Identifying Modifiable Predictors of Patient Outcomes After Intracerebral Hemorrhage with Machine Learning

CT-based deep learning model for predicting hospital discharge outcome in spontaneous intracerebral hemorrhage

OBJECTIVES

To predict the functional outcome of patients with intracerebral hemorrhage (ICH) using deep learning models based on computed tomography (CT) images.

METHODS

A retrospective, bi-center study of ICH patients was conducted. Firstly, a custom 3D convolutional model was built for predicting the functional outcome of ICH patients based on CT scans from randomly selected ICH patients in H training dataset collected from H hospital. Secondly, clinical data and radiological features were collected at admission and the Extreme Gradient Boosting (XGBoost) algorithm was used to establish a second model, named the XGBoost model. Finally, the Convolution model and XGBoost model were fused to build the third "Fusion model." Favorable outcome was defined as modified Rankin Scale score of 0-3 at discharge. The prognostic predictive accuracy of the three models was evaluated using an H test dataset and an external Y dataset, and compared with the performance of ICH score and ICH grading scale (ICH-GS).

RESULTS

A total of 604 patients with ICH were included in this study, of which 450 patients were in the H training dataset, 50 patients in the H test dataset, and 104 patients in the Y dataset. In the Y dataset, the areas under the curve (AUCs) of the Convolution model, XGBoost model, and Fusion model were 0.829, 0.871, and 0.905, respectively. The Fusion model prognostic performance exceeded that of ICH score and ICH-GS (p = 0.043 and p = 0.045, respectively).

CONCLUSIONS

Deep learning models have good accuracy for predicting functional outcome of patients with spontaneous intracerebral hemorrhage.

CLINICAL RELEVANCE STATEMENT

The proposed deep learning Fusion model may assist clinicians in predicting functional outcome and developing treatment strategies, thereby improving the survival and quality of life of patients with spontaneous intracerebral hemorrhage.

KEY POINTS

• Integrating clinical presentations, CT images, and radiological features to establish deep learning model for functional outcome prediction of patients with intracerebral hemorrhage. • Deep learning applied to CT images provides great help in prognosing functional outcome of intracerebral hemorrhage patients. • The developed deep learning model performs better than clinical prognostic scores in predicting functional outcome of patients with intracerebral hemorrhage.

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.

Predicting the recurrence of spontaneous intracerebral hemorrhage using a machine learning model

Background

Recurrence can worsen conditions and increase mortality in ICH patients. Predicting the recurrence risk and preventing or treating these patients is a rational strategy to improve outcomes potentially. A machine learning model with improved performance is necessary to predict recurrence.

Methods

We collected data from ICH patients in two hospitals for our retrospective training cohort and prospective testing cohort. The outcome was the recurrence within one year. We constructed logistic regression, support vector machine (SVM), decision trees, Voting Classifier, random forest, and XGBoost models for prediction.

Results

The model included age, NIHSS score at discharge, hematoma volume at admission and discharge, PLT, AST, and CRP levels at admission, use of hypotensive drugs and history of stroke. In internal validation, logistic regression demonstrated an AUC of 0.89 and precision of 0.81, SVM showed an AUC of 0.93 and precision of 0.90, the random forest achieved an AUC of 0.95 and precision of 0.93, and XGBoost scored an AUC of 0.95 and precision of 0.92. In external validation, logistic regression achieved an AUC of 0.81 and precision of 0.79, SVM obtained an AUC of 0.87 and precision of 0.76, the random forest reached an AUC of 0.92 and precision of 0.86, and XGBoost recorded an AUC of 0.93 and precision of 0.91.

Conclusion

The machine learning models performed better in predicting ICH recurrence than traditional statistical models. The XGBoost model demonstrated the best comprehensive performance for predicting ICH recurrence in the external testing cohort.

