Using predictive analytics to identify drug-resistant epilepsy patients

Abstract

Epilepsy is one of the most common brain disorders that greatly affects patients’ quality of life and poses serious risks to their health. While the majority of the patients positively respond to the existing anti-epilepsy drugs, others who developed the refractory type of epilepsy show resistance against drug therapy and need to undergo advance treatments such as surgery. Given that identifying such patients is not a straightforward process and requires long courses of trial and error with anti-epilepsy drugs, this study aims at predicting those at-risk patients using clinical and demographic data obtained from electronic medical records. Specifically, the study employs several predictive analytics machine-learning methods, equipped with a novel approach for data balancing, to identify drug-resistant patients using their comorbidities and demographic information along with the initial epilepsy-related diagnosis made by their physician. The promising results we obtained highlight the potential use of machine-learning techniques in facilitating medical decisions and suggest the possibility of extending the proposed approach for developing a clinical decision support system for medical professionals.

New era of personalised epilepsy management

“…Recent studies have used drug dispensing databases to develop models to predict drug treatment responses. 14 15 Although these are large datasets, the models do not capture detailed information about the individual or the disease and therefore lack potentially important data on treatment outcomes. Medical records, on the other hand, include comprehensive clinical information on epilepsy management and are a fuller repository of factors potentially linked to treatment outcomes.…”

Precision Medicine: Academic dreaming or clinical reality?

“…Application of validated nomenclature systems for epilepsy can yield cohorts in the tens of thousands with varying degrees of granular clinical data. 34 Use of these data alone has yielded important insights into premature mortality 35 and potential interventions to reduce risk, 36 bidirectional relationships between psychiatric disease and epilepsy, 34,37 and prediction models for individualized risks of psychiatric adverse effects from levetiracetam 19 and drug-resistant epilepsy 38 .…”

Application of transformers for predicting epilepsy treatment response

“…There has also been a recent paper that employed a manually tuned mathematical model to determine which ASM to use. 8 A few studies have also looked at developing individualized prediction models for early diagnosis of DRE 14,15 and for selecting the most appropriate ASM. 6,14,15 In recent advances of deep learning, attention-based models have shown promise in various applications such as NLP, image recognition, and even electronic health records (EHR) data.…”

“…8 A few studies have also looked at developing individualized prediction models for early diagnosis of DRE 14,15 and for selecting the most appropriate ASM. 6,14,15 In recent advances of deep learning, attention-based models have shown promise in various applications such as NLP, image recognition, and even electronic health records (EHR) data. 4,10,16 Attention-based models is a method in deep learning that allows the network to break down complex inputs into smaller parts for a sequence of data.…”

“…There has also been a recent paper that employed a manually tuned mathematical model to determine which ASM to use. 8 A few studies have also looked at developing individualized prediction models for early diagnosis of DRE 14, 15 and for selecting the most appropriate ASM. 6, 14, 15…”