Prediction of hematoma expansion in spontaneous intracerebral hemorrhage using support vector machine

Quantitative imaging for predicting hematoma expansion in intracerebral hemorrhage: A multimodel comparison

BACKGROUND

Several studies report that radiomics provides additional information for predicting hematoma expansion in intracerebral hemorrhage (ICH). However, the comparison of diagnostic performance of radiomics for predicting revised hematoma expansion (RHE) remains unclear.

METHODS

The cohort comprised 312 consecutive patients with ICH. A total of 1106 radiomics features from seven categories were extracted using Python software. Support vector machines achieved the best performance in both the training and validation datasets. Clinical factors models were constructed to predict RHE. Receiver operating characteristic curve analysis was used to assess the abilities of non-contrast computed tomography (NCCT) signs, radiomics features, and combined models to predict RHE.

RESULTS

We finally selected the top 21 features for predicting RHE. After univariate analysis, 4 clinical factors and 5 NCCT signs were selected for inclusion in the prediction models. In the training and validation dataset, radiomics features had a higher predictive value for RHE (AUC = 0.83) than a single NCCT sign and expansion-prone hematoma. The combined prediction model including radiomics features, clinical factors, and NCCT signs achieved higher predictive performances for RHE (AUC = 0.88) than other combined models.

CONCLUSIONS

NCCT radiomics features have a good degree of discrimination for predicting RHE in ICH patients. Combined prediction models that include quantitative imaging significantly improve the prediction of RHE, which may assist in the risk stratification of ICH patients for anti-expansion treatments.

HE-Mind: A model for automatically predicting hematoma expansion after spontaneous intracerebral hemorrhage

PURPOSE

To develop and validate an end-to-end model for automatically predicting hematoma expansion (HE) after spontaneous intracerebral hemorrhage (sICH) using a novel deep learning framework.

METHODS

This multicenter retrospective study collected cranial noncontrast computed tomography (NCCT) images of 490 patients with sICH at admission for model training (n = 236), internal testing (n = 60), and external testing (n = 194). A HE-Mind model was designed to predict HE, which consists of a densely connected U-net for segmentation process, a multi-instance learning strategy for resolving label ambiguity and a Siamese network for classification process. Two radiomics models based on support vector machine or logistic regression and two deep learning models based on residual network or Swin transformer were developed for performance comparison. Reader experiments including physician diagnosis mode and artificial intelligence mode were conducted for efficiency comparison.

RESULTS

The HE-Mind model showed better performance compared to the comparative models in predicting HE, with areas under the curve of 0.849 and 0.809 in the internal and external test sets respectively. With the assistance of the HE-Mind model, the predictive accuracy and work efficiency of the emergency physician, junior radiologist, and senior radiologist were significantly improved, with accuracies of 0.768, 0.789, and 0.809 respectively, and reporting times of 7.26 s, 5.08 s, and 3.99 s respectively.

CONCLUSIONS

The HE-Mind model could rapidly and automatically process the NCCT data and predict HE after sICH within three seconds, indicating its potential to assist physicians in the clinical diagnosis workflow of HE.

Hematoma expansion prediction based on SMOTE and XGBoost algorithm

Hematoma expansion (HE) is a high risky symptom with high rate of occurrence for patients who have undergone spontaneous intracerebral hemorrhage (ICH) after a major accident or illness. Correct prediction of the occurrence of HE in advance is critical to help the doctors to determine the next step medical treatment. Most existing studies focus only on the occurrence of HE within 6 h after the occurrence of ICH, while in reality a considerable number of patients have HE after the first 6 h but within 24 h. In this study, based on the medical doctors recommendation, we focus on prediction of the occurrence of HE within 24 h, as well as the occurrence of HE every 6 h within 24 h. Based on the demographics and computer tomography (CT) image extraction information, we used the XGBoost method to predict the occurrence of HE within 24 h. In this study, to solve the issue of highly imbalanced data set, which is a frequent case in medical data analysis, we used the SMOTE algorithm for data augmentation. To evaluate our method, we used a data set consisting of 582 patients records, and compared the results of proposed method as well as few machine learning methods. Our experiments show that XGBoost achieved the best prediction performance on the balanced dataset processed by the SMOTE algorithm with an accuracy of 0.82 and F1-score of 0.82. Moreover, our proposed method predicts the occurrence of HE within 6, 12, 18 and 24 h at the accuracy of 0.89, 0.82, 0.87 and 0.94, indicating that the HE occurrence within 24 h can be predicted accurately by the proposed method.

Predicting hematoma expansion using machine learning: An exploratory analysis of the ATACH 2 trial

INTRODUCTION

Hematoma expansion (HE) in patients with intracerebral hemorrhage (ICH) is a key predictor of poor prognosis and potentially amenable to treatment. This study aimed to build a classification model to predict HE in patients with ICH using deep learning algorithms without using advanced radiological features.

METHODS

Data from the ATACH-2 trial (Antihypertensive Treatment of Acute Cerebral Hemorrhage) was utilized. Variables included in the models were chosen as per literature consensus on salient variables associated with HE. HE was defined as increase in either >33% or 6 mL in hematoma volume in the first 24 h. Multiple machine learning algorithms were employed using iterative feature selection and outcome balancing methods. 70% of patients were used for training and 30% for internal validation. We compared the ML models to a logistic regression model and calculated AUC, accuracy, sensitivity and specificity for the internal validation models respective models.

RESULTS

Among 1000 patients included in the ATACH-2 trial, 924 had the complete parameters which were included in the analytical cohort. The median [interquartile range (IQR)] initial hematoma volume was 9.93.mm3 [5.03-18.17] and 25.2% had HE. The best performing model across all feature selection groups and sampling cohorts was using an artificial neural network (ANN) for HE in the testing cohort with AUC 0.702 [95% CI, 0.631-0.774] with 8 hidden layer nodes The traditional logistic regression yielded AUC 0.658 [95% CI, 0.641-0.675]. All other models performed with less accuracy and lower AUC. Initial hematoma volume, time to initial CT head, and initial SBP emerged as most relevant variables across all best performing models.

CONCLUSION

We developed multiple ML algorithms to predict HE with the ANN classifying the best without advanced radiographic features, although the AUC was only modestly better than other models. A larger, more heterogenous dataset is needed to further build and better generalize the models.

Machine learning for predicting hematoma expansion in spontaneous intracerebral hemorrhage: a systematic review and meta-analysis

PURPOSE

Early identification of hematoma enlargement and persistent hematoma expansion (HE) in patients with cerebral hemorrhage is increasingly crucial for determining clinical treatments. However, due to the lack of clinically effective tools, radiomics has been gradually introduced into the early identification of hematoma enlargement. Though, radiomics has limited predictive accuracy due to variations in procedures. Therefore, we conducted a systematic review and meta-analysis to explore the value of radiomics in the early detection of HE in patients with cerebral hemorrhage.

METHODS

Eligible studies were systematically searched in PubMed, Embase, Cochrane and Web of Science from inception to April 8, 2024. English articles are considered eligible. The radiomics quality scoring (RQS) tool was used to evaluate included studies.

RESULTS

A total of 34 studies were identified with sample sizes ranging from 108 to 3016. Eleven types of models were involved, and the types of modeling contained mainly clinical, radiomic, and radiomic plus clinical features. The radiomics models seem to have better performance (0.77 and 0.73 C-index in the training cohort and validation cohort, respectively) than the clinical models (0.69 C-index in the training cohort and 0.70 C-index in the validation cohort) in discriminating HE. However, the C-index was the highest for the combined model in both the training (0.82) and validation (0.79) cohorts.

CONCLUSIONS

Machine learning based on radiomic plus clinical features has the best predictive performance for HE, followed by machine learning based on radiomic features, and can be used as a potential tool to assist clinicians in early judgment.

