and to generate risk scores to support quality of care.

Methods and results

De-identified patient data were obtained from Vanderbilt University Medical Center. Patients with breast, kidney, and B-cell lymphoma cancers were targeted. Additionally, the study included patients who received immunotherapy drugs for treatment of melanoma, lung cancer, or kidney cancer. Random forest (RF) and artificial neural network (ANN) ML models were applied to analyse each cohort: A total of 20 023 records were analysed (breast cancer, 6299; B-cell lymphoma, 9227; kidney cancer, 2047; and immunotherapy for three covered cancers, 2450). Data were divided randomly into training (80%) and test (20%) data sets. Random forest and ANN performed over 90% for accuracy and area under the curve (AUC). All ANN models performed better than RF models and produced accurate referrals.

Conclusion

Predictive models are ready for translation into oncology practice to identify and care for patients who are at risk of cardiovascular disease. The models are being integrated with electronic health record application as a report of patients who should be referred to cardio-oncology for monitoring and/or tailored treatments. Models operationally support cardio-oncology practice. Limited validation identified 86% of the lymphoma and 58% of the kidney cancer patients with major risk for cardiotoxicity who were not referred to cardio-oncology.

Machine Learning Methods for Small Data Challenges in Molecular Science

Small data are often used in scientific and engineering research due to the presence of various constraints, such as time, cost, ethics, privacy, security, and technical limitations in data acquisition. However, big data have been the focus for the past decade, small data and their challenges have received little attention, even though they are technically more severe in machine learning (ML) and deep learning (DL) studies. Overall, the small data challenge is often compounded by issues, such as data diversity, imputation, noise, imbalance, and high-dimensionality. Fortunately, the current big data era is characterized by technological breakthroughs in ML, DL, and artificial intelligence (AI), which enable data-driven scientific discovery, and many advanced ML and DL technologies developed for big data have inadvertently provided solutions for small data problems. As a result, significant progress has been made in ML and DL for small data challenges in the past decade. In this review, we summarize and analyze several emerging potential solutions to small data challenges in molecular science, including chemical and biological sciences. We review both basic machine learning algorithms, such as linear regression, logistic regression (LR), k-nearest neighbor (KNN), support vector machine (SVM), kernel learning (KL), random forest (RF), and gradient boosting trees (GBT), and more advanced techniques, including artificial neural network (ANN), convolutional neural network (CNN), U-Net, graph neural network (GNN), Generative Adversarial Network (GAN), long short-term memory (LSTM), autoencoder, transformer, transfer learning, active learning, graph-based semi-supervised learning, combining deep learning with traditional machine learning, and physical model-based data augmentation. We also briefly discuss the latest advances in these methods. Finally, we conclude the survey with a discussion of promising trends in small data challenges in molecular science.

Classification of imbalanced data using machine learning algorithms to predict the risk of renal graft failures in Ethiopia

INTRODUCTION

The prevalence of end-stage renal disease has raised the need for renal replacement therapy over recent decades. Even though a kidney transplant offers an improved quality of life and lower cost of care than dialysis, graft failure is possible after transplantation. Hence, this study aimed to predict the risk of graft failure among post-transplant recipients in Ethiopia using the selected machine learning prediction models.

METHODOLOGY

The data was extracted from the retrospective cohort of kidney transplant recipients at the Ethiopian National Kidney Transplantation Center from September 2015 to February 2022. In response to the imbalanced nature of the data, we performed hyperparameter tuning, probability threshold moving, tree-based ensemble learning, stacking ensemble learning, and probability calibrations to improve the prediction results. Merit-based selected probabilistic (logistic regression, naive Bayes, and artificial neural network) and tree-based ensemble (random forest, bagged tree, and stochastic gradient boosting) models were applied. Model comparison was performed in terms of discrimination and calibration performance. The best-performing model was then used to predict the risk of graft failure.

RESULTS

A total of 278 completed cases were analyzed, with 21 graft failures and 3 events per predictor. Of these, 74.8% are male, and 25.2% are female, with a median age of 37. From the comparison of models at the individual level, the bagged tree and random forest have top and equal discrimination performance (AUC-ROC = 0.84). In contrast, the random forest has the best calibration performance (brier score = 0.045). Under testing the individual model as a meta-learner for stacking ensemble learning, the result of stochastic gradient boosting as a meta-learner has the top discrimination (AUC-ROC = 0.88) and calibration (brier score = 0.048) performance. Regarding feature importance, chronic rejection, blood urea nitrogen, number of post-transplant admissions, phosphorus level, acute rejection, and urological complications are the top predictors of graft failure.

CONCLUSIONS

Bagging, boosting, and stacking, with probability calibration, are good choices for clinical risk predictions working on imbalanced data. The data-driven probability threshold is more beneficial than the natural threshold of 0.5 to improve the prediction result from imbalanced data. Integrating various techniques in a systematic framework is a smart strategy to improve prediction results from imbalanced data. It is recommended for clinical experts in kidney transplantation to use the final calibrated model as a decision support system to predict the risk of graft failure for individual patients.

Responsible Use of Artificial Intelligence in Dentistry: Survey on Dentists' and Final-Year Undergraduates' Perspectives

The introduction of artificial intelligence (AI)-based dental applications into clinical practice could play a significant role in improving diagnostic accuracy and reforming dental care, but its implementation relies on the readiness of dentists, as well as the health system, to adopt it in everyday practice. A cross-sectional anonymous online survey was conducted among experienced dentists and final-year undergraduate students from the School of Dental Medicine at the University of Belgrade (n = 281) in order to investigate their current perspectives and readiness to accept AI into practice. Responders (n = 193) in the present survey, especially final-year undergraduates (n = 76), showed a lack of knowledge about AI (only 7.9% of them were familiar with AI use) and were skeptical (only 34% of them believed that AI should be used), and the underlying reasons, as shown by logistic regression analyses, were a lack of knowledge about the AI technology associated with a fear of being replaced by AI, as well as a lack of regulatory policy. Female dentists perceived ethical issues more significantly than men regarding AI implementation in the practice. The present results encourage an ethical debate on education/training and regulatory policies for AI as a prerequisite for regular AI use in dental practice.

