Risk models to predict late-onset seizures after stroke: A systematic review

Large datasets from Electronic Health Records predict seizures after ischemic strokes: A Machine Learning approach

Objective

To develop an artificial intelligence, machine learning prediction model for estimating the risk of seizures 1 year and 5 years after ischemic stroke (IS) using a large dataset from Electronic Health Records.

Background

Seizures are frequent after ischemic strokes and are associated with increased mortality, poor functional outcomes, and lower quality of life. Separating patients at high risk of seizures from those at low risk of seizures is needed for treatment and clinical trial planning, but remains challenging. Machine learning (ML) is a potential approach to solve this paradigm.

Design/Methods

We identified patients (aged ≥18 years) with IS without a prior diagnosis of seizures from 2015 until inception (08/09/22) in the TriNetX Research Network, using the International Classification of Diseases, Tenth Revision (ICD-10) I63, excluding I63.6 (venous infarction). The outcome of interest was any ICD-10 diagnosis of seizures (G40/G41) at 1 year and 5 years following the index IS. We applied a conventional logistic regression and a Light Gradient Boosted Machine algorithm to predict the risk of seizures at 1 year and 5 years. The performance of the model was assessed using the area under the receiver operating characteristics (AUROC), the area under the precision-recall curve (AUPRC), F1 statistic, model accuracy, balanced accuracy, precision, and recall, with and without anti-seizure medication use in the models.

Results

Our study cohort included 430,254 IS patients. Seizures were present in 18,502 (4.3%) and (5.3%) patients within 1 and 5 years after IS, respectively. At 1-year, the AUROC, AUPRC, F1 statistic, accuracy, balanced-accuracy, precision, and recall were respectively 0.7854 (standard error: 0.0038), 0.2426 (0.0048), 0.2299 (0.0034), 0.8236 (0.001), 0.7226 (0.0049), 0.1415 (0.0021), and 0.6122, (0.0095). Corresponding metrics at 5 years were 0.7607 (0.0031), 0.247 (0.0064), 0.2441 (0.0032), 0.8125 (0.0013), 0.7001 (0.0045), 0.155 (0.002) and 0.5745 (0.0095).

Conclusion

Our findings suggest that ML models show good model performance for predicting seizures after IS.

[Algorithms for the diagnosis and treatment of cognitive impairment and dysphagia in stroke patients]

Stroke is a socially significant neurological disease, the second most common cause of disability and mortality. A wide range of neurological problems that occur after stroke: cognitive, motor, speech, and language disfunction, neuropsychiatric, swallowing disorders and others, complicate rehabilitation, impair social and everyday adaptation, and reduce the quality of life of patients and their caregivers. Cognitive impairment (CI) is one of the most significant and common complications of stroke. Stroke increases the risk of their development by 5-8 times. Dysphagia is also a common symptom of stroke, the cause of aspiration complications (pneumonia), and nutritional imbalance. It increases the possibility of developing CI and dementia, and contributes to an increase in mortality. Older adults with CI are at a higher risk of developing dysphagia, therefore the early symptoms of dysphagia (presbyphagia) should be diagnosed. In recent years, the connection between CI and dysphagia has been actively studied. It is extremely important to identify CI and swallowing disorders as early as possible in patients both before and at all stages after stroke; as well as to develop combined multidisciplinary protocols for the rehabilitation of patients with these disorders with pharmacological support for the process.

Risk Factors Associated with Epilepsy Related to Cerebrovascular Disease: A Systematic Review and Meta-Analysis

Background and Objective

Stroke is one of the most frequent neurological syndromes in the adult population and the cause of 10% of all diagnosed epilepsies. It is attributed to the origin of up to 50% of them in adults >60 years of age. Although a few risk factors have been described and considered when modeling predictive tools, this aspect is still clinically complex. The objective of this study is to describe and compare predictor scales of post stroke epilepsy (PSE) in adult patients with better performance.

Methods

A systematic review and meta-analysis were performed of studies published between 2010 and 2020 and found in PubMed, Scopus, EMBASE, LILACS, BVS, Google Scholar, and CENTRAL databases. Sixteen studies were included with a total of 298,694 patients with a diagnosis of stroke, 5590 presented late seizures (LS).