Using 30-day modified rankin scale score to predict 90-day score in patients with intracranial hemorrhage: Derivation and validation of prediction model

Whether 30-day modified Rankin Scale (mRS) scores can predict 90-day scores is unclear. This study derived and validated a model to predict ordinal 90-day mRS score in an intracerebral hemorrhage (ICH) population using 30-day mRS values and routinely available baseline variables. Adults enrolled in the Antihypertensive Treatment of Acute Cerebral Hemorrhage-2 (ATACH-2) trial between May 2011 and September 2015 with acute ICH, who were alive at 30 days and had mRS scores reported at both 30 and 90 days were included in this post-hoc analysis. A proportional odds regression model for predicting ordinal 90-day mRS scores was developed and internally validated using bootstrapping. Variables in the model included: mRS score at 30 days, age (years), hematoma volume (cm3), hematoma location (deep [basal ganglia, thalamus], lobar, or infratentorial), presence of intraventricular hemorrhage (IVH), baseline Glasgow Coma Scale (GCS) score, and National Institutes of Health Stroke Scale (NIHSS) score at randomization. We assessed model fit, calibration, discrimination, and agreement (ordinal, dichotomized functional independence), and EuroQol-5D ([EQ-5D] utility weighted) between predicted and observed 90-day mRS. A total of 898/1000 participants were included. Following bootstrap internal validation, our model (calibration slope = 0.967) had an optimism-corrected c-index of 0.884 (95% CI = 0.873-0.896) and R2 = 0.712 for 90-day mRS score. The weighted ĸ for agreement between observed and predicted ordinal 90-day mRS score was 0.811 (95% CI = 0.787-0.834). Agreement between observed and predicted functional independence (mRS score of 0-2) at 90 days was 74.3% (95% CI = 69.9-78.7%). The mean ± SD absolute difference between predicted and observed EQ-5D-weighted mRS score was negligible (0.005 ± 0.145). This tool allows practitioners and researchers to utilize clinically available information along with the mRS score 30 days after ICH to reliably predict the mRS score at 90 days.

Guidelines for Neuroprognostication in Critically Ill Adults with Intracerebral Hemorrhage

BACKGROUND

The objective of this document is to provide recommendations on the formal reliability of major clinical predictors often associated with intracerebral hemorrhage (ICH) neuroprognostication.

METHODS

A narrative systematic review was completed using the Grading of Recommendations Assessment, Development, and Evaluation methodology and the Population, Intervention, Comparator, Outcome, Timing, Setting questions. Predictors, which included both individual clinical variables and prediction models, were selected based on clinical relevance and attention in the literature. Following construction of the evidence profile and summary of findings, recommendations were based on Grading of Recommendations Assessment, Development, and Evaluation criteria. Good practice statements addressed essential principles of neuroprognostication that could not be framed in the Population, Intervention, Comparator, Outcome, Timing, Setting format.

RESULTS

Six candidate clinical variables and two clinical grading scales (the original ICH score and maximally treated ICH score) were selected for recommendation creation. A total of 347 articles out of 10,751 articles screened met our eligibility criteria. Consensus statements of good practice included deferring neuroprognostication-aside from the most clinically devastated patients-for at least the first 48-72 h of intensive care unit admission; understanding what outcomes would have been most valued by the patient; and counseling of patients and surrogates whose ultimate neurological recovery may occur over a variable period of time. Although many clinical variables and grading scales are associated with ICH poor outcome, no clinical variable alone or sole clinical grading scale was suggested by the panel as currently being reliable by itself for use in counseling patients with ICH and their surrogates, regarding functional outcome at 3 months and beyond or 30-day mortality.

CONCLUSIONS

These guidelines provide recommendations on the formal reliability of predictors of poor outcome in the context of counseling patients with ICH and surrogates and suggest broad principles of neuroprognostication. Clinicians formulating their judgments of prognosis for patients with ICH should avoid anchoring bias based solely on any one clinical variable or published clinical grading scale.

Development and validation of a random forest model to predict functional outcome in patients with intracerebral hemorrhage

OBJECTIVE

To develop and validate a machine learning (ML)-based model to predict functional outcome in Chinese patients with intracerebral hemorrhage (ICH).