Breath analysis by ultra-sensitive broadband laser spectroscopy detects SARS-CoV-2 infection

Rapid testing is essential to fighting pandemics such as coronavirus disease 2019 (COVID-19), the disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Exhaled human breath contains multiple volatile molecules providing powerful potential for non-invasive diagnosis of diverse medical conditions. We investigated breath detection of SARS-CoV-2 infection using cavity-enhanced direct frequency comb spectroscopy (CE-DFCS), a state-of-the-art laser spectroscopic technique capable of a real-time massive collection of broadband molecular absorption features at ro-vibrational quantum state resolution and at parts-per-trillion volume detection sensitivity. Using a total of 170 individual breath samples (83 positive and 87 negative with SARS-CoV-2 based on reverse transcription polymerase chain reaction tests), we report excellent discrimination capability for SARS-CoV-2 infection with an area under the receiver-operating-characteristics curve of 0.849(4). Our results support the development of CE-DFCS as an alternative, rapid, non-invasive test for COVID-19 and highlight its remarkable potential for optical diagnoses of diverse biological conditions and disease states.

Fully Complex-Valued Dendritic Neuron Model

A single dendritic neuron model (DNM) that owns the nonlinear information processing ability of dendrites has been widely used for classification and prediction. Complex-valued neural networks that consist of a number of multiple/deep-layer McCulloch-Pitts neurons have achieved great successes so far since neural computing was utilized for signal processing. Yet no complex value representations appear in single neuron architectures. In this article, we first extend DNM from a real-value domain to a complex-valued one. Performance of complex-valued DNM (CDNM) is evaluated through a complex XOR problem, a non-minimum phase equalization problem, and a real-world wind prediction task. Also, a comparative analysis on a set of elementary transcendental functions as an activation function is implemented and preparatory experiments are carried out for determining hyperparameters. The experimental results indicate that the proposed CDNM significantly outperforms real-valued DNM, complex-valued multi-layer perceptron, and other complex-valued neuron models.

Machine learning at the edge for AI-enabled multiplexed pathogen detection

Multiplexed detection of biomarkers in real-time is crucial for sensitive and accurate diagnosis at the point of use. This scenario poses tremendous challenges for detection and identification of signals of varying shape and quality at the edge of the signal-to-noise limit. Here, we demonstrate a robust target identification scheme that utilizes a Deep Neural Network (DNN) for multiplex detection of single particles and molecular biomarkers. The model combines fast wavelet particle detection with Short-Time Fourier Transform analysis, followed by DNN identification on an AI-specific edge device (Google Coral Dev board). The approach is validated using multi-spot optical excitation of Klebsiella Pneumoniae bacterial nucleic acids flowing through an optofluidic waveguide chip that produces fluorescence signals of varying amplitude, duration, and quality. Amplification-free 3× multiplexing in real-time is demonstrated with excellent specificity, sensitivity, and a classification accuracy of 99.8%. These results show that a minimalistic DNN design optimized for mobile devices provides a robust framework for accurate pathogen detection using compact, low-cost diagnostic devices.

Artificial Intelligence in Resuscitation: A Scoping Review

Introduction: Cardiac arrest is a significant cause of premature mortality and severe disability. Despite the death rate steadily decreasing over the previous decade, only 22% of survivors achieve good clinical status and only 25% of patients survive until their discharge from the hospital. The objective of this scoping review was to review relevant AI modalities and the main potential applications of AI in resuscitation. Methods: We conducted the literature search for related studies in PubMed, EMBASE, and Google Scholar. We included peer-reviewed publications and articles in the press, pooling and characterizing the data by their model types, goals, and benefits. Results: After identifying 268 original studies, we chose 59 original studies (reporting 1,817,419 patients) to include in the qualitative synthesis. AI-based methods appear to be superior to traditional methods in achieving high-level performance. Conclusion: AI might be useful in predicting cardiac arrest, heart rhythm disorders, and post-cardiac arrest outcomes, as well as in the delivery of drone-delivered defibrillators and notification of dispatchers. AI-powered technologies could be valuable assistants to continuously track patient conditions. Healthcare professionals should assist in the research and development of AI-powered technologies as well as their implementation into clinical practice.

Machine learning-based model for predicting the esophagogastric variceal bleeding risk in liver cirrhosis patients

BACKGROUND

Liver cirrhosis patients are at risk for esophagogastric variceal bleeding (EGVB). Herein, we aimed to estimate the EGVB risk in patients with liver cirrhosis using an artificial neural network (ANN).

METHODS

We included 999 liver cirrhosis patients hospitalized at the Beijing Ditan Hospital, Capital Medical University in the training cohort and 101 patients from Shuguang Hospital in the validation cohort. The factors independently affecting EGVB occurrence were determined via univariate analysis and used to develop an ANN model.

RESULTS

The 1-year cumulative EGVB incidence rates were 11.9 and 11.9% in the training and validation groups, respectively. A total of 12 independent risk factors, including gender, drinking and smoking history, decompensation, ascites, location and size of varices, alanine aminotransferase (ALT), γ-glutamyl transferase (GGT), hematocrit (HCT) and neutrophil-lymphocyte ratio (NLR) levels as well as red blood cell (RBC) count were evaluated and used to establish the ANN model, which estimated the 1-year EGVB risk. The ANN model had an area under the curve (AUC) of 0.959, which was significantly higher than the AUC for the North Italian Endoscopic Club (NIEC) (0.669) and revised North Italian Endoscopic Club (Rev-NIEC) indices (0.725) (all P <  0.001). Decision curve analyses revealed improved net benefits of the ANN compared to the NIEC and Rev-NIEC indices.

CONCLUSIONS

The ANN model accurately predicted the 1-year risk for EGVB in liver cirrhosis patients and might be used as a basis for risk-based EGVB surveillance strategies.

Risk factor identification and prediction models for prolonged length of stay in hospital after acute ischemic stroke using artificial neural networks

Background

Accurate estimation of prolonged length of hospital stay after acute ischemic stroke provides crucial information on medical expenditure and subsequent disposition. This study used artificial neural networks to identify risk factors and build prediction models for a prolonged length of stay based on parameters at the time of hospitalization.

