AI-based predictive biomarkers for chronic neurological diseases: the rAIdD prospective, multicenter, observational study protocol.

Research overview

The text below is the source article’s original abstract under a verified reuse license. cc by

BackgroundChronic neurological disorders such as Multiple Sclerosis (MS), Parkinson's disease (PD), and Alzheimer's Disease (AD) represent a major global health burden characterized by progressive neurodegeneration, functional disability, and cognitive decline. Despite differences in etiology and clinical presentation, these conditions share multifactorial pathophysiological mechanisms influenced by genetic, environmental, and lifestyle-related factors. Advances in artificial intelligence (AI), wearable technologies, and multimodal clinical data integration offer new opportunities for identifying predictive digital biomarkers and improving personalized disease management. The rAIdD project ("eHealth Network: AI and new ICT technology equipment for digital diagnosis") aims to develop an interoperable digital infrastructure to support early diagnosis, monitoring, and risk stratification in chronic neurological diseases. This study protocol describes the neurological component of the rAIdD network focusing on MS, PD, and AD.Methods and analysisThis prospective, multicenter, observational study involves six Italian academic and clinical centers and will enroll 780 participants: 300 MS, 150 PD, 150 AD, and 180 healthy controls. Participants will be followed for 18 months within a 48-month study period. Standardized clinical, neuropsychological, neuroimaging, and digital assessments will be performed at baseline and at 6-, 12-, and 18-month follow-ups. Clinical evaluation includes disease-specific disability and functional scales, mood and quality-of-life assessments, and lifestyle and environmental risk factor profiling. Continuous digital monitoring will be conducted using wearable sensors to collect biometric and behavioral data, including physical activity, sleep patterns, and cardiovascular parameters. Structural neuroimaging will be acquired longitudinally and integrated with clinical and digital data through a centralized web-based electronic data capture platform. Machine learning approaches will be applied to identify multimodal predictive biomarkers and model disease progression patterns across disorders.Ethics and disseminationThe study has been approved by the Ethics Committee of the coordinating center and by local ethics committees of all participating institutions. Written informed consent is obtained from all participants in accordance with the Declaration of Helsinki and the General Data Protection Regulation (GDPR 2016/679). Results will be disseminated through peer-reviewed publications, scientific conferences, and digital communication platforms to support knowledge translation and implementation of precision neurology approaches.

Related topics