METHODS

This retrospective cohort study enrolled patients with ICH between November 2017 and November 2020. The follow-up period ended in February 2021. The study population was divided into training and testing sets with a ratio of 7:3. All variables were included in the least absolute shrinkage and selection operator (LASSO) regression for feature selection. The selected variables were incorporated into the random forest algorithm to construct the prediction model. The predictive performance of the model was evaluated via the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and calibration curve.

RESULTS

A total of 412 ICH patients were included, with 288 in the training set, and 124 in the testing set. Twelve attributes were selected: neurological deterioration, Glasgow Coma Scale (GCS) score at 24 h, baseline GCS score, time from onset to the emergency room, blood glucose, diastolic blood pressure (DBP) change in 24 h, hematoma volume change in 24 h, systemic immune-inflammatory index (SII), systolic blood pressure (SBP) change in 24 h, serum creatinine, serum sodium, and age. In the testing set, the accuracy, AUC, sensitivity, specificity, PPV, and NPV of the model were 0.895, 0.964, 0.872, 0.906, 0.810, and 0.939, respectively. The calibration curves showed a good calibration capability of the model.

CONCLUSION

This developed random forest model performed well in predicting 3-month poor functional outcome for Chinese ICH patients.

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.

Developing DELPHI expert consensus rules for a digital twin model of acute stroke care in the neuro critical care unit

INTRODUCTION

Digital twins, a form of artificial intelligence, are virtual representations of the physical world. In the past 20 years, digital twins have been utilized to track wind turbines' operations, monitor spacecraft's status, and even create a model of the Earth for climate research. While digital twins hold much promise for the neurocritical care unit, the question remains on how to best establish the rules that govern these models. This model will expand on our group's existing digital twin model for the treatment of sepsis.

METHODS

The authors of this project collaborated to create a Direct Acyclic Graph (DAG) and an initial series of 20 DELPHI statements, each with six accompanying sub-statements that captured the pathophysiology surrounding the management of acute ischemic strokes in the practice of Neurocritical Care (NCC). Agreement from a panel of 18 experts in the field of NCC was collected through a 7-point Likert scale with consensus defined a-priori by ≥ 80% selection of a 6 ("agree") or 7 ("strongly agree"). The endpoint of the study was defined as the completion of three separate rounds of DELPHI consensus. DELPHI statements that had met consensus would not be included in subsequent rounds of DELPHI consensus. The authors refined DELPHI statements that did not reach consensus with the guidance of de-identified expert comments for subsequent rounds of DELPHI. All DELPHI statements that reached consensus by the end of three rounds of DELPHI consensus would go on to be used to inform the construction of the digital twin model.

RESULTS

After the completion of three rounds of DELPHI, 93 (77.5%) statements reached consensus, 11 (9.2%) statements were excluded, and 16 (13.3%) statements did not reach a consensus of the original 120 DELPHI statements.

CONCLUSION

This descriptive study demonstrates the use of the DELPHI process to generate consensus among experts and establish a set of rules for the development of a digital twin model for use in the neurologic ICU. Compared to associative models of AI, which develop rules based on finding associations in datasets, digital twin AI created by the DELPHI process are easily interpretable models based on a current understanding of underlying physiology.

Big Data in Stroke: How to Use Big Data to Make the Next Management Decision

The last decade has seen significant advances in the accumulation of medical data, the computational techniques to analyze that data, and corresponding improvements in management. Interventions such as thrombolytics and mechanical thrombectomy improve patient outcomes after stroke in selected patients; however, significant gaps remain in our ability to select patients, predict complications, and understand outcomes. Big data and the computational methods needed to analyze it can address these gaps. For example, automated analysis of neuroimaging to estimate the volume of brain tissue that is ischemic and salvageable can help triage patients for acute interventions. Data-intensive computational techniques can perform complex risk calculations that are too cumbersome to be completed by humans, resulting in more accurate and timely prediction of which patients require increased vigilance for adverse events such as treatment complications. To handle the accumulation of complex medical data, a variety of advanced computational techniques referred to as machine learning and artificial intelligence now routinely complement traditional statistical inference. In this narrative review, we explore data-intensive techniques in stroke research, how it has informed the management of stroke patients, and how current work could shape clinical practice in the future.

