Dynamic Detection of Delayed Cerebral Ischemia: A Study in 3 Centers

Integrating Clinical Data and Radiomics and Deep Learning Features for End-to-End Delayed Cerebral Ischemia Prediction on Noncontrast CT

BACKGROUND AND PURPOSE

Delayed cerebral ischemia is hard to diagnose early due to gradual, symptomless development. This study aimed to develop an automated model for predicting delayed cerebral ischemia following aneurysmal SAH on NCCT.

MATERIALS AND METHODS

This retrospective study included 400 patients with aneurysmal SAH (156 with delayed cerebral ischemia) who underwent NCCT. The study used ATT-Deeplabv3+ for automatically segmenting hemorrhagic regions using semisupervised learning. Principal component analysis was used for reducing the dimensionality of deep learning features extracted from the average pooling layer of ATT-DeepLabv3+. The classification model integrated clinical data, radiomics, and deep learning features to predict delayed cerebral ischemia. Feature selection involved Pearson correlation coefficients, least absolute shrinkage, and selection operator regression. We developed models based on clinical features, clinical-radiomics, and a combination of clinical, radiomics, and deep learning. The study selected logistic regression, Naive Bayes, Adaptive Boosting (AdaBoost), and multilayer perceptron as classifiers. The performance of segmentation and classification models was evaluated on their testing sets using the Dice similarity coefficient for segmentation, and the area under the receiver operating characteristic curve (AUC) and calibration curves for classification.

RESULTS

The segmentation process achieved a Dice similarity coefficient of 0.91 and the average time of 0.037 s/image. Seventeen features were selected to calculate the radiomics score. The clinical-radiomics-deep learning model with multilayer perceptron achieved the highest AUC of 0.84 (95% CI, 0.72-0.97), which outperformed the clinical-radiomics model (P = .002) and the clinical features model (P = .001) with multilayer perceptron. The performance of clinical-radiomics-deep learning model using AdaBoost was significantly superior to its clinical-radiomics model (P = .027). The performance of the clinical-radiomics-deep learning model and the clinical-radiomics model with logistic regression notably exceeded that of the model based solely on clinical features (P = .028; P = .046). The AUC of the clinical-radiomics-deep learning model with multilayer perceptron (P < .001) and the clinical-radiomics model with logistic regression (P = .046) were significantly higher than the clinical model with logistic regression. Of all models, the clinical-radiomics-deep learning model with multilayer perceptron showed best calibration.

CONCLUSIONS

The proposed 2-stage end-to-end model not only achieves rapid and accurate segmentation but also demonstrates superior diagnostic performance with high AUC values and good calibration in the clinical-radiomics-deep learning model, suggesting its potential to enhance delayed cerebral ischemia detection and treatment strategies.

Machine Learning Algorithms to Predict Delayed Cerebral Ischemia After Subarachnoid Hemorrhage: A Systematic Review and Meta-analysis

Delayed cerebral ischemia (DCI) is a common and severe complication after subarachnoid hemorrhage (SAH). Logistic regression (LR) is the primary method to predict DCI, but it has low accuracy. This study assessed whether other machine learning (ML) models can predict DCI after SAH more accurately than conventional LR. PubMed, Embase, and Web of Science were systematically searched for studies directly comparing LR and other ML algorithms to forecast DCI in patients with SAH. Our main outcome was the accuracy measurement, represented by sensitivity, specificity, and area under the receiver operating characteristic. In the six studies included, comprising 1828 patients, about 28% (519) developed DCI. For LR models, the pooled sensitivity was 0.71 (95% confidence interval [CI] 0.57-0.84; p < 0.01) and the pooled specificity was 0.63 (95% CI 0.42-0.85; p < 0.01). For ML models, the pooled sensitivity was 0.74 (95% CI 0.61-0.86; p < 0.01) and the pooled specificity was 0.78 (95% CI 0.71-0.86; p = 0.02). Our results suggest that ML algorithms performed better than conventional LR at predicting DCI.Trial Registration: PROSPERO (International Prospective Register of Systematic Reviews) CRD42023441586; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=441586.

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.