Methods

We retrieved the medical records of patients who received acute ischemic stroke diagnoses and were treated at a stroke center between January 2016 and June 2020, and a retrospective analysis of these data was performed. Prolonged length of stay was defined as a hospital stay longer than the median number of days. We applied artificial neural networks to derive prediction models using parameters associated with the length of stay that was collected at admission, and a sensitivity analysis was performed to assess the effect of each predictor. We applied 5-fold cross-validation and used the validation set to evaluate the classification performance of the artificial neural network models.

Results

Overall, 2,240 patients were enrolled in this study. The median length of hospital stay was 9 days. A total of 1,101 patients (49.2%) had a prolonged hospital stay. A prolonged length of stay is associated with worse neurological outcomes at discharge. Univariate analysis identified 14 baseline parameters associated with prolonged length of stay, and with these parameters as input, the artificial neural network model achieved training and validation areas under the curve of 0.808 and 0.788, respectively. The mean accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of prediction models were 74.5, 74.9, 74.2, 75.2, and 73.9%, respectively. The key factors associated with prolonged length of stay were National Institutes of Health Stroke Scale scores at admission, atrial fibrillation, receiving thrombolytic therapy, history of hypertension, diabetes, and previous stroke.

Conclusion

The artificial neural network model achieved adequate discriminative power for predicting prolonged length of stay after acute ischemic stroke and identified crucial factors associated with a prolonged hospital stay. The proposed model can assist in clinically assessing the risk of prolonged hospitalization, informing decision-making, and developing individualized medical care plans for patients with acute ischemic stroke.

Toward Predicting 30-Day Readmission Among Oncology Patients: Identifying Timely and Actionable Risk Factors

PURPOSE

Predicting 30-day readmission risk is paramount to improving the quality of patient care. In this study, we compare sets of patient-, provider-, and community-level variables that are available at two different points of a patient's inpatient encounter (first 48 hours and the full encounter) to train readmission prediction models and identify possible targets for appropriate interventions that can potentially reduce avoidable readmissions.

METHODS

Using electronic health record data from a retrospective cohort of 2,460 oncology patients and a comprehensive machine learning analysis pipeline, we trained and tested models predicting 30-day readmission on the basis of data available within the first 48 hours of admission and from the entire hospital encounter.

RESULTS

Leveraging all features, the light gradient boosting model produced higher, but comparable performance (area under receiver operating characteristic curve [AUROC]: 0.711) with the Epic model (AUROC: 0.697). Given features in the first 48 hours, the random forest model produces higher AUROC (0.684) than the Epic model (AUROC: 0.676). Both models flagged patients with a similar distribution of race and sex; however, our light gradient boosting and random forest models were more inclusive, flagging more patients among younger age groups. The Epic models were more sensitive to identifying patients with an average lower zip income. Our 48-hour models were powered by novel features at various levels: patient (weight change over 365 days, depression symptoms, laboratory values, and cancer type), hospital (winter discharge and hospital admission type), and community (zip income and marital status of partner).

CONCLUSION

We developed and validated models comparable with the existing Epic 30-day readmission models with several novel actionable insights that could create service interventions deployed by the case management or discharge planning teams that may decrease readmission rates over time.

A comprehensive deep learning method for empirical spectral prediction and its quantitative validation of nano-structured dimers

Nanophotonics exploits the best of photonics and nanotechnology which has transformed optics in recent years by allowing subwavelength structures to enhance light-matter interactions. Despite these breakthroughs, design, fabrication, and characterization of such exotic devices have remained through iterative processes which are often computationally costly, memory-intensive, and time-consuming. In contrast, deep learning approaches have recently shown excellent performance as practical computational tools, providing an alternate avenue for speeding up such nanophotonics simulations. This study presents a DNN framework for transmission, reflection, and absorption spectra predictions by grasping the hidden correlation between the independent nanostructure properties and their corresponding optical responses. The proposed DNN framework is shown to require a sufficient amount of training data to achieve an accurate approximation of the optical performance derived from computational models. The fully trained framework can outperform a traditional EM solution using on the COMSOL Multiphysics approach in terms of computational cost by three orders of magnitude. Furthermore, employing deep learning methodologies, the proposed DNN framework makes an effort to optimise design elements that influence the geometrical dimensions of the nanostructure, offering insight into the universal transmission, reflection, and absorption spectra predictions at the nanoscale. This paradigm improves the viability of complicated nanostructure design and analysis, and it has a lot of potential applications involving exotic light-matter interactions between nanostructures and electromagnetic fields. In terms of computational times, the designed algorithm is more than 700 times faster as compared to conventional FEM method (when manual meshing is used). Hence, this approach paves the way for fast yet universal methods for the characterization and analysis of the optical response of nanophotonic systems.

The role of artificial neural network and machine learning in utilizing spatial information

In this age of the fourth industrial revolution 4.0, the digital world has a plethora of data, including the internet of things, mobile, cybersecurity, social media, forecasts, health data, and so on. The expertise of machine learning and artificial intelligence (AI) is required to soundly evaluate the data and develop related smart and automated applications, These fields use a variety of machine learning techniques including supervised, unsupervised, and reinforcement learning. The objective of the study is to present the role of artificial neural networks and machine learning in utilizing spatial information. Machine learning and AI play an increasingly important role in disaster risk reduction from hazard mapping and forecasting severe occurrences to real-time event detection, situational awareness, and decision assistance. Some of the applications employed in the study to analyze the various ANN domains included weather forecasting, medical diagnosis, aerospace, facial recognition, stock market, social media, signature verification, forensics, robotics, electronics hardware, defense, and seismic data gathering. Machine learning determines the many prediction models for problems involving classification, regression, and clustering using known variables and locations from the training dataset, spatial data that is based on tabular data creates different observations that are geographically related to one another for unknown factors and places. The study presents that the Recurrent neural network and convolutional neural network are the best method in spatial information processing, healthcare, and weather forecasting with greater than 90% accuracy.

Application of machine learning-based models to boost the predictive power of the SPAN index

BACKGROUND

This study re-explored the predictive validity of Stroke Prognostication using Age and National Institutes of Health Stroke Scale (SPAN) index in patients who received different treatments for acute ischemic stroke (AIS) and developed machine learning-boosted outcome prediction models.