An efficient edge/cloud medical system for rapid detection of level of consciousness in emergency medicine based on explainable machine learning models

Emergency medicine (EM) is one of the attractive research fields in which researchers investigate their efforts to diagnose and treat unforeseen illnesses or injuries. There are many tests and observations are involved in EM. Detection of the level of consciousness is one of these observations, which can be detected using several methods. Among these methods, the automatic estimation of the Glasgow coma scale (GCS) is studied in this paper. The GCS is a medical score used to describe a patient’s level of consciousness. This type of scoring system requires medical examination that may not be available with the shortage of the medical expert. Therefore, the automatic medical calculation for a patient’s level of consciousness is highly needed. Artificial intelligence has been deployed in several applications and appears to have a high performance regarding providing automatic solutions. The main objective of this work is to introduce the edge/cloud system to improve the efficiency of the consciousness measurement through efficient local data processing. Moreover, an efficient machine learning (ML) model to predict the level of consciousness of a certain patient based on the patient’s demographic, vital signs, and laboratory tests is proposed, as well as maintaining the explainability issue using Shapley additive explanations (SHAP) that provides natural language explanation in a form that helps the medical expert to understand the final prediction. The developed ML model is validated using vital signs and laboratory tests extracted from the MIMIC III dataset, and it achieves superior performance (mean absolute error (MAE) = 0.269, mean square error (MSE) = 0.625, R² score = 0.964). The resulting model is accurate, medically intuitive, and trustworthy.

Radiomics features from perihematomal edema for prediction of prognosis in the patients with basal ganglia hemorrhage

Objective

We developed and validated a clinical-radiomics nomogram to predict the prognosis of basal ganglia hemorrhage patients.

Methods

Retrospective analyses were conducted in 197 patients with basal ganglia hemorrhage (training cohort: n = 136, test cohort: n = 61) who were admitted to The First Affiliated Hospital of Shandong First Medical University (Shandong Provincial Qianfoshan Hospital) and underwent computed tomography (CT) scan. According to different prognoses, patients with basal ganglia hemorrhage were divided into two groups. Independent clinical risk factors were derived with univariate and multivariate regression analysis. Radiomics signatures were obtained using least absolute shrinkage and selection operator. A radiomics score (Rad-score) was generated by 12 radiomics signatures of perihematomal edema (PHE) from CT images that were correlated with the prognosis of basal ganglia hemorrhage patients. A clinical-radiomics nomogram was conducted by combing the Rad-score and clinical risk factors using logistic regression analysis. The prediction performance of the nomogram was tested in the training cohort and verified in the test cohort.

Results

The clinical model conducted by four clinical risk factors and 12 radiomcis features were used to establish the Rad-score. The clinical-radiomics nomogram outperformed the clinical model in the training cohort [area under the curve (AUC), 0.92 vs. 0.85] and the test cohort (AUC, 0.91 vs 0.85). The clinical-radiomics nomogram showed good calibration and clinical benefit in both the training and test cohorts.

Conclusion

Radiomics features of PHE in patients with basal ganglia hemorrhage could contribute to the outcome prediction. The clinical-radiomics nomogram may help first-line clinicians to make individual clinical treatment decisions for patients with basal ganglia hemorrhage.

Predicting Early Seizures After Intracerebral Hemorrhage with Machine Learning

BACKGROUND

Seizures are a harmful complication of acute intracerebral hemorrhage (ICH). "Early" seizures in the first week after ICH are a risk factor for deterioration, later seizures, and herniation. Ideally, seizure medications after ICH would only be administered to patients with a high likelihood to have seizures. We developed and validated machine learning (ML) models to predict early seizures after ICH.

METHODS

We used two large datasets to train and then validate our models in an entirely independent test set. The first model ("CAV") predicted early seizures from a subset of variables of the CAVE score (a prediction rule for later seizures)-cortical hematoma location, age less than 65 years, and hematoma volume greater than 10 mL-whereas early seizure was the dependent variable. We attempted to improve on the "CAV" model by adding anticoagulant use, antiplatelet use, Glasgow Coma Scale, international normalized ratio, and systolic blood pressure ("CAV + "). For each model we used logistic regression, lasso regression, support vector machines, boosted trees (Xgboost), and random forest models. Final model performance was reported as the area under the receiver operating characteristic curve (AUC) using receiver operating characteristic models for the test data. The setting of the study was two large academic institutions: institution 1, 634 patients; institution 2, 230 patients. There were no interventions.

