The BRAINTEASER Datasets: Clinical, Wearable and Environmental Data for ALS & MS Progression Modeling

Area of Science:

• Neuroscience
• Medical Informatics
• Artificial Intelligence

Background:

• Amyotrophic lateral sclerosis (ALS) and multiple sclerosis (MS) are progressive neurological diseases.
• Predictive modeling for ALS and MS progression is crucial for patient care but limited by data availability.
• Artificial Intelligence (AI) holds potential for improving disease progression modeling.

Purpose of the Study:

• To curate and validate comprehensive datasets for AI-driven disease progression modeling in ALS and MS.
• To address the data scarcity challenge hindering the development of predictive tools.
• To support the creation of AI models for personalized patient care and clinical decision-making.

Main Methods:

• Curated four datasets from the H2020 BRAINTEASER project, including clinical, environmental, and wearable data.
• Collected data from 2,290 ALS patients and 723 MS patients from real-world clinical practice.
• Validated datasets through three editions of the intelligent Disease Progression Prediction challenges at CLEF, alongside automated and manual quality checks.

Main Results:

• Established large, clinically relevant datasets for ALS and MS patient progression.
• Datasets encompass diverse data types, offering a realistic representation for AI model training.
• Community validation through challenges ensures dataset quality and utility for predictive modeling.

Conclusions:

• The BRAINTEASER datasets provide a valuable resource for advancing AI in neurological disease research.
• These datasets facilitate the development and validation of AI tools for predicting ALS and MS progression.
• Improved predictive tools can enhance patient outcomes and support clinical management of these debilitating conditions.