METHODS

We evaluated the prognostic relevance of SPAN index in patients with AIS who received intravenous tissue-type plasminogen activator (IV-tPA), intra-arterial thrombolysis (IAT) or non-thrombolytic treatments (non-tPA), and applied machine learning algorithms to develop SPAN-based outcome prediction models in a cohort of 2145 hospitalized AIS patients. The performance of the models was assessed and compared using the area under the receiver operating characteristic curves (AUCs).

RESULTS

SPAN index ≥100 was associated with higher mortality rate and higher modified Rankin Scale at discharge in AIS patients who received the different treatments. Compared to the lower AUCs for the SPAN-alone model across all groups, the AUCs of the logistic regression-boosted model were 0.838, 0.857, 0.766 and 0.875 for the whole cohort, non-tPA, IV-tPA and IAT groups, respectively. Similarly, the AUCs of the generated artificial neural network were 0.846, 0.858, 0.785 and 0.859 for the whole cohort, non-tPA, IV-tPA and IAT groups, respectively, while for gradient boosting decision tree model, we computed 0.850, 0.863, 0.779 and 0.815.

CONCLUSIONS

SPAN index has prognostic relevance in patients with AIS who received different treatments. The generated machine learning-based models exhibit good performance for predicting the functional recovery of AIS; thus, their proposed clinical application to aid outcome prediction and decision-making for the patients with AIS.

Artificial Neural Networks Coupled with MALDI-TOF MS Serum Fingerprinting To Classify and Diagnose Pathological Pain Subtypes in Preclinical Models

Pathological pain subtypes can be classified as either neuropathic pain, caused by a somatosensory nervous system lesion or disease, or nociplastic pain, which develops without evidence of somatosensory system damage. Since there is no gold standard for the diagnosis of pathological pain subtypes, the proper classification of individual patients is currently an unmet challenge for clinicians. While the determination of specific biomarkers for each condition by current biochemical techniques is a complex task, the use of multimolecular techniques, such as matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), combined with artificial intelligence allows specific fingerprints for pathological pain-subtypes to be obtained, which may be useful for diagnosis. We analyzed whether the information provided by the mass spectra of serum samples of four experimental models of neuropathic and nociplastic pain combined with their functional pain outcomes could enable pathological pain subtype classification by artificial neural networks. As a result, a simple and innovative clinical decision support method has been developed that combines MALDI-TOF MS serum spectra and pain evaluation with its subsequent data analysis by artificial neural networks and allows the identification and classification of pathological pain subtypes in experimental models with a high level of specificity.

Conceptual Structure and Current Trends in Artificial Intelligence, Machine Learning, and Deep Learning Research in Sports: A Bibliometric Review

Artificial intelligence and its subcategories of machine learning and deep learning are gaining increasing importance and attention in the context of sports research. This has also meant that the number of corresponding publications has become complex and unmanageably large in human terms. In the current state of the research field, there is a lack of bibliometric analysis, which would prove useful for obtaining insights into the large amounts of available literature. Therefore, the present work aims to identify important research issues, elucidate the conceptual structure of the research field, and unpack the evolutionary trends and the direction of hot topics regarding key themes in the research field of artificial intelligence in sports. Using the Scopus database, 1215 documents (reviews and articles) were selected. Bibliometric analysis was performed using VOSviewer and bibliometrix R package. The main findings are as follows: (a) the literature and research interest concerning AI and its subcategories is growing exponentially; (b) the top 20 most cited works comprise 32.52% of the total citations; (c) the top 10 journals are responsible for 28.64% of all published documents; (d) strong collaborative relationships are present, along with small, isolated collaboration networks of individual institutions; (e) the three most productive countries are China, the USA, and Germany; (f) different research themes can be characterized using author keywords with current trend topics, e.g., in the fields of biomechanics, injury prevention or prediction, new algorithms, and learning approaches. AI research activities in the fields of sports pedagogy, sports sociology, and sports economics seem to have played a subordinate role thus far. Overall, the findings of this study expand knowledge on the research situation as well as the development of research topics regarding the use of artificial intelligence in sports, and may guide researchers to identify currently relevant topics and gaps in the research.

Analysis of Unconfined Compressive Strength of Rammed Earth Mixes Based on Artificial Neural Network and Statistical Analysis

Earth materials have been used in construction as safe, healthy and environmentally sustainable. It is often challenging to develop an optimum soil mix because of the significant variations in soil properties from one soil to another. The current study analyzed the soil properties, including the grain size distribution, Atterberg limits, compaction characteristics, etc., using multilinear regression (MLR) and artificial neural networks (ANN). Data collected from previous studies (i.e., 488 cases) for stabilized (with either cement or lime) and unstabilized soils were considered and analyzed. Missing data were estimated by correlations reported in previous studies. Then, different ANNs were designed (trained and validated) using Levenberg-Marquardt (L-M) algorithms. Using the MLR, several models were developed to estimate the compressive strength of both unstabilized and stabilized soils with a Pearson Coefficient of Correlation (R²) equal to 0.2227 and 0.766, respectively. On the other hand, developed ANNs gave a higher value for R² than MLR (with the highest value achieved at 0.9883). Thereafter, an experimental program was carried out to validate the results achieved in this study. Finally, a sensitivity analysis was carried out using the resulting networks to assess the effect of different soil properties on the unconfined compressive strength (UCS). Moreover, suitable recommendations for earth materials mixes were presented.

Development and validation of prediction models for hypertension risks: A cross-sectional study based on 4,287,407 participants

Objective

To develop an optimal screening model to identify the individuals with a high risk of hypertension in China by comparing tree-based machine learning models, such as classification and regression tree, random forest, adaboost with a decision tree, extreme gradient boosting decision tree, and other machine learning models like an artificial neural network, naive Bayes, and traditional logistic regression models.

Methods

A total of 4,287,407 adults participating in the national physical examination were included in the study. Features were selected using the least absolute shrinkage and selection operator regression. The Borderline synthetic minority over-sampling technique was used for data balance. Non-laboratory and semi-laboratory analyses were carried out in combination with the selected features. The tree-based machine learning models, other machine learning models, and traditional logistic regression models were constructed to identify individuals with hypertension, respectively. Top features selected using the best algorithm and the corresponding variable importance score were visualized.