Research advances in predicting the expansion of hypertensive intracerebral hemorrhage based on CT images: an overview

Hematoma expansion (HE) is an important risk factor for death or poor prognosis in patients with hypertensive intracerebral hemorrhage (HICH). Accurately predicting the risk of HE in patients with HICH is of great clinical significance for timely intervention and improving patient prognosis. Many imaging signs reported in literatures showed the important clinical value for predicting HE. In recent years, the development of radiomics and artificial intelligence has provided new methods for HE prediction with high accuracy. Therefore, this article reviews the latest research progress in CT imaging, radiomics, and artificial intelligence of HE, in order to help identify high-risk patients for HE in clinical practice.

Artificial Intelligence and Machine Learning Applications in Critically Ill Brain Injured Patients

The utilization of Artificial Intelligence (AI) and Machine Learning (ML) is paving the way for significant strides in patient diagnosis, treatment, and prognostication in neurocritical care. These technologies offer the potential to unravel complex patterns within vast datasets ranging from vast clinical data and EEG (electroencephalogram) readings to advanced cerebral imaging facilitating a more nuanced understanding of patient conditions. Despite their promise, the implementation of AI and ML faces substantial hurdles. Historical biases within training data, the challenge of interpreting multifaceted data streams, and the "black box" nature of ML algorithms present barriers to widespread clinical adoption. Moreover, ethical considerations around data privacy and the need for transparent, explainable models remain paramount to ensure trust and efficacy in clinical decision-making.This article reflects on the emergence of AI and ML as integral tools in neurocritical care, discussing their roles from the perspective of both their scientific promise and the associated challenges. We underscore the importance of extensive validation in diverse clinical settings to ensure the generalizability of ML models, particularly considering their potential to inform critical medical decisions such as withdrawal of life-sustaining therapies. Advancement in computational capabilities is essential for implementing ML in clinical settings, allowing for real-time analysis and decision support at the point of care. As AI and ML are poised to become commonplace in clinical practice, it is incumbent upon health care professionals to understand and oversee these technologies, ensuring they adhere to the highest safety standards and contribute to the realization of personalized medicine. This engagement will be pivotal in integrating AI and ML into patient care, optimizing outcomes in neurocritical care through informed and data-driven decision-making.

The significance of admission blood lactate and fibrinogen in pediatric traumatic brain injury: a single-center clinical study

BACKGROUND

Traumatic brain injury (TBI) is a significant cause of morbidity and mortality in pediatric patients, leading to long-term physical, cognitive, and psychological impairments. Blood lactate and fibrinogen levels have emerged as potential biomarkers associated with tissue hypoperfusion and coagulation dysfunction, respectively. However, limited research has specifically focused on the significance of these biomarkers in pediatric TBI. This study aimed to investigate the clinical significance of blood lactate and fibrinogen levels upon admission in pediatric patients with traumatic brain injury.

METHODS

The medical records of 80 children with a traumatic brain injury who were admitted from January 2017 to January 2021 were retrospectively analyzed. The two groups were compared according to whether the blood lactate in the admission arterial blood gas increased and the fibrinogen content in the coagulation function decreased. The clinical data of the children in the two groups were different, and then they were divided into a good prognosis group and a poor prognosis group according to the GOS prognostic score, and the differences in the clinical indicators of the two groups were compared.

RESULTS

Among the 80 patients, 33 had elevated blood lactate levels, 34 had decreased fibrinogen levels, and 29 had an unfavorable outcome (GOS < 4). Compared to the normal blood lactate group, there were no statistically significant differences in age, sex ratio, or platelet count in the elevated blood lactate group (P > 0.05). However, the elevated blood lactate group had lower Glasgow Coma Scale (GCS) scores upon admission, higher blood lactate levels, lower fibrinogen levels, longer hospital stay, lower GOS scores, and a higher proportion of GOS < 4 (P < 0.05). Compared to the normal fibrinogen group, there were no statistically significant differences in age, sex ratio, or platelet count in the decreased fibrinogen group (P > 0.05). However, the decreased fibrinogen group had lower GCS scores upon admission, higher blood lactate levels, lower fibrinogen levels, longer hospital stays, lower GOS scores, and a higher proportion of GOS < 4 (P < 0.05). Compared to the favorable outcome group, there were no statistically significant differences in age, sex ratio, or platelet count in the unfavorable outcome group (P > 0.05). However, the unfavorable outcome group had lower GCS scores upon admission, higher blood lactate levels, lower fibrinogen levels, longer hospital stays, a higher incidence of pulmonary infection, a higher incidence of stress ulcers, and lower GOS scores (P < 0.05).

CONCLUSION

The levels of blood lactate and fibrinogen may represent the severity of children with traumatic brain injury and may be risk factors for poor prognosis of children with traumatic brain injury.

Predicting hematoma expansion in acute spontaneous intracerebral hemorrhage: integrating clinical factors with a multitask deep learning model for non-contrast head CT

PURPOSE

To predict hematoma growth in intracerebral hemorrhage patients by combining clinical findings with non-contrast CT imaging features analyzed through deep learning.

METHODS

Three models were developed to predict hematoma expansion (HE) in 572 patients. We utilized multi-task learning for both hematoma segmentation and prediction of expansion: the Image-to-HE model processed hematoma slices, extracting features and computing a normalized DL score for HE prediction. The Clinical-to-HE model utilized multivariate logistic regression on clinical variables. The Integrated-to-HE model combined image-derived and clinical data. Significant clinical variables were selected using forward selection in logistic regression. The two models incorporating clinical variables were statistically validated.

RESULTS

For hematoma detection, the diagnostic performance of the developed multi-task model was excellent (AUC, 0.99). For expansion prediction, three models were evaluated for predicting HE. The Image-to-HE model achieved an accuracy of 67.3%, sensitivity of 81.0%, specificity of 64.0%, and an AUC of 0.76. The Clinical-to-HE model registered an accuracy of 74.8%, sensitivity of 81.0%, specificity of 73.3%, and an AUC of 0.81. The Integrated-to-HE model, merging both image and clinical data, excelled with an accuracy of 81.3%, sensitivity of 76.2%, specificity of 82.6%, and an AUC of 0.83. The Integrated-to-HE model, aligning closest to the diagonal line and indicating the highest level of calibration, showcases superior performance in predicting HE outcomes among the three models.

CONCLUSION

The integration of clinical findings with non-contrast CT imaging features analyzed through deep learning showed the potential for improving the prediction of HE in acute spontaneous intracerebral hemorrhage patients.

Independent Factors Associated with 30-Day In-Hospital Mortality from Acute Spontaneous Intracerebral Hemorrhage

OBJECTIVE

This study aims to investigate independent factors associated with 30-day mortality in patients with acute spontaneous intracerebral hemorrhage (SICH) before treatment.

METHODS

A retrospective analysis was performed on medical records of patients hospitalized with acute SICH between 2019 and 2021. Data included personal history, hospital stay duration, symptom onset, chief complaint, underlying diseases, medication, and alcohol/smoking habits. Physical examination records comprised baseline blood pressure, Glasgow Coma Scale assessment, and pupil reaction evaluation. Diagnostic imaging, specifically computed tomography brain scans, was examined for hemorrhage details. Multivariable logistic analysis was utilized for data analysis.

RESULTS

Among 663 cases, 185 (27.9%) experienced mortality. Risk factors for mortality included chronic kidney disease, ischemic heart disease, loss of follow-up in hypertension clinic, and pontine hemorrhage. Conversely, motor response (m), reactive pupils, and basal cistern persistence significantly decreased the risk of mortality in multivariable analysis. Receiver operating characteristic analysis identified a m score of 5 as the cutoff for predicting survival.

CONCLUSIONS

Chronic kidney disease, ischemic heart disease, loss of hypertension follow-up, m, reactive pupils, pontine hemorrhage, and basal cistern persistence were independent variables associated with the 30-day mortality rate in SICH patients before treatment initiation. A m, pupil reaction, and basal cistern persistence serve as predictive tools for assessing mortality in SICH before treatment.

