ALAMEDA Data: Bridging the Early Diagnosis and Treatment Gaps of Brain Diseases (Parkinson's Disease, Multiple Slerosis and Stroke)

ALAMEDA is an Horizon 2020 Research and Innovation project that aims to bridge the early diagnosis and treatment gap of brain diseases via smart, connected, proactive and evidence-based technological interventions. Its vision is to research and prototype new generation Artificial Intelligence (AI) systems to support brain disorders patients' healthcare, focusing on Parkinson's Disease (PD), Multiple Sclerosis (MS) and Stroke.

To this end, three (one for each disease) small scale validation pilots were performed in real world settings. Throughout these pilots, various types of data, such as accelerometer, gyroscopic, heart rate, etc., were collected via smart wearable sensors from the patients enrolled. The smart devices that were employed include: a Fitbit smartwatch, a GENEActiv smart bracelet, Novel Loadsol insole sensors and a prototype smart belt with triaxial accelerometers and gyroscopes embedded. Moreover, the patients underwent several clinical assessments and filled in numerous both disease-specific and non-disease-specific questionnaires.

In this record, both raw and processed sensory data are combined with both clinical and patient reported outcomes (PROs) to form different disease-specific datasets. More specifically:

For Parkinson's disease: Three datasets are provided (one for tremor detection, one for dyskinesia detection, and one for Hoehn & Yahr score estimation) alongside the vertical ground reaction force recordings.

For Multiple Sclerosis: Two datasets are provided (one for Expanded Disability Status Scale (EDSS) scores classification and one that accumulates clinical data and individual scores from various MS-related questionnaires) alongside the vertical ground reaction force and the smart belt recordings.

For Stroke: Two datasets are provided (one for rehabilitation exercises' recognition and one for walking classification, both with and without manual annotations) alongside the smart belt recordings.

More information about the datasets provided can be found in the respective READ ME files that are included in the current record.