Results

A total of 24 variables were finally included for analyses after the least absolute shrinkage and selection operator regression model. The sample size of hypertensive patients in the training set was expanded from 689,025 to 2,312,160 using the borderline synthetic minority over-sampling technique algorithm. The extreme gradient boosting decision tree algorithm showed the best results (area under the receiver operating characteristic curve of non-laboratory: 0.893 and area under the receiver operating characteristic curve of semi-laboratory: 0.894). This study found that age, systolic blood pressure, waist circumference, diastolic blood pressure, albumin, drinking frequency, electrocardiogram, ethnicity (uyghur, hui, and other), body mass index, sex (female), exercise frequency, diabetes mellitus, and total bilirubin are important factors reflecting hypertension. Besides, some algorithms included in the semi-laboratory analyses showed less improvement in the predictive performance compared to the non-laboratory analyses.

Conclusion

Using multiple methods, a more significant prediction model can be built, which discovers risk factors and provides new insights into the prediction and prevention of hypertension.

Hybrid stacked ensemble combined with genetic algorithms for diabetes prediction

Diabetes is currently one of the most common, dangerous, and costly diseases globally caused by increased blood sugar or a decrease in insulin in the body. Diabetes can have detrimental effects on people’s health if diagnosed late. Today, diabetes has become one of the challenges for health and government officials. Prevention is a priority, and taking care of people’s health without compromising their comfort is an essential need. In this study, the ensemble training methodology based on genetic algorithms was used to diagnose and predict the outcomes of diabetes mellitus accurately. This study uses the experimental data, actual data on Indian diabetics on the University of California website. Current developments in ICT, such as the Internet of Things, machine learning, and data mining, allow us to provide health strategies with more intelligent capabilities to accurately predict the outcomes of the disease in daily life and the hospital and prevent the progression of this disease and its many complications. The results show the high performance of the proposed method in diagnosing the disease, which has reached 98.8%, and 99% accuracy in this study.

Develop a diagnostic tool for dementia using machine learning and non-imaging features

Background

Early identification of Alzheimer’s disease or mild cognitive impairment can help guide direct prevention and supportive treatments, improve outcomes, and reduce medical costs. Existing advanced diagnostic tools are mostly based on neuroimaging and suffer from certain problems in cost, reliability, repeatability, accessibility, ease of use, and clinical integration. To address these problems, we developed, evaluated, and implemented an early diagnostic tool using machine learning and non-imaging factors.

Methods and results

A total of 654 participants aged 65 or older from the Nursing Home in Hangzhou, China were identified. Information collected from these patients includes dementia status and 70 demographic, cognitive, socioeconomic, and clinical features. Logistic regression, support vector machine (SVM), neural network, random forest, extreme gradient boosting (XGBoost), least absolute shrinkage and selection operator (LASSO), and best subset models were trained, tuned, and internally validated using a novel double cross validation algorithm and multiple evaluation metrics. The trained models were also compared and externally validated using a separate dataset with 1,100 participants from four communities in Zhejiang Province, China. The model with the best performance was then identified and implemented online with a friendly user interface. For the nursing dataset, the top three models are the neural network (AUROC = 0.9435), XGBoost (AUROC = 0.9398), and SVM with the polynomial kernel (AUROC = 0.9213). With the community dataset, the best three models are the random forest (AUROC = 0.9259), SVM with linear kernel (AUROC = 0.9282), and SVM with polynomial kernel (AUROC = 0.9213). The F1 scores and area under the precision-recall curve showed that the SVMs, neural network, and random forest were robust on the unbalanced community dataset. Overall the SVM with the polynomial kernel was found to be the best model. The LASSO and best subset models identified 17 features most relevant to dementia prediction, mostly from cognitive test results and socioeconomic characteristics.

Conclusion

Our non-imaging-based diagnostic tool can effectively predict dementia outcomes. The tool can be conveniently incorporated into clinical practice. Its online implementation allows zero barriers to its use, which enhances the disease’s diagnosis, improves the quality of care, and reduces costs.

Diagnostic Strategies for Breast Cancer Detection: From Image Generation to Classification Strategies Using Artificial Intelligence Algorithms

Breast cancer is one the main death causes for women worldwide, as 16% of the diagnosed malignant lesions worldwide are its consequence. In this sense, it is of paramount importance to diagnose these lesions in the earliest stage possible, in order to have the highest chances of survival. While there are several works that present selected topics in this area, none of them present a complete panorama, that is, from the image generation to its interpretation. This work presents a comprehensive state-of-the-art review of the image generation and processing techniques to detect Breast Cancer, where potential candidates for the image generation and processing are presented and discussed. Novel methodologies should consider the adroit integration of artificial intelligence-concepts and the categorical data to generate modern alternatives that can have the accuracy, precision and reliability expected to mitigate the misclassifications.

Feature-level ensemble approach for COVID-19 detection using chest X-ray images

Severe acute respiratory syndrome coronavirus 2 (SARS CoV-2), also known as the coronavirus disease 2019 (COVID-19), has threatened many human beings around the world and capsized economies at unprecedented magnitudes. Therefore, the detection of this disease using chest X-ray modalities has played a pivotal role in producing fast and accurate medical diagnoses, especially in countries that are unable to afford laboratory testing kits. However, identifying and distinguishing COVID-19 from virtually similar thoracic abnormalities utilizing medical images is challenging because it is time-consuming, demanding, and susceptible to human-based errors. Therefore, artificial-intelligence-driven automated diagnoses, which excludes direct human intervention, may potentially be used to achieve consistently accurate performances. In this study, we aimed to (i) obtain a customized dataset composed of a relatively small number of images collected from publicly available datasets; (ii) present the efficient integration of the shallow handcrafted features obtained from local descriptors, radiomics features specialized for medical images, and deep features aggregated from pre-trained deep learning architectures; and (iii) distinguish COVID-19 patients from healthy controls and pneumonia patients using a collection of conventional machine learning classifiers. By conducting extensive experiments, we demonstrated that the feature-based ensemble approach provided the best classification metrics, and this approach explicitly outperformed schemes that used only either local, radiomic, or deep features. In addition, our proposed method achieved state-of-the-art multi-class classification results compared to the baseline reference for the currently available COVID-19 datasets.