Activated Prothrombin Complex Concentrates for the Treatment of Factor Xa Inhibitor-Associated Spontaneous Intracerebral Hemorrhage

Background: Anticoagulant-associated intracerebral hemorrhage (ICH) is a significant cause of morbidity and mortality. Despite approval of a specific reversal agent for factor Xa inhibitors, there is still much interest in nonspecific reversal agents, such as activated prothrombin complex concentrates (aPCCs). Objective: The objective of this study was to describe ICH expansion in a cohort of patients with factor Xa inhibitor-associated ICH who were treated with aPCC. Methods: This was a retrospective cohort study conducted at an academic medical center designated as a comprehensive stroke center. Consecutive patients admitted for ICH who reported use of apixaban or rivaroxaban prior to admission were considered for inclusion in the study. Patients were treated with 25 to 50 units/kg of aPCC. Intracerebral hemorrhage volume was measured before administration of aPCC and then again within 36 hours of aPCC administration. Results: A total of 40 patients were included in the final analysis. Overall, the cohort was predominantly male (24 [60%]), white (27 [67.5%]), and the mean age was 75.3 ± 10.5 years. Most patients reported taking apixaban prior to admission (31 [77.5%]) and a large proportion were also taking aspirin (13 [32.5%]). The mean change in ICH volume was 1.12 ± 6.03 mL (P = 0.2475). Conclusions and Relevance: There was a nonsignificant change in mean ICH volume and no reported cases of thromboembolism. Due to the relatively high proportion of patients with significant hematoma expansion, more studies are needed on which patient population would best benefit from treatment with aPCC.

Development of Cloud-Based Telemedicine Platform for Acute Intracerebral Hemorrhage in Gangwon-do : Concept and Protocol

We aimed to develop a cloud-based telemedicine platform for patients with intracerebral hemorrhage (ICH) at local hospitals in rural and underserved areas in Gangwon-do using artificial intelligence and non-face-to-face collaboration treatment technology. This is a prospective and multi-center development project in which neurosurgeons from four university hospitals in Gangwondo will participate. Information technology experts will verify and improve the performance of the cloud-based telemedicine collaboration platform while treating ICH patients in the actual medical field. Problems identified will be resolved, and the function, performance, security, and safety of the telemedicine platform will be checked through an accredited certification authority. The project will be carried out over 4 years and consists of two phases. The first phase will be from April 2022 to December 2023, and the second phase will be from April 2024 to December 2025. The platform will be developed by dividing the work of the neurosurgeons and information technology experts by setting the order of items through mutual feedback. This article provides information on a project to develop a cloud-based telemedicine platform for acute ICH patients in Gangwon-do.

Using Radiomics and Convolutional Neural Networks for the Prediction of Hematoma Expansion After Intracerebral Hemorrhage

Background

Hematoma enlargement (HE) is a common complication following acute intracerebral hemorrhage (ICH) and is associated with early deterioration and unfavorable clinical outcomes. This study aimed to evaluate the predictive performance of a computed tomography (CT) based model that utilizes deep learning features in identifying HE.

Methods

A total of 408 patients were retrospectively enrolled between January 2015 and December 2020 from our institution. We designed an automatic model that could mask the hematoma area and fusion features of radiomics, clinical data, and convolutional neural network (CNN) in a hybrid model. We assessed the model's performance by using confusion matrix metrics (CM), the area under the receiver operating characteristics curve (AUC), and other statistical indicators.

Results

After automated masking, 408 patients were randomly divided into two cohorts with 204 patients in the training set and 204 patients in the validation set. The first cohort trained the CNN model, from which we then extracted radiomics, clinical data, and CNN features for the second validation cohort. After feature selection by K-highest score, a support vector machines (SVM) model classification was used to predict HE. Our hybrid model exhibited a high AUC of 0.949, and 0.95 of precision, 0.83 of recall, and 0.94 of average precision (AP). The CM found that only 5 cases were misidentified by the model.

Conclusion

The automatic hybrid model we developed is an end-to-end method and can assist in clinical decision-making, thereby facilitating personalized treatment for patients with ICH.

Non-contrast CT radiomics and machine learning for outcomes prediction of patients with acute ischemic stroke receiving conventional treatment

PURPOSE

Accurate prediction of outcomes for patients with acute ischemic stroke (AIS) is crucial for clinical decision-making. In this study, we developed prediction models based on non-contrast computed tomography (NCCT) radiomics and clinical features to predict the modified Rankin Scale (mRS) six months after hospital discharge.

METHOD

A two-center retrospective cohort of 240 AIS patients receiving conventional treatment was included. Radiomics features of the infarct area were extracted from baseline NCCT scans. We applied Kruskal-Wallis (KW) test and recursive feature elimination (RFE) to select features for developing clinical, radiomics, and fusion models (with clinical data and radiomics features), using support vector machine (SVM) algorithm. The prediction performance of the models was assessed by accuracy, sensitivity, specificity, F1 score, and receiver operating characteristic (ROC) curve. Shapley Additive exPlanations (SHAP) was applied to analyze the interpretability and predictor importance of the model.

RESULTS

A total of 1454 texture features were extracted from the NCCT images. In the test cohort, the ROC analysis showed that the radiomics model and the fusion model showed AUCs of 0.705 and 0.857, which outperformed the clinical model (0.643), with the fusion model exhibiting the best performance. Additionally, the accuracy and sensitivity of the fusion model were also the best among the models (84.8% and 93.8%, respectively).

CONCLUSIONS

The model based on NCCT radiomics and machine learning has high predictive efficiency for the prognosis of AIS patients receiving conventional treatment, which can be used to assist early personalized clinical therapy.

Machine learning prediction of motor function in chronic stroke patients: a systematic review and meta-analysis

Background

Recent studies have reported that machine learning (ML), with a relatively strong capacity for processing non-linear data and adaptive ability, could improve the accuracy and efficiency of prediction. The article summarizes the published studies on ML models that predict motor function 3-6 months post-stroke.

Methods

A systematic literature search was conducted in PubMed, Embase, Cochorane and Web of Science as of April 3, 2023 for studies on ML prediction of motor function in stroke patients. The quality of the literature was assessed using the Prediction model Risk Of Bias Assessment Tool (PROBAST). A random-effects model was preferred for meta-analysis using R4.2.0 because of the different variables and parameters.

Results

A total of 44 studies were included in this meta-analysis, involving 72,368 patients and 136 models. Models were categorized into subgroups according to the predicted outcome Modified Rankin Scale cut-off value and whether they were constructed based on radiomics. C-statistics, sensitivity, and specificity were calculated. The random-effects model showed that the C-statistics of all models were 0.81 (95% CI: 0.79; 0.83) in the training set and 0.82 (95% CI: 0.80; 0.85) in the validation set. According to different Modified Rankin Scale cut-off values, C-statistics of ML models predicting Modified Rankin Scale>2(used most widely) in stroke patients were 0.81 (95% CI: 0.78; 0.84) in the training set, and 0.84 (95% CI: 0.81; 0.87) in the validation set. C-statistics of radiomics-based ML models in the training set and validation set were 0.81 (95% CI: 0.78; 0.84) and 0.87 (95% CI: 0.83; 0.90), respectively.

Conclusion

ML can be used as an assessment tool for predicting the motor function in patients with 3-6 months of post-stroke. Additionally, the study found that ML models with radiomics as a predictive variable were also demonstrated to have good predictive capabilities. This systematic review provides valuable guidance for the future optimization of ML prediction systems that predict poor motor outcomes in stroke patients.

Systematic review registration

https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022335260, identifier: CRD42022335260.

Noncontrast Computed Tomography Markers Associated with Hematoma Expansion: Analysis of a Multicenter Retrospective Study

BACKGROUND

Hematoma expansion (HE) is a significant predictor of poor outcomes in patients with intracerebral hemorrhage (ICH). Non-contrast computed tomography (NCCT) markers in ICH are promising predictors of HE. We aimed to determine the association of the NCCT markers with HE by using different temporal HE definitions.