Results

Hemorrhage, cortical involvement, and early seizure were the elements most associated with the risk of presenting late seizures. The SeLECT score demonstrated a low risk of bias with a high predictive ability in patients with ischemic stroke (AUC: 0.77 [95% CI: 0.71-0.82]). In patients with hemorrhagic stroke, the CAVE score demonstrated adequate predictive ability (AUC: 0.81 [95% CI: 0.76-0.86]), but an uncertain risk of bias. Research has established risk factors for post ictal epilepsy; however, the numerous ways of assessing data in studies and the difference in their designs make the task of producing a predictive scale that covers the most important risk factors and is reliable for application in the clinical setting, regardless of stroke etiology, very arduous.

Conclusion

Hemorrhage, cortical involvement, and early seizure are associated with an increased risk of post ictal epilepsy. Also, elements such as age, traditional vascular risk factors, and functional assessment failed to reflect statistical significance. Finally, further research is required to refine the available predictive tools.

Machine Learning and Artificial Intelligence Applications to Epilepsy: a Review for the Practicing Epileptologist

PURPOSE OF REVIEW

Machine Learning (ML) and Artificial Intelligence (AI) are data-driven techniques to translate raw data into applicable and interpretable insights that can assist in clinical decision making. Some of these tools have extremely promising initial results, earning both great excitement and creating hype. This non-technical article reviews recent developments in ML/AI in epilepsy to assist the current practicing epileptologist in understanding both the benefits and limitations of integrating ML/AI tools into their clinical practice.

RECENT FINDINGS

ML/AI tools have been developed to assist clinicians in almost every clinical decision including (1) predicting future epilepsy in people at risk, (2) detecting and monitoring for seizures, (3) differentiating epilepsy from mimics, (4) using data to improve neuroanatomic localization and lateralization, and (5) tracking and predicting response to medical and surgical treatments. We also discuss practical, ethical, and equity considerations in the development and application of ML/AI tools including chatbots based on Large Language Models (e.g., ChatGPT). ML/AI tools will change how clinical medicine is practiced, but, with rare exceptions, the transferability to other centers, effectiveness, and safety of these approaches have not yet been established rigorously. In the future, ML/AI will not replace epileptologists, but epileptologists with ML/AI will replace epileptologists without ML/AI.

Post-stroke Everything

PURPOSE OF REVIEW

This review aims at providing updates on selected post-stroke complications. We examined recent advances in diagnosing and treating the following post-stroke complications: cognitive impairment, epilepsy, depression, fatigue, tremors, dysphagia, and pain.

RECENT FINDINGS

Advances in understanding the mechanisms of post-stroke complications, in general, are needed despite advances made in understanding, treating, and preventing these complications. There are growing progresses in integrating new tools to diagnose post-stroke cognitive impairment. The potential role of acute stroke reperfusion treatment in post-stroke epilepsy and its impact on other stroke complications is getting more transparent. Post-stroke depression remains underestimated and new tools to diagnose depression after stroke are being developed. New promising pharmacological approaches to treating post-stroke pain are emerging. Tremors related to stroke are poorly understood and under-evaluated, while treatment towards post-stroke dysphagia has benefited from new non-pharmacological to pharmacological approaches.

CONCLUSIONS

An integrative approach to stroke complications and collaborations between providers across specialties are more likely to improve stroke outcomes.

Inflammation Mediated Epileptogenesis as Possible Mechanism Underlying Ischemic Post-stroke Epilepsy

Post-stroke Epilepsy (PSE) is one of the most common forms of acquired epilepsy, especially in the elderly population. As people get increasingly older, the number of stroke patients is expected to rise and concomitantly the number of people with PSE. Although many patients are affected by post-ischemic epileptogenesis, not much is known about the underlying pathomechanisms resulting in the development of chronic seizures. A common hypothesis is that persistent neuroinflammation and glial scar formation cause aberrant neuronal firing. Here, we summarize the clinical features of PSE and describe in detail the inflammatory changes after an ischemic stroke as well as the chronic changes reported in epilepsy. Moreover, we discuss alterations and disturbances in blood-brain-barrier leakage, astrogliosis, and extracellular matrix changes in both, stroke and epilepsy. In the end, we provide an overview of commonalities of inflammatory reactions and cellular processes in the post-ischemic environment and epileptic brain and discuss how these research questions should be addressed in the future.