RESULTS

Early seizures were predicted across the ML models by the CAV score in test data, (AUC 0.72, 95% confidence interval 0.62-0.82). The ML model that predicted early seizure better in the test data was Xgboost (AUC 0.79, 95% confidence interval 0.71-0.87, p = 0.04) compared with the CAV model AUC.

CONCLUSIONS

Early seizures after ICH are predictable. Models using cortical hematoma location, age less than 65 years, and hematoma volume greater than 10 mL had a good accuracy rate, and performance improved with more independent variables. Additional methods to predict seizures could improve patient selection for monitoring and prophylactic seizure medications.

Machine learning model prediction of 6-month functional outcome in elderly patients with intracerebral hemorrhage

Spontaneous intracerebral hemorrhage (ICH) has an increasing incidence and a worse outcome in elderly patients. The ability to predict the functional outcome in these patients can be helpful in supporting treatment decisions and establishing prognostic expectations. We evaluated the performance of a machine learning (ML) model to predict the 6-month functional status in elderly patients with ICH leveraging the predictive value of the clinical characteristics at hospital admission. Data were extracted by a retrospective multicentric database of patients ≥ 70 years of age consecutively admitted for the management of spontaneous ICH between January 1, 2014 and December 31, 2019. Relevant demographic, clinical, and radiological variables were selected by a feature selection algorithm (Boruta) and used to build a ML model. Outcome was determined according to the Glasgow Outcome Scale (GOS) at 6 months from ICH: dead (GOS 1), poor outcome (GOS 2–3: vegetative status/severe disability), and good outcome (GOS 4–5: moderate disability/good recovery). Ten features were selected by Boruta with the following relative importance order in the ML model: Glasgow Coma Scale, Charlson Comorbidity Index, ICH score, ICH volume, pupillary status, brainstem location, age, anticoagulant/antiplatelet agents, intraventricular hemorrhage, and cerebellar location. Random forest prediction model, evaluated on the hold-out test set, achieved an AUC of 0.96 (0.94–0.98), 0.89 (0.86–0.93), and 0.93 (0.90–0.95) for dead, poor, and good outcome classes, respectively, demonstrating high discriminative ability. A random forest classifier was successfully trained and internally validated to stratify elderly patients with spontaneous ICH into prognostic subclasses. The predictive value is enhanced by the ability of ML model to identify synergy among variables.

Can Artificial Intelligence Be Applied to Diagnose Intracerebral Hemorrhage under the Background of the Fourth Industrial Revolution? A Novel Systemic Review and Meta-Analysis

Aim

We intended to provide the clinical evidence that artificial intelligence (AI) could be used to assist doctors in the diagnosis of intracerebral hemorrhage (ICH).

Methods

Studies published in 2021 were identified after the literature search of PubMed, Embase, and Cochrane. Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) was used to perform the quality assessment of studies. Data extraction of diagnosis effect included accuracy (ACC), sensitivity (SEN), specificity (SPE), positive predictive value (PPV), negative predictive value (NPV), area under curve (AUC), and Dice scores (Dices). The pooled effect with its 95% confidence interval (95%CI) was calculated by the random effects model. I-Square (I 2) was used to test heterogeneity. To check the stability of the overall results, sensitivity analysis was conducted by recalculating the pooled effect of the remaining studies after omitting the study with the highest quality or the random effects model was switched to the fixed effects model. Funnel plot was used to evaluate publication bias. To reduce heterogeneity, recalculating the pooled effect of the remaining studies after omitting the study with the lowest quality or perform subgroup analysis.