METHODS

We utilized Risa-MIS-ICH trial data (risk stratification and minimally invasive surgery in acute intracerebral hemorrhage). We defined four HE types based on the time to baseline CT (BCT) and the time to follow-up CT (FCT). Hematoma volume was measured by software with a semi-automatic edge detection tool. HE was defined as a follow-up CT hematoma volume increase of >6 mL or a 33% hematoma volume increase relative to the baseline CT. Multivariable regression analyses were used to determine the HE parameters. The prediction potential of indicators for HE was evaluated using receiver-operating characteristic analysis.

RESULTS

The study enrolled 158 patients in total. The time to baseline CT was independently associated with HE in one type (odds ratio (OR) 0.234, 95% confidence interval (CI) 0.077-0.712, p = 0.011), and the blend sign was independently associated with HE in two types (OR, 6.203-6.985, both p < 0.05). Heterogeneous density was independently associated with HE in all types (OR, 6.465-88.445, all p < 0.05) and was the optimal type for prediction, with an area under the curve of 0.674 (p = 0.004), a sensitivity of 38.9%, and specificity of 96.0%.

CONCLUSION

In specific subtypes, the time to baseline CT, blend sign, and heterogeneous density were independently associated with HE. The association between NCCT markers and HE is influenced by the temporal definition of HE. Heterogeneous density is a stable and robust predictor of HE in different subtypes of hematoma expansion.

Prediction of poor outcome in stroke patients using radiomics analysis of intraparenchymal and intraventricular hemorrhage and clinical factors

PURPOSE

To build three prognostic models using radiomics analysis of the hemorrhagic lesions, clinical variables, and their combination, to predict the outcome of stroke patients with spontaneous intracerebral hemorrhage (sICH).

MATERIALS AND METHODS

Eighty-three sICH patients were included. Among them, 40 patients (48.2%) had poor prognosis with modified Rankin scale (mRS) of 5 and 6 at discharge, and the prognostic model was built to differentiate mRS ≤ 4 vs. 5 + 6. The region of interest (ROI) of intraparenchymal hemorrhage (IPH) and intraventricular hemorrhage (IVH) were separately segmented. Features were extracted using PyRadiomics, and the support vector machine was applied to select features and build radiomics models based on IPH and IPH + IVH. The clinical models were built using multivariate logistic regression, and then the radiomics scores were combined with clinical variables to build the combined model.

RESULTS

When using IPH, the AUC for radiomics, clinical, and combined model was 0.78, 0.82, and 0.87, respectively. When using IPH + IVH, the AUC was increased to 0.80, 0.84, and 0.90, respectively. The combined model had a significantly improved AUC compared to the radiomics by DeLong test. A clinical prognostic model based on the ICH score of 0-1 only achieved AUC of 0.71.

CONCLUSIONS

The combined model using the radiomics score derived from IPH + IVH and the clinical factors could achieve a high accuracy in prediction of sICH patients with poor outcome, which may be used to assist in making the decision about the optimal care.

Intracerebral haemorrhage expansion: definitions, predictors, and prevention

Haematoma expansion affects a fifth of patients within 24 h of the onset of acute intracerebral haemorrhage and is associated with death and disability, which makes it an appealing therapeutic target. The time in which active intervention can be done is short as expansion occurs mostly within the first 3 h after onset. Baseline haemorrhage volume, antithrombotic treatment, and CT angiography spot signs are each associated with increased risk of haematoma expansion. Non-contrast CT features are promising predictors of haematoma expansion, but their potential contribution to current models is under investigation. Blood pressure lowering and haemostatic treatment minimise haematoma expansion but have not led to improved functional outcomes in randomised clinical trials. Ultra-early enrolment and selection of participants on the basis of non-contrast CT imaging markers could focus future clinical trials to show clinical benefit in people at high risk of expansion or investigate heterogeneity of treatment effects in clinical trials with broad inclusion criteria.

The Added Value of Intraventricular Hemorrhage on the Radiomics Analysis for the Prediction of Hematoma Expansion of Spontaneous Intracerebral Hemorrhage

Background: Among patients undergoing head computed tomography (CT) scans within 3 h of spontaneous intracerebral hemorrhage (sICH), 28% to 38% have hematoma expansion (HE) on follow-up CT. This study aimed to predict HE using radiomics analysis and investigate the impact of intraventricular hemorrhage (IVH) compared with the conventional approach based on intraparenchymal hemorrhage (IPH) alone. Methods: This retrospective study enrolled 127 patients with baseline and follow-up non-contrast CT (NCCT) within 4~72 h of sICH. IPH and IVH were outlined separately for performing radiomics analysis. HE was defined as an absolute hematoma growth > 6 mL or percentage growth > 33% of either IPH (HEP) or a combination of IPH and IVH (HEP+V) at follow-up. Radiomic features were extracted using PyRadiomics, and then the support vector machine (SVM) was used to build the classification model. For each case, a radiomics score was generated to indicate the probability of HE. Results: There were 57 (44.9%) HEP and 70 (55.1%) non-HEP based on IPH alone, and 58 (45.7%) HEP+V and 69 (54.3%) non-HEP+V based on IPH + IVH. The majority (>94%) of HE patients had poor early outcomes (death or modified Rankin Scale > 3 at discharge). The radiomics model built using baseline IPH to predict HEP (RMP) showed 76.4% accuracy and 0.73 area under the ROC curve (AUC). The other model using IPH + IVH to predict HEP+V (RMP+V) had higher accuracy (81.9%) with AUC = 0.80, and this model could predict poor outcomes. The sensitivity/specificity of RMP and RMP+V for HE prediction were 71.9%/80.0% and 79.3%/84.1%, respectively. Conclusion: The proposed radiomics approach with additional IVH information can improve the accuracy in prediction of HE, which is associated with poor clinical outcomes. A reliable radiomics model may provide a robust tool to help manage ICH patients and to enroll high-risk ICH cases into anti-expansion or neuroprotection drug trials.

Transfer Learning-Based Classification Comparison of Stroke

One type of brain disease that significantly harms people's lives and health is stroke. The diagnosis and management of strokes both heavily rely on the quantitative analysis of brain Magnetic Resonance (MR) images. The early diagnosis process is of great importance for the prevention of stroke cases. Stroke prediction is made possible by deep neural networks with the capacity for enormous data learning. Therefore, in thus study, several deep neural network models, including DenseNet121, ResNet50, Xception, MobileNet, VGG16, and EfficientNetB2 are proposed for transfer learning to classify MR images into two categories (stroke and non-stroke) in order to study the characteristics of the stroke lesions and achieve full intelligent automatic detection. The study dataset comprises of 1901 training images, 475 validation images, and 250 testing images. On the training and validation sets, data augmentation was used to increase the number of images to improve the models’ learning. The experimental results outperform all the state of arts that were used the same dataset. The overall accuracy of the best model is 98.8% and the same value for precision, recall, and f1-score using the EfficientNetB2 model for transfer learning.

Deep Learning-Based Prediction of Hematoma Expansion Using a Single Brain Computed Tomographic Slice in Patients With Spontaneous Intracerebral Hemorrhages

OBJECTIVES

We aimed to predict hematoma expansion in intracerebral hemorrhage (ICH) patients by using the deep learning technique.

METHODS

We retrospectively collected data from ICH patients treated between May 2015 and May 2019. Head computed tomography (CT) scans were performed at admission, and 6 hours, 24 hours, and 72 hours after admission. CT scans were mandatory when neurologic deficits occurred. Univariate and multivariate analyses were conducted to illustrate the association between clinical variables and hematoma expansion. Convolutional neural network (CNN) was adopted to predict hematoma expansion based on brain CT slices. In addition, 5 machine learning methods, including support vector machine, multi-layer perceptron, naive Bayes, decision tree, and random forest, were also performed to predict hematoma expansion based on clinical variables for comparisons.