Addressing Binary Classification over Class Imbalanced Clinical Datasets Using Computationally Intelligent Techniques

Nowadays, healthcare is the prime need of every human being in the world, and clinical datasets play an important role in developing an intelligent healthcare system for monitoring the health of people. Mostly, the real-world datasets are inherently class imbalanced, clinical datasets also suffer from this imbalance problem, and the imbalanced class distributions pose several issues in the training of classifiers. Consequently, classifiers suffer from low accuracy, precision, recall, and a high degree of misclassification, etc. We performed a brief literature review on the class imbalanced learning scenario. This study carries the empirical performance evaluation of six classifiers, namely Decision Tree, k-Nearest Neighbor, Logistic regression, Artificial Neural Network, Support Vector Machine, and Gaussian Naïve Bayes, over five imbalanced clinical datasets, Breast Cancer Disease, Coronary Heart Disease, Indian Liver Patient, Pima Indians Diabetes Database, and Coronary Kidney Disease, with respect to seven different class balancing techniques, namely Undersampling, Random oversampling, SMOTE, ADASYN, SVM-SMOTE, SMOTEEN, and SMOTETOMEK. In addition to this, the appropriate explanations for the superiority of the classifiers as well as data-balancing techniques are also explored. Furthermore, we discuss the possible recommendations on how to tackle the class imbalanced datasets while training the different supervised machine learning methods. Result analysis demonstrates that SMOTEEN balancing method often performed better over all the other six data-balancing techniques with all six classifiers and for all five clinical datasets. Except for SMOTEEN, all other six balancing techniques almost had equal performance but moderately lesser performance than SMOTEEN.

Applications of Neural Networks in Biomedical Data Analysis

Neural networks for deep-learning applications, also called artificial neural networks, are important tools in science and industry. While their widespread use was limited because of inadequate hardware in the past, their popularity increased dramatically starting in the early 2000s when it became possible to train increasingly large and complex networks. Today, deep learning is widely used in biomedicine from image analysis to diagnostics. This also includes special topics, such as forensics. In this review, we discuss the latest networks and how they work, with a focus on the analysis of biomedical data, particularly biomarkers in bioimage data. We provide a summary on numerous technical aspects, such as activation functions and frameworks. We also present a data analysis of publications about neural networks to provide a quantitative insight into the use of network types and the number of journals per year to determine the usage in different scientific fields.

Hybrid neural networks with virtual flows in in medical risk classifiers

Modern medical risk classification systems focus on traditional risk factors and modeling methods. The available modeling tools do not allow reliable prediction of the of disease severity. In this study we develop prediction model of recurrent myocardial infarction in the rehabilitation period using several health variables generated in virtual flows. Hybrid decision modules with health data flows were used to build prognostic model for the prediction of disease. The vector of input information features consists of two subvectors: the first reflects real flows, the second reflects virtual flows. Complex interrelations among input data are modelled using Neural Network structure. The model classification quality of the intellectual cardiovascular catastrophe prediction system was tested on a sample composed of 230 patients who had acute myocardial infarction. For prediction, three categories of risk factors were identified: traditional factors, factors associated with stressful overloads, and risk factors derived from bio-impedance studies. During the rehabilitation period, the level of molecular products of lipid peroxidation and the antioxidant potential of blood serum were also studied. Experimental studies of various modifications of the proposed classifier model were conducted, consisting of sequential disconnection from the aggregator of solutions of “weak” classifiers at various hierarchical levels. The mathematical model show predictions accuracy of correct prognosis for the risk of myocardial infarction exceeding 0.86. Prediction quality indicators are higher than the known ASCORE cardiovascular catastrophe prediction system, on average, by 14%.

Artificial neural network-boosted Cardiac Arrest Survival Post-Resuscitation In-hospital (CASPRI) score accurately predicts outcome in cardiac arrest patients treated with targeted temperature management

Existing prognostic models to predict the neurological recovery in patients with cardiac arrest receiving targeted temperature management (TTM) either exhibit moderate accuracy or are too complicated for clinical application. This necessitates the development of a simple and generalizable prediction model to inform clinical decision-making for patients receiving TTM. The present study explores the predictive validity of the Cardiac Arrest Survival Post-resuscitation In-hospital (CASPRI) score in cardiac arrest patients receiving TTM, regardless of cardiac event location, and uses artificial neural network (ANN) algorithms to boost the prediction performance. This retrospective observational study evaluated the prognostic relevance of the CASPRI score and applied ANN to develop outcome prediction models in a cohort of 570 patients with cardiac arrest and treated with TTM between 2014 and 2019 in a nationwide multicenter registry in Taiwan. In univariate logistic regression analysis, the CASPRI score was significantly associated with neurological outcome, with the area under the receiver operating characteristics curve (AUC) of 0.811. The generated ANN model, based on 10 items of the CASPRI score, achieved a training AUC of 0.976 and validation AUC of 0.921, with the accuracy, precision, sensitivity, and specificity of 89.2%, 91.6%, 87.6%, and 91.2%, respectively, for the validation set. CASPRI score has prognostic relevance in patients who received TTM after cardiac arrest. The generated ANN-boosted, CASPRI-based model exhibited good performance for predicting TTM neurological outcome, thus, we propose its clinical application to improve outcome prediction, facilitate decision-making, and formulate individualized therapeutic plans for patients receiving TTM.