Results

Twenty-five diagnostic tests of ICH via AI and doctors with overall high quality were included. Pooled ACC, SEN, SPE, PPV, NPV, AUC, and Dices were 0.88 (0.83∼0.93), 0.85 (0.81∼0.89), 0.90 (0.88∼0.92), 0.80 (0.75∼0.85), 0.93 (0.91∼0.95), 0.84 (0.80∼0.89), and 0.90 (0.85∼0.95), respectively. There was no publication bias. All of results were stable as revealed by sensitivity analysis and were accordant as outcomes via subgroups analysis.

Conclusion

Under the background of the fourth industrial revolution, AI might be an effective and efficient tool to assist doctors in the clinical diagnosis of ICH.

Enhancing Robustness of Machine Learning Integration With Routine Laboratory Blood Tests to Predict Inpatient Mortality After Intracerebral Hemorrhage

Objective: The accurate evaluation of outcomes at a personalized level in patients with intracerebral hemorrhage (ICH) is critical clinical implications. This study aims to evaluate how machine learning integrates with routine laboratory tests and electronic health records (EHRs) data to predict inpatient mortality after ICH.

Methods: In this machine learning-based prognostic study, we included 1,835 consecutive patients with acute ICH between October 2010 and December 2018. The model building process incorporated five pre-implant ICH score variables (clinical features) and 13 out of 59 available routine laboratory parameters. We assessed model performance according to a range of learning metrics, such as the mean area under the receiver operating characteristic curve [AUROC]. We also used the Shapley additive explanation algorithm to explain the prediction model.

Results: Machine learning models using laboratory data achieved AUROCs of 0.71–0.82 in a split-by-year development/testing scheme. The non-linear eXtreme Gradient Boosting model yielded the highest prediction accuracy. In the held-out validation set of development cohort, the predictive model using comprehensive clinical and laboratory parameters outperformed those using clinical alone in predicting in-hospital mortality (AUROC [95% bootstrap confidence interval], 0.899 [0.897–0.901] vs. 0.875 [0.872–0.877]; P <0.001), with over 81% accuracy, sensitivity, and specificity. We observed similar performance in the testing set.

Conclusions: Machine learning integrated with routine laboratory tests and EHRs could significantly promote the accuracy of inpatient ICH mortality prediction. This multidimensional composite prediction strategy might become an intelligent assistive prediction for ICH risk reclassification and offer an example for precision medicine.

Machine Learning-Based Approaches for Prediction of Patients’ Functional Outcome and Mortality after Spontaneous Intracerebral Hemorrhage

Spontaneous intracerebral hemorrhage (SICH) has been common in China with high morbidity and mortality rates. This study aims to develop a machine learning (ML)-based predictive model for the 90-day evaluation after SICH. We retrospectively reviewed 751 patients with SICH diagnosis and analyzed clinical, radiographic, and laboratory data. A modified Rankin scale (mRS) of 0–2 was defined as a favorable functional outcome, while an mRS of 3–6 was defined as an unfavorable functional outcome. We evaluated 90-day functional outcome and mortality to develop six ML-based predictive models and compared their efficacy with a traditional risk stratification scale, the intracerebral hemorrhage (ICH) score. The predictive performance was evaluated by the areas under the receiver operating characteristic curves (AUC). A total of 553 patients (73.6%) reached the functional outcome at the 3rd month, with the 90-day mortality rate of 10.2%. Logistic regression (LR) and logistic regression CV (LRCV) showed the best predictive performance for functional outcome (AUC = 0.890 and 0.887, respectively), and category boosting presented the best predictive performance for the mortality (AUC = 0.841). Therefore, ML might be of potential assistance in the prediction of the prognosis of SICH.

Association Between Intraventricular Alteplase Use and Parenchymal Hematoma Volume in Patients With Spontaneous Intracerebral Hemorrhage and Intraventricular Hemorrhage

IMPORTANCE

Intraventricular thrombolysis reduces intraventricular hemorrhage (IVH) volume in patients with spontaneous intracerebral hemorrhage (ICH), but it is unclear if a similar association with parenchymal ICH volume exists.