Sex-Related Differences in Mortality, Delayed Cerebral Ischemia, and Functional Outcomes in Patients with Aneurysmal Subarachnoid Hemorrhage: A Systematic Review and Meta-Analysis

Background/Objective: Sex-related differences among patients with aneurysmal subarachnoid hemorrhage (aSAH) and their potential clinical implications have been insufficiently investigated. To address this knowledge gap, we conduct a comprehensive systematic review and meta-analysis. Methods: Sex-specific differences in patients with aSAH, including mortality, delayed cerebral ischemia (DCI), and functional outcomes were assessed. The functional outcome was dichotomized into favorable or unfavorable based on the modified Rankin Scale (mRS), Glasgow Outcome Scale (GOS), and Glasgow Outcome Scale Extended (GOSE). Results: Overall, 2823 studies were identified in EMBASE, MEDLINE, PubMed, and by manual search on 14 February 2024. After an initial assessment, 74 studies were included in the meta-analysis. In the analysis of mortality, including 18,534 aSAH patients, no statistically significant differences could be detected (risk ratio (RR) 0.99; 95% CI, 0.90-1.09; p = 0.91). In contrast, the risk analysis for DCI, including 23,864 aSAH patients, showed an 11% relative risk reduction in DCI in males versus females (RR, 0.89; 95% CI, 0.81-0.97; p = 0.01). The functional outcome analysis (favorable vs. unfavorable), including 7739 aSAH patients, showed a tendency towards better functional outcomes in men than women; however, this did not reach statistical significance (RR, 1.02; 95% CI, 0.98-1.07; p = 0.34). Conclusions: In conclusion, the available data suggest that sex/gender may play a significant role in the risk of DCI in patients with aSAH, emphasizing the need for sex-specific management strategies.

A generalizable physiological model for detection of Delayed Cerebral Ischemia using Federated Learning

Delayed cerebral ischemia (DCI) is a complication seen in patients with subarachnoid hemorrhage stroke. It is a major predictor of poor outcomes and is detected late. Machine learning models are shown to be useful for early detection, however training such models suffers from small sample sizes due to rarity of the condition. Here we propose a Federated Learning approach to train a DCI classifier across three institutions to overcome challenges of sharing data across hospitals. We developed a framework for federated feature selection and built a federated ensemble classifier. We compared the performance of FL model to that obtained by training separate models at each site. FL significantly improved performance at only two sites. We found that this was due to feature distribution differences across sites. FL improves performance in sites with similar feature distributions, however, FL can worsen performance in sites with heterogeneous distributions. The results highlight both the benefit of FL and the need to assess dataset distribution similarity before conducting FL.

Artificial Intelligence and Neurosurgery: Tracking Antiplatelet Response Patterns for Endovascular Intervention

Platelets play a critical role in blood clotting and the development of arterial blockages. Antiplatelet therapy is vital for preventing recurring events in conditions like coronary artery disease and strokes. However, there is a lack of comprehensive guidelines for using antiplatelet agents in elective neurosurgery. Continuing therapy during surgery poses a bleeding risk, while discontinuing it before surgery increases the risk of thrombosis. Discontinuation is recommended in neurosurgical settings but carries an elevated risk of ischemic events. Conversely, maintaining antithrombotic therapy may increase bleeding and the need for transfusions, leading to a poor prognosis. Artificial intelligence (AI) holds promise in making difficult decisions regarding antiplatelet therapy. This paper discusses current clinical guidelines and supported regimens for antiplatelet therapy in neurosurgery. It also explores methodologies like P2Y12 reaction units (PRU) monitoring and thromboelastography (TEG) mapping for monitoring the use of antiplatelet regimens as well as their limitations. The paper explores the potential of AI to overcome such limitations associated with PRU monitoring and TEG mapping. It highlights various studies in the field of cardiovascular and neuroendovascular surgery which use AI prediction models to forecast adverse outcomes such as ischemia and bleeding, offering assistance in decision-making for antiplatelet therapy. In addition, the use of AI to improve patient adherence to antiplatelet regimens is also considered. Overall, this research aims to provide insights into the use of antiplatelet therapy and the role of AI in optimizing treatment plans in neurosurgical settings.

Solving the explainable AI conundrum by bridging clinicians' needs and developers' goals

Explainable artificial intelligence (XAI) has emerged as a promising solution for addressing the implementation challenges of AI/ML in healthcare. However, little is known about how developers and clinicians interpret XAI and what conflicting goals and requirements they may have. This paper presents the findings of a longitudinal multi-method study involving 112 developers and clinicians co-designing an XAI solution for a clinical decision support system. Our study identifies three key differences between developer and clinician mental models of XAI, including opposing goals (model interpretability vs. clinical plausibility), different sources of truth (data vs. patient), and the role of exploring new vs. exploiting old knowledge. Based on our findings, we propose design solutions that can help address the XAI conundrum in healthcare, including the use of causal inference models, personalized explanations, and ambidexterity between exploration and exploitation mindsets. Our study highlights the importance of considering the perspectives of both developers and clinicians in the design of XAI systems and provides practical recommendations for improving the effectiveness and usability of XAI in healthcare.