RESULTS

A total of 223 patients were included. It was revealed that patients' older age (odds ratio [95% confidence interval]: 1.783 [1.417-1.924]), cerebral hemorrhage and breaking into the ventricle (2.524 [1.291-1.778]), coagulopathy (2.341 [1.677-3.454]), and baseline National Institutes of Health Stroke Scale (1.545 [1.132-3.203]) and Glasgow Coma Scale scores (0.782 [0.432-0.918]) independently associated with hematoma expanding. After 4-5 epochs, the CNN framework was well trained. The average sensitivity, specificity, and accuracy of CNN prediction are 0.9197, 0.8837, and 0.9058, respectively. Compared with 5 machine learning methods based on clinical variables, CNN can also achieve better performance.

CONCLUSIONS

More than 90% of hematomas with or without expansion can be precisely classified by deep learning technology within this study, which is better than other methods based on clinical variables only. Deep learning technology could favorably predict hematoma expansion from non-contrast CT scan images.

Artificial intelligence and machine learning in precision and genomic medicine

The advancement of precision medicine in medical care has led behind the conventional symptom-driven treatment process by allowing early risk prediction of disease through improved diagnostics and customization of more effective treatments. It is necessary to scrutinize overall patient data alongside broad factors to observe and differentiate between ill and relatively healthy people to take the most appropriate path toward precision medicine, resulting in an improved vision of biological indicators that can signal health changes. Precision and genomic medicine combined with artificial intelligence have the potential to improve patient healthcare. Patients with less common therapeutic responses or unique healthcare demands are using genomic medicine technologies. AI provides insights through advanced computation and inference, enabling the system to reason and learn while enhancing physician decision making. Many cell characteristics, including gene up-regulation, proteins binding to nucleic acids, and splicing, can be measured at high throughput and used as training objectives for predictive models. Researchers can create a new era of effective genomic medicine with the improved availability of a broad range of datasets and modern computer techniques such as machine learning. This review article has elucidated the contributions of ML algorithms in precision and genome medicine.

Artificial intelligence and machine learning in precision and genomic medicine

The advancement of precision medicine in medical care has led behind the conventional symptom-driven treatment process by allowing early risk prediction of disease through improved diagnostics and customization of more effective treatments. It is necessary to scrutinize overall patient data alongside broad factors to observe and differentiate between ill and relatively healthy people to take the most appropriate path toward precision medicine, resulting in an improved vision of biological indicators that can signal health changes. Precision and genomic medicine combined with artificial intelligence have the potential to improve patient healthcare. Patients with less common therapeutic responses or unique healthcare demands are using genomic medicine technologies. AI provides insights through advanced computation and inference, enabling the system to reason and learn while enhancing physician decision making. Many cell characteristics, including gene up-regulation, proteins binding to nucleic acids, and splicing, can be measured at high throughput and used as training objectives for predictive models. Researchers can create a new era of effective genomic medicine with the improved availability of a broad range of datasets and modern computer techniques such as machine learning. This review article has elucidated the contributions of ML algorithms in precision and genome medicine.

Radiomics Features on Computed Tomography Combined With Clinical-Radiological Factors Predicting Progressive Hemorrhage of Cerebral Contusion

Background

Traumatic brain injury (TBI) is the main cause of death and severe disability in young adults worldwide. Progressive hemorrhage (PH) worsens the disease and can cause a poor neurological prognosis. Radiomics analysis has been used for hematoma expansion of hypertensive intracerebral hemorrhage. This study attempts to develop an optimal radiomics model based on non-contrast CT to predict PH by machine learning (ML) methods and compare its prediction performance with clinical-radiological models.

Methods

We retrospectively analyzed 165 TBI patients, including 89 patients with PH and 76 patients without PH, whose data were randomized into a training set and a testing set at a ratio of 7:3. A total of 10 different machine learning methods were used to predict PH. Univariate and multivariable logistic regression analyses were implemented to screen clinical-radiological factors and to establish a clinical-radiological model. Then, a combined model combining clinical-radiological factors with the radiomics score was constructed. The area under the receiver operating characteristic curve (AUC), accuracy and F1 score, sensitivity, and specificity were used to evaluate the models.

Results

Among the 10 various ML algorithms, the support vector machine (SVM) had the best prediction performance based on 12 radiomics features, including the AUC (training set: 0.918; testing set: 0.879) and accuracy (training set: 0.872; test set: 0.834). Among the clinical and radiological factors, the onset-to-baseline CT time, the scalp hematoma, and fibrinogen were associated with PH. The radiomics model's prediction performance was better than the clinical-radiological model, while the predictive nomogram combining the radiomics features with clinical-radiological characteristics performed best.

Conclusions

The radiomics model outperformed the traditional clinical-radiological model in predicting PH. The nomogram model of the combined radiomics features and clinical-radiological factors is a helpful tool for PH.

Automatic and Efficient Prediction of Hematoma Expansion in Patients with Hypertensive Intracerebral Hemorrhage Using Deep Learning Based on CT Images

Patients with hypertensive intracerebral hemorrhage (ICH) have a high hematoma expansion (HE) incidence. Noninvasive prediction HE helps doctors take effective measures to prevent accidents. This study retrospectively analyzed 253 cases of hypertensive intraparenchymal hematoma. Baseline non-contrast-enhanced CT scans (NECTs) were collected at admission and compared with subsequent CTs to determine the presence of HE. An end-to-end deep learning method based on CT was proposed to automatically segment the hematoma region, region of interest (ROI) feature extraction, and HE prediction. A variety of algorithms were employed for comparison. U-Net with attention performs best in the task of segmenting hematomas, with the mean Intersection overUnion (mIoU) of 0.9025. ResNet-34 achieves the most robust generalization capability in HE prediction, with an area under the receiver operating characteristic curve (AUC) of 0.9267, an accuracy of 0.8827, and an F1 score of 0.8644. The proposed method is superior to other mainstream models, which will facilitate accurate, efficient, and automated HE prediction.

Prediction and Risk Assessment Models for Subarachnoid Hemorrhage: A Systematic Review on Case Studies

Subarachnoid hemorrhage (SAH) is one of the major health issues known to society and has a higher mortality rate. The clinical factors with computed tomography (CT), magnetic resonance image (MRI), and electroencephalography (EEG) data were used to evaluate the performance of the developed method. In this paper, various methods such as statistical analysis, logistic regression, machine learning, and deep learning methods were used in the prediction and detection of SAH which are reviewed. The advantages and limitations of SAH prediction and risk assessment methods are also being reviewed. Most of the existing methods were evaluated on the collected dataset for the SAH prediction. In some researches, deep learning methods were applied, which resulted in higher performance in the prediction process. EEG data were applied in the existing methods for the prediction process, and these methods demonstrated higher performance. However, the existing methods have the limitations of overfitting problems, imbalance data problems, and lower efficiency in feature analysis. The artificial neural network (ANN) and support vector machine (SVM) methods have been applied for the prediction process, and considerably higher performance is achieved by using this method.

Machine Learning in Action: Stroke Diagnosis and Outcome Prediction

The application of machine learning has rapidly evolved in medicine over the past decade. In stroke, commercially available machine learning algorithms have already been incorporated into clinical application for rapid diagnosis. The creation and advancement of deep learning techniques have greatly improved clinical utilization of machine learning tools and new algorithms continue to emerge with improved accuracy in stroke diagnosis and outcome prediction. Although imaging-based feature recognition and segmentation have significantly facilitated rapid stroke diagnosis and triaging, stroke prognostication is dependent on a multitude of patient specific as well as clinical factors and hence accurate outcome prediction remains challenging. Despite its vital role in stroke diagnosis and prognostication, it is important to recognize that machine learning output is only as good as the input data and the appropriateness of algorithm applied to any specific data set. Additionally, many studies on machine learning tend to be limited by small sample size and hence concerted efforts to collate data could improve evaluation of future machine learning tools in stroke. In the present state, machine learning technology serves as a helpful and efficient tool for rapid clinical decision making while oversight from clinical experts is still required to address specific aspects not accounted for in an automated algorithm. This article provides an overview of machine learning technology and a tabulated review of pertinent machine learning studies related to stroke diagnosis and outcome prediction.

