Is Mild Really Mild?: Generating Longitudinal Profiles of Stroke Survivor Impairment and Impact Using Unsupervised Machine Learning

Abstract

The National Institute of Health Stroke Scale (NIHSS) is used worldwide to classify stroke severity as ‘mild’, ‘moderate’, or ‘severe’ based on neurological impairment. Yet, stroke survivors argue that the classification of ‘mild’ does not represent the holistic experience and impact of stroke on their daily lives. In this observational cohort study, we aimed to identify different types of impairment profiles among stroke survivors classified as ‘mild’. We used survivors of mild stroke’ data from the START longitudinal stroke cohort (n = 73), with measures related to sensorimotor, cognition, depression, functional disability, physical activity, work, and social adjustment over 12 months. Given the multisource, multigranular, and unlabeled nature of the data, we utilized a structure-adapting, unsupervised machine learning approach, the growing self-organizing map (GSOM) algorithm, to generate distinct clinical profiles. These diverse impairment profiles revealed that survivors of mild stroke experience varying degrees of impairment and impact (cognitive, depression, physical activity, work/social adjustment) at different time points, despite the uniformity implied by their NIHSS-classified ‘mild’ stroke. This emphasizes the necessity of creating a holistic and more comprehensive representation of survivors of mild stroke’ needs over the first year after stroke to improve rehabilitation and poststroke care.

Keywords

• mild stroke
• artificial intelligence
• patient profiling
• unsupervised learning
• personalized healthcare