Pathophysiology of Early Brain Injury and Its Association with Delayed Cerebral Ischemia in Aneurysmal Subarachnoid Hemorrhage: A Review of Current Literature

Background: Delayed cerebral ischemia (DCI) is a common and serious complication of aneurysmal subarachnoid hemorrhage (aSAH). Though many clinical trials have looked at therapies for DCI and vasospasm in aSAH, along with reducing rebleeding risks, none have led to improving outcomes in this patient population. We present an up-to-date review of the pathophysiology of DCI and its association with early brain injury (EBI). Recent Findings: Recent studies have demonstrated that EBI, as opposed to delayed brain injury, is the main contributor to downstream pathophysiological mechanisms that play a role in the development of DCI. New predictive models, including advanced monitoring and neuroimaging techniques, can help detect EBI and improve the clinical management of aSAH patients. Summary: EBI, the severity of subarachnoid hemorrhage, and physiological/imaging markers can serve as indicators for potential early therapeutics in aSAH. The microcellular milieu and hemodynamic pathomechanisms should remain a focus of researchers and clinicians. With the advancement in understanding the pathophysiology of DCI, we are hopeful that we will make strides toward better outcomes for this unique patient population.

Spontaneous subarachnoid haemorrhage

Subarachnoid haemorrhage (SAH) is the third most common subtype of stroke. Incidence has decreased over past decades, possibly in part related to lifestyle changes such as smoking cessation and management of hypertension. Approximately a quarter of patients with SAH die before hospital admission; overall outcomes are improved in those admitted to hospital, but with elevated risk of long-term neuropsychiatric sequelae such as depression. The disease continues to have a major public health impact as the mean age of onset is in the mid-fifties, leading to many years of reduced quality of life. The clinical presentation varies, but severe, sudden onset of headache is the most common symptom, variably associated with meningismus, transient or prolonged unconsciousness, and focal neurological deficits including cranial nerve palsies and paresis. Diagnosis is made by CT scan of the head possibly followed by lumbar puncture. Aneurysms are commonly the underlying vascular cause of spontaneous SAH and are diagnosed by angiography. Emergent therapeutic interventions are focused on decreasing the risk of rebleeding (ie, preventing hypertension and correcting coagulopathies) and, most crucially, early aneurysm treatment using coil embolisation or clipping. Management of the disease is best delivered in specialised intensive care units and high-volume centres by a multidisciplinary team. Increasingly, early brain injury presenting as global cerebral oedema is recognised as a potential treatment target but, currently, disease management is largely focused on addressing secondary complications such as hydrocephalus, delayed cerebral ischaemia related to microvascular dysfunction and large vessel vasospasm, and medical complications such as stunned myocardium and hospital acquired infections.

Dynamic Measurements of Cerebral Blood Flow Responses to Cortical Spreading Depolarization in the Murine Endovascular Perforation Subarachnoid Hemorrhage Model

Delayed cerebral ischemia (DCI) is the most severe complication after subarachnoid hemorrhage (SAH), and cortical spreading depolarization (CSD) is believed to play a vital role in it. However, the dynamic changes in cerebral blood flow (CBF) in response to CSD in typical SAH models have not been well investigated. Here, SAH was established in mice with endovascular perforation. Subsequently, the spontaneous CBF dropped instantly and then returned to baseline rapidly. After KCl application to the cortex, subsequent hypoperfusion waves occurred across the groups, while a lower average perfusion level was found in the SAH groups (days 1-7). Moreover, in the SAH groups, the number of CSD decreased within day 7, and the duration and spreading velocity of the CSD increased within day 3 and day 14, respectively. Next, we continuously monitored the local field potential (LFP) in the prefrontal cortex. The results showed that the decrease in the percentage of gamma oscillations lasted throughout the whole process in the SAH group. In the chronic phase after SAH, we found that the mice still had cognitive deficits but experienced no obvious tissue damage. In summary, SAH negatively affects the CBF responses to CSD and the spontaneous LFP activity and causes long-term cognitive deficits in mice. Based on these findings, in the specific phase after SAH, DCI is induced or exacerbated more easily by potential causers of CSD in clinical practice (edema, erythrocytolysis, inflammation), which may lead to neurological deterioration.

Neutrophil Extracellular Traps and Delayed Cerebral Ischemia in Aneurysmal Subarachnoid Hemorrhage

IMPORTANCE

Myeloperoxidase (MPO)-DNA complexes, biomarkers of neutrophil extracellular traps (NETs), have been associated with arterial and venous thrombosis. Their role in aneurysmal subarachnoid hemorrhage (aSAH) is unknown.