Machine Learning Models Prognosticate Functional Outcomes Better than Clinical Scores in Spontaneous Intracerebral Haemorrhage

OBJECTIVE

This study aims to develop and compare the use of deep neural networks (DNN) and support vector machines (SVM) to clinical prognostic scores for prognosticating 30-day mortality and 90-day poor functional outcome (PFO) in spontaneous intracerebral haemorrhage (SICH).

MATERIALS AND METHODS

We conducted a retrospective cohort study of 297 SICH patients between December 2014 and May 2016. Clinical data was collected from electronic medical records using standardized data collection forms. The machine learning workflow included imputation of missing data, dimensionality reduction, imbalanced-class correction, and evaluation using cross-validation and comparison of accuracy against clinical prognostic scores.

RESULTS

32 (11%) patients had 30-day mortality while 177 (63%) patients had 90-day PFO. For prognosticating 30-day mortality, the class-balanced accuracies for DNN (0.875; 95% CI 0.800-0.950; McNemar's p-value 1.000) and SVM (0.848; 95% CI 0.767-0.930; McNemar's p-value 0.791) were comparable to that of the original ICH score (0.833; 95% CI 0.748-0.918). The c-statistics for DNN (0.895; DeLong's p-value 0.715), and SVM (0.900; DeLong's p-value 0.619), though greater than that of the original ICH score (0.862), were not significantly different. For prognosticating 90-day PFO, the class-balanced accuracies for DNN (0.853; 95% CI 0.772-0.934; McNemar's p-value 0.003) and SVM (0.860; 95% CI 0.781-0.939; McNemar's p-value 0.004) were better than that of the ICH-Grading Scale (0.706; 95% CI 0.600-0.812). The c-statistic for SVM (0.883; DeLong's p-value 0.022) was significantly greater than that of the ICH-Grading Scale (0.778), while the c-statistic for DNN was 0.864 (DeLong's p-value 0.055).

CONCLUSION

We showed that the SVM model performs significantly better than clinical prognostic scores in predicting 90-day PFO in SICH.

Artificial Intelligence: A Shifting Paradigm in Cardio-Cerebrovascular Medicine

The future of healthcare is an organic blend of technology, innovation, and human connection. As artificial intelligence (AI) is gradually becoming a go-to technology in healthcare to improve efficiency and outcomes, we must understand our limitations. We should realize that our goal is not only to provide faster and more efficient care, but also to deliver an integrated solution to ensure that the care is fair and not biased to a group of sub-population. In this context, the field of cardio-cerebrovascular diseases, which encompasses a wide range of conditions-from heart failure to stroke-has made some advances to provide assistive tools to care providers. This article aimed to provide an overall thematic review of recent development focusing on various AI applications in cardio-cerebrovascular diseases to identify gaps and potential areas of improvement. If well designed, technological engines have the potential to improve healthcare access and equitability while reducing overall costs, diagnostic errors, and disparity in a system that affects patients and providers and strives for efficiency.

Noncontrast Computed Tomography Markers of Cerebral Hemorrhage Expansion: Diagnostic Accuracy Meta-Analysis

BACKGROUND AND PURPOSE

Assess the diagnostic accuracy of noncontrast computed tomography (NCCT) markers of hematoma expansion in patients with primary intracerebral hemorrhage.

METHODS

We performed a meta-analysis of observational studies and randomized controlled trials with available data for calculation of sensitivity and specificity of NCCT markers for hematoma expansion (absolute growth >6 or 12.5 mL and/or relative growth >33%). The following NCCT markers were analyzed: irregular shape, island sign (shape-related features); hypodensity, heterogeneous density, blend sign, black hole sign, and swirl sign (density-related features). Pooled accuracy values for each marker were derived from hierarchical logistic regression models.

RESULTS

A total of 10,363 subjects from 23 eligible studies were included. Significant risk of bias of included studies was noted. Hematoma expansion frequency ranged from 7% to 40%, mean intracerebral hemorrhage volume from 9 to 27.8 ml, presence of NCCT markers from 9% (island sign) to 82% (irregular shape). Among shape features, sensitivity ranged from 0.32 (95%CI = 0.20-0.47) for island sign to 0.68 (95%CI = 0.57-0.77) for irregular shape, specificity ranged from 0.47 (95%CI = 0.36-0.59) for irregular shape to 0.92 (95%CI = 0.85-0.96) for island sign; among density features sensitivity ranged from 0.28 (95%CI = 0.21-0.35) for black hole sign to 0.63 (95%CI = 0.44-0.78) for hypodensity, specificity ranged from 0.65 (95%CI = 0.56-0.73) for heterogeneous density to 0.89 (95%CI = 0.85-0.92) for blend sign.

CONCLUSION

Diagnostic accuracy of NCCT markers remains suboptimal for implementation in clinical trials although density features performed better than shape-related features. This analysis may help in better tailoring patients' selection for hematoma expansion targeted trials.

Machine learning-based modified BAT score in predicting hematoma enlargement after spontaneous intracerebral hemorrhage

OBJECTIVE

The BAT score is an easy-to-use prediction tool to detect hematoma enlargement after spontaneous intracerebral hemorrhage. Machine learning technique has potential predictive gains in accuracy over regression models. We sought to use machine learning technique to improve the BAT score for the prediction of hematoma enlargement.

METHODS

Totally 232 patients with spontaneous intracerebral hemorrhage were enrolled from our hospital between 2015 and 2020. The BAT score was calculated to identify high-risk patients with hematoma enlargement. Using the same variables of the original BAT score and 5 common machine learning algorithms, the modified BAT scores were constructed in the training subset (n = 162) and validated in the testing subset (n = 70). Receiver operating characteristic curves were performed to evaluate the discriminative ability of all BAT scores.

RESULTS

Among 5 modified BAT scores, the modified BAT score based on Naive Bayes algorithm performed best, with the area under the receiver operating characteristic curve (AUC) of 0.83 in the training subset and 0.77 in the testing subset. The DeLong test showed that the performances of the modified BAT score based on Naive Bayes algorithm were significantly better than that of the BAT score (AUC = 0.57) in the training and testing subsets (both P < 0.001).

CONCLUSIONS

Machine learning technique could improve the identification performance of the original BAT score using the same variables. The modified BAT score based on Naive Bayes algorithm could be used as an effective prediction tool for identifying high-risk patients with hematoma enlargement.

An optimal deep learning framework for multi-type hemorrhagic lesions detection and quantification in head CT images for traumatic brain injury

Traumatic Brain Injury (TBI) could lead to intracranial hemorrhage (ICH), which has now been identified as a major cause of death after trauma if it is not adequately diagnosed and properly treated within the first 24 hours. CT examination is widely preferred for urgent ICH diagnosis, which enables the fast identification and detection of ICH regions. However, the use of it requires the clinical interpretation by experts to identify the subtypes of ICH. Besides, it is unable to provide the details needed to conduct quantitative assessment, such as the volume and thickness of hemorrhagic lesions, which may have prognostic importance to the decision-making on emergency treatment. In this paper, an optimal deep learning framework is proposed to assist the quantitative assessment for ICH diagnosis and the accurate detection of different subtypes of ICH through head CT scan. Firstly, the format of raw input data is converted from 3D DICOM to NIfTI. Secondly, a pre-trained multi-class semantic segmentation model is applied to each slice of CT images, so as to obtain a precise 3D mask of the whole ICH region. Thirdly, a fine-tuned classification neural network is employed to extract the key features from the raw input data and identify the subtypes of ICH. Finally, a quantitative assessment algorithm is adopted to automatically measure both thickness and volume via the 3D shape mask combined with the output probabilities of the classification network. The results of our extensive experiments demonstrate the effectiveness of the proposed framework where the average accuracy of 96.21 percent is achieved for three types of hemorrhage. The capability of our optimal classification model to distinguish between different types of lesion plays a significant role in reducing the false-positive rate in the existing work. Furthermore, the results suggest that our automatic quantitative assessment algorithm is effective in providing clinically relevant quantification in terms of volume and thickness. It is more important than the qualitative assessment conducted through visual inspection to the decision-making on emergency surgical treatment.