OBJECTIVE

To evaluate the association between intraventricular alteplase use and ICH volume as well as the association between a change in parenchymal ICH volume and long-term functional outcomes.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study was a post hoc exploratory analysis of data from the Clot Lysis: Evaluating Accelerated Resolution of Intraventricular Hemorrhage phase 3 randomized clinical trial with blinded outcome assessments. Between September 1, 2009, and January 31, 2015, patients with ICH and IVH were randomized to receive either intraventricular alteplase or normal saline via an external ventricular drain. Participants with primary IVH were excluded. Data analyses were performed between January 1 and June 30, 2021.

EXPOSURE

Randomization to receive intraventricular alteplase.

MAIN OUTCOMES AND MEASURES

The primary outcome was the change in parenchymal ICH volume between the hematoma stability and end-of-treatment computed tomography scans. Secondary outcomes were a modified Rankin Scale score higher than 3 and mortality, both of which were assessed at 6 months. The association between alteplase and change in parenchymal ICH volume was assessed using multiple linear regression, whereas the associations between change in parenchymal ICH volume and 6-month outcomes were assessed using multiple logistic regression. Prespecified subgroup analyses were performed for baseline IVH volume, admission ICH volume, and ICH location.

RESULTS

A total of 454 patients (254 men [55.9%]; mean [SD] age, 59 [11] years) were included in the study. Of these patients, 230 (50.7%) were randomized to receive alteplase and 224 (49.3%) to receive normal saline. The alteplase group had a greater mean (SD) reduction in parenchymal ICH volume compared with the saline group (1.8 [0.2] mL vs 0.4 [0.1] mL; P < .001). In the primary analysis, alteplase use was associated with a change in the parenchymal ICH volume in the unadjusted analysis per 1-mL change (β, 1.37; 95% CI, 0.92-1.81; P < .001) and in multivariable linear regression analysis that was adjusted for demographic characteristics, stability ICH and IVH volumes, ICH location, and time to first dose of study drug per 1-mL change (β, 1.20; 95% CI, 0.79-1.62; P < .001). In the secondary analyses, no association was found between change in parenchymal ICH volume and poor outcome (odds ratio [OR], 0.97; 95% CI 0.87-1.10; P = .64) or mortality (OR, 0.97; 95% CI 0.99-1.08; P = .59). Similar results were observed in the subgroup analyses.

CONCLUSIONS AND RELEVANCE

This study found that intraventricular alteplase use in patients with a large IVH was associated with a small reduction in parenchymal ICH volume, but this association did not translate into improved functional outcomes or mortality. Intraventricular thrombolysis should be examined in patients with moderate to large ICH with IVH, especially in a thalamic location.

Postsurgical functional outcome prediction model using deep learning framework (Prediction One, Sony Network Communications Inc.) for hypertensive intracerebral hemorrhage

BACKGROUND

Reliable prediction models of intracerebral hemorrhage (ICH) outcomes are needed for decision-making of the treatment. Statistically making such prediction models needs a large number of samples and time-consuming statistical analysis. Deep learning (DL), one of the artificial intelligence, is attractive, but there were no reports on DL-based functional outcome prediction models for ICH outcomes after surgery. We herein made a functional outcome prediction model using DLframework, Prediction One (Sony Network Communications Inc., Tokyo, Japan), and compared it to original ICH score, ICH Grading Scale, and FUNC score.

METHODS

We used 140 consecutive hypertensive ICH patients' data in our hospital between 2012 and 2019. All patients were surgically treated. Modified Rankin Scale 0-3 at 6 months was defined as a favorable outcome. We randomly divided them into 100 patients training dataset and 40 patients validation dataset. Prediction One made the prediction model using the training dataset with 5-fold cross-validation. We calculated area under the curves (AUCs) regarding the outcome using the DL-based model, ICH score, ICH Grading Scale, and FUNC score. The AUCs were compared.

RESULTS

The model made by Prediction One using 64 variables had AUC of 0.997 in the training dataset and that of 0.884 in the validation dataset. These AUCs were superior to those derived from ICH score, ICH Grading Scale, and FUNC score.

CONCLUSION

We easily and quickly made prediction models using Prediction One, even with a small single-center dataset. The accuracy of the DL-based model was superior to those of previous statistically calculated models.