Clustering Approaches for Gait Analysis within Neurological Disorders: A Narrative Review.

Area of Science:

• Neurology
• Biomedical Engineering
• Data Science

Background:

• Increasing prevalence of neurological disorders necessitates objective gait analysis for deficit identification.
• Existing gait analysis technologies are often impractical for clinical use due to cost, time, and processing limitations.

Purpose of the Study:

• To evaluate existing patient clustering techniques for neurological disorders to aid clinical treatment optimization.
• To identify gaps and suggest advancements for improving the clinical utility of gait analysis.

Main Methods:

• Conducted a narrative review of thirteen relevant studies on patient clustering for neurological disorders.
• Characterized and evaluated methods against seven criteria, summarizing results in comprehensive tables.

Main Results:

• Only three reviewed approaches demonstrated medium or high process maturity.
• Merely two approaches exhibited high clinical applicability, indicating a need for improvement.

Conclusions:

• Advancements are required in markerless optical tracking, experimental design optimization, and external validation.
• Enhanced clustering techniques can improve the identification of patient disparities, ultimately benefiting patient outcomes.