GGA-MLP: A Greedy Genetic Algorithm to Optimize Weights and Biases in Multilayer Perceptron

The task of designing an Artificial Neural Network (ANN) can be thought of as an optimization problem that involves many parameters whose optimal value needs to be computed in order to improve the classification accuracy of an ANN. Two of the major parameters that need to be determined during the design of an ANN are weights and biases. Various gradient-based optimization algorithms have been proposed by researchers in the past to generate an optimal set of weights and biases. However, due to the tendency of gradient-based algorithms to get trapped in local minima, researchers have started exploring metaheuristic algorithms as an alternative to the conventional techniques. In this paper, we propose the GGA-MLP (Greedy Genetic Algorithm-Multilayer Perceptron) approach, a learning algorithm, to generate an optimal set of weights and biases in multilayer perceptron (MLP) using a greedy genetic algorithm. The proposed approach increases the performance of the traditional genetic algorithm (GA) by using a greedy approach to generate the initial population as well as to perform crossover and mutation. To evaluate the performance of GGA-MLP in classifying nonlinear input patterns, we perform experiments on datasets of varying complexities taken from the University of California, Irvine (UCI) repository. The experimental results of GGA-MLP are compared with the existing state-of-the-art techniques in terms of classification accuracy. The results show that the performance of GGA-MLP is better than or comparable to the existing state-of-the-art techniques.

Deep Learning-Based Multilevel Classification of Alzheimer’s Disease Using Non-invasive Functional Near-Infrared Spectroscopy

The timely diagnosis of Alzheimer’s disease (AD) and its prodromal stages is critically important for the patients, who manifest different neurodegenerative severity and progression risks, to take intervention and early symptomatic treatments before the brain damage is shaped. As one of the promising techniques, functional near-infrared spectroscopy (fNIRS) has been widely employed to support early-stage AD diagnosis. This study aims to validate the capability of fNIRS coupled with Deep Learning (DL) models for AD multi-class classification. First, a comprehensive experimental design, including the resting, cognitive, memory, and verbal tasks was conducted. Second, to precisely evaluate the AD progression, we thoroughly examined the change of hemodynamic responses measured in the prefrontal cortex among four subject groups and among genders. Then, we adopted a set of DL architectures on an extremely imbalanced fNIRS dataset. The results indicated that the statistical difference between subject groups did exist during memory and verbal tasks. This presented the correlation of the level of hemoglobin activation and the degree of AD severity. There was also a gender effect on the hemoglobin changes due to the functional stimulation in our study. Moreover, we demonstrated the potential of distinguished DL models, which boosted the multi-class classification performance. The highest accuracy was achieved by Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) using the original dataset of three hemoglobin types (0.909 ± 0.012 on average). Compared to conventional machine learning algorithms, DL models produced a better classification performance. These findings demonstrated the capability of DL frameworks on the imbalanced class distribution analysis and validated the great potential of fNIRS-based approaches to be further contributed to the development of AD diagnosis systems.

Predicting Characteristics Associated with Breast Cancer Survival Using Multiple Machine Learning Approaches

Breast cancer is one of the most commonly diagnosed female disorders globally. Numerous studies have been conducted to predict survival markers, although the majority of these analyses were conducted using simple statistical techniques. In lieu of that, this research employed machine learning approaches to develop models for identifying and visualizing relevant prognostic indications of breast cancer survival rates. A comprehensive hospital-based breast cancer dataset was collected from the National Cancer Institute's SEER Program's November 2017 update, which offers population-based cancer statistics. The dataset included female patients diagnosed between 2006 and 2010 with infiltrating duct and lobular carcinoma breast cancer (SEER primary cites recode NOS histology codes 8522/3). The dataset included nine predictor factors and one predictor variable that were linked to the patients' survival status (alive or dead). To identify important prognostic markers associated with breast cancer survival rates, prediction models were constructed using K-nearest neighbor (K-NN), decision tree (DT), gradient boosting (GB), random forest (RF), AdaBoost, logistic regression (LR), voting classifier, and support vector machine (SVM). All methods yielded close results in terms of model accuracy and calibration measures, with the lowest achieved from logistic regression (accuracy = 80.57 percent) and the greatest acquired from the random forest (accuracy = 94.64 percent). Notably, the multiple machine learning algorithms utilized in this research achieved high accuracy, suggesting that these approaches might be used as alternative prognostic tools in breast cancer survival studies, especially in the Asian area.

Deep learning time series prediction models in surveillance data of hepatitis incidence in China

Background

Precise incidence prediction of Hepatitis infectious disease is critical for early prevention and better government strategic planning. In this paper, we presented different prediction models using deep learning methods based on the monthly incidence of Hepatitis through a national public health surveillance system in China mainland.

Methods

We assessed and compared the performance of three deep learning methods, namely, Long Short-Term Memory (LSTM) prediction model, Recurrent Neural Network (RNN) prediction model, and Back Propagation Neural Network (BPNN) prediction model. The data collected from 2005 to 2018 were used for the training and prediction model, while the data are split via 5-Fold cross-validation. The performance was evaluated based on three metrics: mean square error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE).

Results

Among the year 2005–2018, 20,924,951 cases and 11,892 deaths were supervised in the system. Hepatitis B (HB) is the most disease-causing incidence and death, and the proportion is greater than 70 percent, while the percentage of the incidence and deaths is decreased much in 2018 compared with 2005. Based on the measured errors and the visualization of the three neural networks, there is no one model predicting the incidence cases that can be completely superior to other models. When predicting the number of incidence cases for HB, the performance ranking of the three models from high to low is LSTM, BPNN, RNN, while it is LSTM, RNN, BPNN for Hepatitis C (HC). while the MAE, MSE and MAPE of the LSTM model for HB, HC are 3.84*10⁻⁰⁶, 3.08*10⁻¹¹, 4.981, 8.84*10⁻⁰⁶, 1.98*10⁻¹²,5.8519, respectively.

Conclusions

The deep learning time series predictive models show their significance to forecast the Hepatitis incidence and have the potential to assist the decision-makers in making efficient decisions for the early detection of the disease incidents, which would significantly promote Hepatitis disease control and management.

