Machine learning and wearable sensors in early cognitive change detection in CADASIL

Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy (CADASIL) is a hereditary small vessel disease that often leads to progressive cognitive decline. Early detection of cognitive changes remains a clinical challenge, as conventional neuroimaging and neuropsychological assessments may lack sensitivity in the prodromal phase. Recent advances in machine learning and wearable sensor technologies offer promising possibilities for noninvasive, real-time monitoring of subtle cognitive shifts. By analyzing multimodal data, such as gait patterns, speech dynamics, sleep metrics, and physiological signals, machine learning algorithms can identify predictive markers of early cognitive impairment in CADASIL patients. This approach enables continuous, personalized assessment outside clinical settings, potentially transforming disease monitoring and intervention timing. We propose integrating wearable-based digital phenotyping with artificial intelligence-driven analytics to enhance early detection, risk stratification, and longitudinal tracking in CADASIL care.