Dilated Optic Nerve Sheath Diameter Predicts Poor Outcome in Acute Spontaneous Intracerebral Hemorrhage

BACKGROUND AND PURPOSE

Optic nerve sheath diameter (ONSD) enlargement occurs in patients with intracerebral hemorrhage (ICH). However, the relationship between ONSD and prognosis of ICH is uncertain. This study aimed to investigate the predictive value of ONSD on poor outcome of patients with acute spontaneous ICH.

METHODS

We studied 529 consecutive patients with acute spontaneous ICH who underwent initial CT within 6 h of symptom onset between October 2016 and February 2019. The ONSDs were measured 3 mm behind the eyeball on initial CT images. Poor outcome was defined as having a Glasgow Outcome Scale (GOS) score of 1-3, and favorable outcome was defined as having a GOS score of 4-5 at discharge.

RESULTS

The ONSD of the poor outcome group was significantly greater than that of the favorable outcome group (5.87 ± 0.86 vs. 5.21 ± 0.69 mm, p < 0.001). ONSD was related to hematoma volume (r = 0.475, p < 0.001). Adjusting other meaningful predictors, ONSD (OR: 2.83; 95% CI: 1.94-4.15) was associated with poor functional outcome by multivariable logistic regression analysis. Receiver operating characteristic curve showed that the ONSD improved the accuracy of ultraearly hematoma growth in the prediction of poor outcome (AUC: 0.790 vs. 0.755, p = 0.016). The multivariable logistic regression model with all the meaningful predictors showed a better predictive performance than the model without ONSD (AUC: 0.862 vs. 0.831, p = 0.001).

CONCLUSIONS

The dilated ONSD measured on initial CT indicated elevated intracranial pressure and poor outcome, so appropriate intervention should be taken in time.

Neutrophil to lymphocyte ratio predicts early growth of traumatic intracerebral haemorrhage

OBJECTIVE

The neutrophil to lymphocyte ratio (NLR) has been proposed to capture the inflammatory status of patients with various conditions involving the brain. This retrospective study aimed to explore the association between the NLR and the early growth of traumatic intracerebral haemorrhage (tICH) in patients with traumatic brain injury (TBI).

METHODS

A multicentre, observational cohort study was conducted. Patients with cerebral contusion undergoing baseline computed tomography for haematoma volume analysis within 6 h after primary injury and follow-up visits within 48 h were included. Routine blood tests were performed upon admission, and early growth of tICH was assessed. Prediction accuracies of the NLR for the early growth of tICH and subsequent surgical intervention in patients were analysed.

RESULTS

There were a total of 1077 patients who met the criteria included in the study cohort. Univariate analysis results showed that multiple risk factors were associated with the early growth of tICH and included in the following multivariate analysis models. The multivariate logistic regression analysis results revealed that the NLR was highly associated with the early growth of tICH (p < 0.001) while considering other risk factors in the same model. The prediction accuracy of the NLR for the early growth of tICH in patients is 82%.

INTERPRETATION

The NLR is easily calculated and might predict the early growth of tICH for patients suffering from TBI.

Comparison of Supervised Machine Learning Algorithms for Classifying of Home Discharge Possibility in Convalescent Stroke Patients: A Secondary Analysis

OBJECTIVES

Classifying the possibility of home discharge is important during stroke rehabilitation to support decision-making. There have been several studies on supervised machine learning algorithms, but only a few have compared the performance of different algorithms based on the same dataset for the classification of home discharge possibility. Therefore, we aimed to evaluate five supervised machine learning algorithms for the classification of home discharge possibility in stroke patients.

MATERIALS AND METHODS

This was a secondary analysis based on the data of 481 stroke patients from the database of our institution. Five models developed by supervised machine learning algorithms, including decision tree (DT), linear discriminant analysis (LDA), k-nearest neighbors (k-NN), support vector machine (SVM), and random forest (RF) were compared by constructing a classification system based on the same dataset. Several parameters including classification accuracy, area under the curve (AUC), and F1 score (a weighted average of precision and recall) were used for model evaluation.

RESULTS

The k-NN model had the best classification accuracy (84.0%) with a moderate AUC (0.88) and F1 score (87.8). The SVM model also showed high classification accuracy (82.6%) along with the highest AUC (0.91), sensitivity (94.4), negative predictive value (87.5), and negative likelihood ratio (0.088). The DT, LDA, and RF models had high classification accuracies (≥ 79.9%) with moderate AUCs (≥ 0.84) and F1 scores (≥ 83.8).

CONCLUSIONS

Regarding model performance, the k-NN and SVM seemed the best candidate algorithms for classifying the possibility of home discharge in stroke patients.

Emerging role of artificial intelligence in stroke imaging

Introduction: The recognition and therapy of patients with stroke is becoming progressively intricate as additional treatment choices become accessible and new associations between disease characteristics and treatment response are incessantly uncovered. Therefore, clinicians must regularly learn new skill, stay up to date with the literature and integrate advances into daily practice. The application of artificial intelligence (AI) to assist clinical decision making could diminish inter-rater variation in routine clinical practice and accelerate the mining of vital data that could expand recognition of patients with stroke, forecast of treatment responses and patient outcomes.Areas covered: In this review, the authors provide an up-to-date review of AI in stroke, analyzing the latest papers on this subject. These have been divided in two main groups: stroke diagnosis and outcome prediction.Expert opinion: The highest value of AI is its capability to merge, select and condense a large amount of clinical and imaging features of a single patient and to associate these with fitted models that have gone through robust assessment and optimization with large cohorts of data to support clinical decision making.

Deep learning for automatically predicting early haematoma expansion in Chinese patients

Background and purpose

Early haematoma expansion is determinative in predicting outcome of intracerebral haemorrhage (ICH) patients. The aims of this study are to develop a novel prediction model for haematoma expansion by applying deep learning model and validate its prediction accuracy.

Methods

Data of this study were obtained from a prospectively enrolled cohort of patients with primary supratentorial ICH from our centre. We developed a deep learning model to predict haematoma expansion and compared its performance with conventional non-contrast CT (NCCT) markers. To evaluate the predictability of this model, it was also compared with a logistic regression model based on haematoma volume or the BAT score.

Results

A total of 266 patients were finally included for analysis, and 74 (27.8%) of them experienced early haematoma expansion. The deep learning model exhibited highest C statistic as 0.80, compared with 0.64, 0.65, 0.51, 0.58 and 0.55 for hypodensities, black hole sign, blend sign, fluid level and irregular shape, respectively. While the C statistics for swirl sign (0.70; p=0.211) and heterogenous density (0.70; p=0.141) were not significantly higher than that of the deep learning model. Moreover, the predictive value for the deep learning model was significantly superior to that of the logistic model of haematoma volume (0.62; p=0.042) and the BAT score (0.65; p=0.042).

Conclusions

Compared with the conventional NCCT markers and BAT predictive model, the deep learning algorithm showed superiority for predicting early haematoma expansion in ICH patients.

Artificial Intelligence and Acute Stroke Imaging

Artificial intelligence technology is a rapidly expanding field with many applications in acute stroke imaging, including ischemic and hemorrhage subtypes. Early identification of acute stroke is critical for initiating prompt intervention to reduce morbidity and mortality. Artificial intelligence can help with various aspects of the stroke treatment paradigm, including infarct or hemorrhage detection, segmentation, classification, large vessel occlusion detection, Alberta Stroke Program Early CT Score grading, and prognostication. In particular, emerging artificial intelligence techniques such as convolutional neural networks show promise in performing these imaging-based tasks efficiently and accurately. The purpose of this review is twofold: first, to describe AI methods and available public and commercial platforms in stroke imaging, and second, to summarize the literature of current artificial intelligence-driven applications for acute stroke triage, surveillance, and prediction.