The Role of Humans in Surgery Automation

Innovation in healthcare promises unparalleled potential in optimizing the production, distribution, and use of the health workforce and infrastructure, allocating system resources more efficiently, and streamline care pathways and supply chains. A recent innovation contributing to this is robot-assisted surgeries (RAS). RAS causes less damage to the patient's body, less pain and discomfort, shorter hospital stays, quicker recovery times, smaller scars, and less risk of complications. However, introducing a robot in traditional surgeries is not straightforward and brings about new risks that conventional medical instruments did not pose before. For instance, since robots are sophisticated machines capable of acting autonomously, the surgical procedure's outcome is no longer limited to the surgeon but may also extend to the robot manufacturer and the hospital. This article explores the influence of automation on stakeholder responsibility in surgery robotization. To this end, we map how the role of different stakeholders in highly autonomous robotic surgeries is transforming, explore some of the challenges that robot manufacturers and hospital management will increasingly face as surgical procedures become more and more automated, and bring forward potential solutions to ascertain clarity in the role of stakeholders before, during, and after robot-enabled surgeries (i.e. a Robot Impact Assessment (ROBIA), a Robo-Terms framework inspired by the international trade system 'Incoterms', and a standardized adverse event reporting mechanism). In particular, we argue that with progressive robot autonomy, performance, oversight, and support will increasingly be shared between the human surgeon, the support staff, and the robot (and, by extent, the robot manufacturer), blurring the lines of who is responsible if something goes wrong. Understanding the exact role of humans in highly autonomous robotic surgeries is essential to map liability and bring certainty concerning the ascription of responsibility. We conclude that the full benefits the use of robotic innovations and solutions in surgery could bring to healthcare providers and receivers cannot be realized until there is more clarity on the division of responsibilities channeling robot autonomy and human performance, support, and oversight; a transformation on the education and training of medical staff, and betterment on the complex interplay between manufacturers, healthcare providers, and patients.

A methodical survey of mathematical model-based control techniques based on open and closed loop control approach for diabetes management

Disturbance of blood sugar level is controlled through well-known biomechanical feedback loops: high levels of glucose in blood facilitate to release insulin from the pancreas which accelerates the absorption rate of cellular glucose. Low glucose levels encourage to release pancreatic glucagon which induces glycogen breakdown to glucose in the liver. These bio-control systems do not function properly in diabetic patients. Though the control of disease seems intuitively easy, in real life, due to many differences in structure by diet and fasting, exercise, medications, patient’s profile and other stressors, it is not that easy. The mathematical models of the glucose-insulin regulatory system follow the patient’s physiological conditions which make it difficult to identify and estimate all the model parameters. In this paper, we have given a systematic literature review on mathematical models of the diabetic patients, and various kinds of disease control techniques through the development of open and closed loop insulin deliver command system and optimization of exogenous insulin rate. It demonstrates the open and closed loop type model-based control strategies underlying the assumptions of the concerned models. The combination of mathematical model with control strategies such as genetic algorithm (GA), neural network (NN), sliding mode controller (SMC), model predictive controller (MPC), and fuzzy logic control (FLC) has been considered, which provides an overview of this area, highlighting the control profile over the diabetic model with promising clinical results, outlining key challenges, and identifying needs for the future research. Also, the significance of these control algorithms has been discussed in the presence of the noises, the controller’s robustness and various other disturbances. It provides substantial information on diabetes management through various control techniques.

SuPART: supervised projective adapted resonance theory for automatic quality assurance approval of radiotherapy treatment plans

Radiotherapy is a common treatment modality for the treatment of cancer, where treatments must be carefully designed to deliver appropriate dose to targets while avoiding healthy organs. The comprehensive multi-disciplinary quality assurance (QA) process in radiotherapy is designed to ensure safe and effective treatment plans are delivered to patients. However, the plan QA process is expensive, often time-intensive, and requires review of large quantities of complex data, potentially leading to human error in QA assessment. We therefore develop an automated machine learning algorithm to identify ‘acceptable’ plans (plans that are similar to historically approved plans) and ‘unacceptable’ plans (plans that are dissimilar to historically approved plans). This algorithm is a supervised extension of projective adaptive resonance theory, called SuPART, that learns a set of distinctive features, and considers deviations from them indications of unacceptable plans. We test SuPART on breast and prostate radiotherapy datasets from our institution, and find that SuPART outperforms common classification algorithms in several measures of accuracy. When no falsely approved plans are allowed, SuPART can correctly auto-approve 34% of the acceptable breast and 32% of the acceptable prostate plans, and can also correctly reject 53% of the unacceptable breast and 56% of the unacceptable prostate plans. Thus, usage of SuPART to aid in QA could potentially yield significant time savings.

Survey of BERT-Base Models for Scientific Text Classification: COVID-19 Case Study

On 30 January 2020, the World Health Organization announced a new coronavirus, which later turned out to be very dangerous. Since that date, COVID-19 has spread to become a pandemic that has now affected practically all regions in the world. Since then, many researchers in medicine have contributed to fighting COVID-19. In this context and given the great growth of scientific publications related to this global pandemic, manual text and data retrieval has become a challenging task. To remedy this challenge, we are proposing CovBERT, a pre-trained language model based on the BERT model to automate the literature review process. CovBERT relies on prior training on a large corpus of scientific publications in the biomedical domain and related to COVID-19 to increase its performance on the literature review task. We evaluate CovBERT on the classification of short text based on our scientific dataset of biomedical articles on COVID-19 entitled COV-Dat-20. We demonstrate statistically significant improvements by using BERT.

Robust Estimation of the Chronological Age of Children and Adolescents Using Tooth Geometry Indicators and POD-GP

Determining the chronological age of children or adolescents is becoming an extremely necessary and important issue. Correct age-assessment methods are especially important in the process of international adoption and in the case of immigrants without valid documents confirming their identity. It is well known that traditional, analog methods widely used in clinical evaluation are burdened with a high error rate and are characterized by low accuracy. On the other hand, new digital approaches appear in medicine more and more often, which allow the increase of the accuracy of these estimates, and thus equip doctors with a tool for reliable estimation of the chronological age of children and adolescents. In this study, the work on a fast and effective metamodel is continued. Metamodels have one great advantage over all other analog and quasidigital methods—if they are well trained, a priori, on a representative set of samples, then in the age-assessment phase, results are obtained in a fraction of a second and with little error (reduced to ±7.5 months). In the here-proposed method, the standard deviation for each estimate is additionally obtained, which allows the assessment of the certainty of each result. In this study, 619 pantomographic photos of 619 patients (296 girls and 323 boys) of different ages were used. In the numerical procedure, on the other hand, a metamodel based on the Proper Orthogonal Decomposition (POD) and Gaussian processes (GP) were utilized. The accuracy of the trained model was up to 95%.