OBJECTIVES

To assess whether serum MPO-DNA complexes are present in patients with aSAH and whether they are associated with delayed cerebral ischemia (DCI).

DESIGN SETTING AND PARTICIPANTS

Post-hoc analysis of a prospective, observational single-center study, with de novo serum biomarker measurements in consecutive patients with aSAH between July 2018 and September 2020, admitted to a tertiary care neuroscience ICU.

MAIN OUTCOMES AND MEASURES

We analyzed serum obtained at admission and hospital day 4 for concentrations of MPO-DNA complexes. The primary outcome was DCI, defined as new infarction on brain CT. The secondary outcome was clinical vasospasm, a composite of clinical and transcranial Doppler parameters. We used Wilcoxon signed-rank-test to assess for differences between paired measures.

RESULTS

Among 100 patients with spontaneous subarachnoid hemorrhage, mean age 59 years (sd ± 13 yr), 55% women, 78 had confirmed aSAH. Among these, 29 (37%) developed DCI. MPO-DNA complexes were detected in all samples. The median MPO-DNA level was 33 ng/mL (interquartile range [IQR], 18-43 ng/mL) at admission, and 22 ng/mL (IQR, 11-31 ng/mL) on day 4 (unpaired test; p = 0.015). We found a significant reduction in MPO-DNA levels from admission to day 4 in patients with DCI (paired test; p = 0.036) but not in those without DCI (p = 0.17). There was a similar reduction in MPO-DNA levels between admission and day 4 in patients with (p = 0.006) but not in those without clinical vasospasm (p = 0.47).

CONCLUSIONS AND RELEVANCE

This is the first study to detect the NET biomarkers MPO-DNA complexes in peripheral serum of patients with aSAH and to associate them with DCI. A pronounced reduction in MPO-DNA levels might serve as an early marker of DCI. This diagnostic potential of MPO-DNA complexes and their role as potential therapeutic targets in aSAH should be explored further.

Individualized cerebral perfusion pressure in acute neurological injury: are we ready for clinical use?

PURPOSE OF REVIEW

Individualizing cerebral perfusion pressure based on cerebrovascular autoregulation assessment is a promising concept for neurological injuries where autoregulation is typically impaired. The purpose of this review is to describe the status quo of autoregulation-guided protocols and discuss steps towards clinical use.

RECENT FINDINGS

Retrospective studies have indicated an association of impaired autoregulation and poor clinical outcome in traumatic brain injury (TBI), hypoxic-ischemic brain injury (HIBI) and aneurysmal subarachnoid hemorrhage (aSAH). The feasibility and safety to target a cerebral perfusion pressure optimal for cerebral autoregulation (CPPopt) after TBI was recently assessed by the COGITATE trial. Similarly, the feasibility to calculate a MAP target (MAPopt) based on near-infrared spectroscopy was demonstrated for HIBI. Failure to meet CPPopt is associated with the occurrence of delayed cerebral ischemia in aSAH but interventional trials in this population are lacking. No level I evidence is available on potential effects of autoregulation-guided protocols on clinical outcomes.

SUMMARY

The effect of autoregulation-guided management on patient outcomes must still be demonstrated in prospective, randomized, controlled trials. Selection of disease-specific protocols and endpoints may serve to evaluate the overall benefit from such approaches.

Basic Surveillance Parameters Improve the Prediction of Delayed Cerebral Infarction After Aneurysmal Subarachnoid Hemorrhage

Background

To establish a practical risk chart for prediction of delayed cerebral infarction (DCI) after aneurysmal subarachnoid hemorrhage (aSAH) by using information that is available until day 5 after ictus.

Methods

We assessed all consecutive patients with aSAH admitted to our service between September 2008 and September 2015 (n = 417). The data set was randomly split into thirds. Two-thirds were used for model development and one-third was used for validation. Characteristics that were present between the bleeding event and day 5 (i.e., prior to >95% of DCI diagnoses) were assessed to predict DCI by using logistic regression models. A simple risk chart was established and validated.

Results

The amount of cisternal and ventricular blood on admission CT (Hijdra sum score), early sonographic vasospasm (i.e., mean flow velocity of either intracranial artery >160 cm/s until day 5), and a simplified binary level of consciousness score until day 5 were the strongest predictors of DCI. A model combining these predictors delivered a high predictive accuracy [the area under the receiver operating characteristic (AUC) curve of 0.82, Nagelkerke's R² 0.34 in the development cohort]. Validation of the model demonstrated a high discriminative capacity with the AUC of 0.82, Nagelkerke's R² 0.30 in the validation cohort.