Support vector machine based aphasia classification of transcranial magnetic stimulation language mapping in brain tumor patients

Repetitive TMS (rTMS) allows for non-invasive and transient disruption of local neuronal functioning. We used machine learning approaches to assess whether brain tumor patients can be accurately classified into aphasic and non-aphasic groups using their rTMS language mapping results as input features. Given that each tumor affects the subject-specific language networks differently, resulting in heterogenous rTMS functional mappings, we propose the use of machine learning strategies to classify potential patterns of rTMS language mapping results. We retrospectively included 90 patients with left perisylvian world health organization (WHO) grade II-IV gliomas that underwent presurgical navigated rTMS language mapping. Within our cohort, 29 of 90 (32.2%) patients suffered from at least mild aphasia as shown in the Aachen Aphasia Test based Berlin Aphasia Score (BAS). After spatial normalization to MNI 152 of all rTMS spots, we calculated the error rate (ER) in each stimulated cortical area (28 regions of interest, ROI) by automated anatomical labeling parcellation (AAL3) and IIT. We used a support vector machine (SVM) to classify significant areas in relation to aphasia. After feeding the ROIs into the SVM model, it revealed that in addition to age (w = 2.98), the ERs of the left supramarginal gyrus (w = 3.64), left inferior parietal gyrus (w = 2.28) and right pars triangularis (w = 1.34) contributed more than other features to the model. The model's sensitivity was 86.2%, the specificity was 82.0%, the overall accuracy was 85.5% and the AUC was 89.3%. Our results demonstrate an increased vulnerability of right inferior pars triangularis to rTMS in aphasic patients due to left perisylvian gliomas. This finding points towards a functional relevant involvement of the right pars triangularis in response to aphasia. The tumor location feature, specified by calculating overlaps with white and grey matter atlases, did not affect the SVM model. The left supramarginal gyrus as a feature improved our SVM model the most. Additionally, our results could point towards a decreasing potential for neuroplasticity with age.

Island Sign Predicts Hematoma Expansion and Poor Outcome After Intracerebral Hemorrhage: A Systematic Review and Meta-Analysis

Background: Early hematoma expansion (HE) occurs in patients with intracerebral hemorrhage (ICH) within the first few hours from ICH onset. Hematoma expansion has been considered as an independent predictor of poor clinical outcome and mortality after ICH. Island sign (IS) on the non-contrast computed tomography (NCCT) appears to increase the rate of detection of HE. However, there is insufficient evidence to declare that IS is an independent predictor for ICH patients prognosis and classification.

Objectives: To investigate whether IS on NCCT could predict HE and functional outcome following ICH.

Methods: Major databases were systematically searched, including PubMed, EMBASE, Cochrane library, and the Chinese database (CNKI, VIP, and Wanfang databases). Studies about the associations between IS and HE or IS and clinical outcome were included. The pooled result used the odds ratio (OR) with a 95% confidence interval (CI) as effect size. Heterogeneity and publication bias were assessed. Subgroup analysis and meta-regression were applied to detect potential factors of heterogeneity.

Results: Eleven studies with 4,310 patients were included in the final analysis. The average incidence rate of IS and HE were 21.58 and 33%, respectively. The ideal timing for assessing HE was also not uniform or standardized. We separately performed two meta-analyses. First, 10 studies were included to estimate the association between IS and HE. The pooled OR was statistically significant (OR = 7.61, 95% CI = 3.10–18.67, P < 0.001). Second, four studies were included in the meta-analysis, and the pooled result showed that IS had a significantly positive relationship with poor outcome (OR = 3.83, 95% CI = 2.51–5.85, P < 0.001).

Conclusions: This meta-analysis showed that NCCT IS is of great importance and value for evaluation of HE and poor outcome in patients with ICH. Future studies should focus on developing consensus guidelines, and more studies with large sample size and longitudinal design are needed to validate the conclusions.

Inter- and Intrarater Agreement of Spot Sign and Noncontrast CT Markers for Early Intracerebral Hemorrhage Expansion

Background: The aim of this study was to assess the inter- and intrarater reliability of noncontrast CT (NCCT) markers [Black Hole Sign (BH), Blend Sign (BS), Island Sign (IS), and Hypodensities (HD)] and Spot Sign (SS) on CTA in patients with spontaneous intracerebral hemorrhage (ICH). Methods: Patients with spontaneous ICH at three German tertiary stroke centers were retrospectively included. Each CT scan was rated for four NCCT markers and SS on CTA by two radiology residents. Raters were blind to all demographic and outcome data. Inter- and intrarater agreement was determined by Cohen’s kappa (κ) coefficient and percentage of agreement. Results: Interrater agreement was excellent in 473 included patients, ranging from 96% to 99%. Interrater κ ranged from 0.85 (95% CI [0.78–0.91]) to 0.97 (95% CI [0.94–0.99]) for NCCT markers and 0.93 (95% CI [0.88–0.98]) for SS, all p-values < 0.001. Intrarrater agreement ranged from 96% to 100%, with κ ranging from 0.85 (95% CI [0.78–0.91]) to 1.00 (95% CI [0.10–0.85]) for NCCT markers and 0.96 (95% CI [0.92–1.00]) for SS, all p-values < 0.001. Conclusions: NCCT imaging findings and SS on CTA have good-to-excellent inter- and intrarater reliabilities, with the highest agreement for BH and SS.

Association Between Eosinophilic Leukocyte Count and Hematoma Expansion in Acute Spontaneous Intracerebral Hemorrhage

Background/Objective: Hematoma expansion (HE) predicts poor outcome and is an appealing treatment target in spontaneous intracerebral hemorrhage (ICH). Clinical evidence has shown an association of HE with peripheral white blood cells (WBC) count, but the individual contributions of leukocyte subtypes between literatures are described inconsistently. Our aim was to determine the relationship between admission absolute and differential leukocyte counts and HE by using different growth definitions.

Methods: We analyzed spontaneous ICH patients who underwent baseline cranial computed tomography and blood sampling within 6 h of stroke onset in our institution between September 2013 and August 2018. Hematoma volume was calculated using a semiautomated 3-dimensional reconstruction algorithm. According to commonly used absolute or relative growth definitions (>6 mL, >12.5 mL, or >33%), we defined 5 types of HE. A propensity score-matching analysis was performed to evaluate the influence of complete blood count components on HE across the various growth definitions. The receiver operating characteristic analysis assessed the predictive ability of leukocyte counts for HE.

Results: A total of 1,066 patients were included, of whom 11–21% met the 5 HE definitions. After propensity score-matching, except using the definition of >12.5 mL growth or its combination with >33% growth, both WBC and neutrophil count were independently associated with reduced risk of HE (odds ratio [OR] for 10³ cells increase; OR, 0.86–0.99; all p < 0.05) after adjusting confounders in multivariate analyses. However, monocyte count was correlated with increased risk of HE under the usage of >12.5 mL expansion definition only (OR, 1.43; p = 0.024). There was no association between lymphocyte count and HE (all p > 0.05). Regardless of the growth definition, admission eosinophil count was directly associated with the risk of HE (OR, 6.92–31.60; all p < 0.05), and was the best predictive subtype with area under the curve 0.64, sensitivity 69.5%, and specificity 58.9% at the optimal cut-off value of 45 cells/μL.

Conclusions: Growth definition affects the relationship of HE with leukocyte subtypes counting. Eosinophil count robustly predicts HE, and may be a surrogate when using an inflammatory marker to help select acute ICH patients with high expansion risk for hemostasis treatment in clinical trial and practice.