Can Big Data guide prognosis and clinical decisions in epilepsy?

Abstract

Big Data is no longer a novel concept in health care. Its promise of positive impact is not only undiminished, but daily enhanced by seemingly endless possibilities. Epilepsy is a disorder with wide heterogeneity in both clinical and research domains, and thus lends itself to Big Data concepts and techniques. It is therefore inevitable that Big Data will enable multimodal research, integrating various aspects of “‐omics” domains, such as phenome, genome, microbiome, metabolome, and proteome. This scope and granularity have the potential to change our understanding of prognosis and mortality in epilepsy. The scale of new discovery is unprecedented due to the possibilities promised by advances in machine learning, in particular deep learning. The subsequent possibilities of personalized patient care through clinical decision support systems that are evidence‐based, adaptive, and iterative seem to be within reach. A major objective is not only to inform decision‐making, but also to reduce uncertainty in outcomes. Although the adoption of electronic health record (EHR) systems is near universal in the United States, for example, advanced clinical decision support in or ancillary to EHRs remains sporadic. In this review, we discuss the role of Big Data in the development of clinical decision support systems for epilepsy care, prognostication, and discovery.

A multimodal clinical data resource for personalized risk assessment of sudden unexpected death in epilepsy

“…It may be difficult, time-consuming, and costly to clean and reorganize data, and studies' conclusions may be affected by hidden biases. Only 20% of data scientists' time is spent building models, analyzing, visualizing, and analyzing the data, while most of their time (80%) is spent cleaning and preparing data (Patel, 2019 ; Li et al, 2021 ). However, having high-quality data is not sufficient to say the system is data-driven.…”

“…As a final point, and very importantly, the data should be queryable and tools should be developed so that data can be sliced and diced. In order to perform assessments, large amounts of raw data should be filtered, grouped, and aggregated into smaller sets of higher-level and analysis-ready data that can assist clinicians and researchers in gaining insights into topics of interest (Li et al, 2021 ). Ensuring the credibility of the data and improving the knowledge base while maintaining accessibility are key issues for an effective personalized risk assessment.…”

“…With the rapid development of big data in medicine, the development of models for personalized risk assessment based on machine learning and artificial intelligence is now one of the hottest topics in the research field. However, most existing models and methods were built and evaluated on a small and single dataset, which makes it difficult to obtain a generally applicable and robust model (i.e., stable detection or prediction performance for different patients; Bernardi et al, 2016;Li et al, 2019Li et al, , 2021.…”

Economic aspects of treating seizure clusters

“…"Big data" techniques, including machine learning and natural language processing, may prove useful in analyzing ever-growing volumes of heterogeneous datasets to fill at least some of these gaps as well. 2 The objective of this article is to outline the various methods of pharmacoeconomic analysis of seizure clusters, beginning with discussion of data on the chance that a cluster will progress to an adverse outcome, such as a need for emergency care, the costs of such an outcome, the cost of a rescue medication (RM), and the effectiveness of the RM. Indirect costs, such as lost employment for patients and caregivers, must also be considered.…”

“…Data are available for some, but not all, of these components needed for economic analysis of seizure clusters and their treatment; in some cases, when data relating to seizure clusters are unavailable, data for epilepsy or seizures in general may serve as a surrogate until such gaps in data can be addressed. “Big data” techniques, including machine learning and natural language processing, may prove useful in analyzing ever‐growing volumes of heterogeneous datasets to fill at least some of these gaps as well 2 . The objective of this article is to outline the various methods of pharmacoeconomic analysis of seizure clusters, beginning with discussion of data on the costs of illness and RM therapy.…”