Conclusion

Adding level of consciousness and sonographic vasospasm between admission and postbleed day 5 to the initial blood amount allows for simple and precise prediction of DCI. The suggested risk chart may prove useful for selection of appropriate candidates for interventions to prevent DCI.

External Validation of a Neural Network Model in Aneurysmal Subarachnoid Hemorrhage: A Comparison With Conventional Logistic Regression Models

BACKGROUND

Interest in machine learning (ML)-based predictive modeling has led to the development of models predicting outcomes after aneurysmal subarachnoid hemorrhage (aSAH), including the Nijmegen acute subarachnoid hemorrhage calculator (Nutshell). Generalizability of such models to external data remains unclear.

OBJECTIVE

To externally validate the performance of the Nutshell tool while comparing it with the conventional Subarachnoid Hemorrhage International Trialists (SAHIT) models and to review the ML literature on outcome prediction after aSAH and aneurysm treatment.

METHODS

A prospectively maintained database of patients with aSAH presenting consecutively to our institution in the 2013 to 2018 period was used. The web-based Nutshell and SAHIT calculators were used to derive the risks of poor long-term (12-18 months) outcomes and 30-day mortality. Discrimination was evaluated using the area under the curve (AUC), and calibration was investigated using calibration plots. The literature on relevant ML models was surveyed for a synopsis.

RESULTS

In 269 patients with aSAH, the SAHIT models outperformed the Nutshell tool (AUC: 0.786 vs 0.689, P = .025) in predicting long-term functional outcomes. A logistic regression model of the Nutshell variables derived from our data achieved adequate discrimination (AUC = 0.759) of poor outcomes. The SAHIT models outperformed the Nutshell tool in predicting 30-day mortality (AUC: 0.810 vs 0.636, P < .001). Calibration properties were more favorable for the SAHIT models. Most published aneurysm-related ML-based outcome models lack external validation and usable testing platforms.

CONCLUSION

The Nutshell tool demonstrated limited performance on external validation in comparison with the SAHIT models. External validation and the dissemination of testing platforms for ML models must be emphasized.

Multimodality Monitoring for Delayed Cerebral Ischemia in Subarachnoid Hemorrhage: A Mini Review

Aneurysmal subarachnoid hemorrhage is a disease with high mortality and morbidity due in large part to delayed effects of the hemorrhage, including vasospasm, and delayed cerebral ischemia. These two are now recognized as overlapping yet distinct entities, and supportive therapies for delayed cerebral ischemia are predicated on identifying DCI as quickly as possible. The purpose of this overview is to highlight diagnostic tools that are being used in the identification of DCI in the neurocritical care settings.

Easily Created Prediction Model Using Automated Artificial Intelligence Framework (Prediction One, Sony Network Communications Inc., Tokyo, Japan) for Subarachnoid Hemorrhage Outcomes Treated by Coiling and Delayed Cerebral Ischemia

Introduction Reliable prediction models of subarachnoid hemorrhage (SAH) outcomes and delayed cerebral ischemia (DCI) are needed to decide the treatment strategy. Automated artificial intelligence (AutoAI) is attractive, but there are few reports on AutoAI-based models for SAH functional outcomes and DCI. We herein made models using an AutoAI framework, Prediction One (Sony Network Communications Inc., Tokyo, Japan), and compared it to other previous statistical prediction scores. Methods We used an open dataset of 298 SAH patients, who were with non-severe neurological grade and treated by coiling. Modified Rankin Scale 0-3 at six months was defined as a favorable functional outcome and DCI occurrence as another outcome. We randomly divided them into a 248-patient training dataset and a 50-patient test dataset. Prediction One made the model using training dataset with 5-fold cross-validation. We evaluated the model using the test dataset and compared the area under the curves (AUCs) of the created models. Those of the modified SAFIRE score and the Fisher computed tomography (CT) scale to predict the outcomes. Results The AUCs of the AutoAI-based models for functional outcome in the training and test dataset were 0.994 and 0.801, and those for the DCI occurrence were 0.969 and 0.650. AUCs for functional outcome calculated using modified SAFIRE score were 0.844 and 0.892. Those for the DCI occurrence calculated using the Fisher CT scale were 0.577 and 0.544. Conclusions We easily and quickly made AutoAI-based prediction models. The models' AUCs were not inferior to the previous prediction models despite the easiness.