The 2nd E-Prevention Challenge: Psychotic and Non-Psychotic Relapse Detection Using Wearable-Based Digital Phenotyping

Panagiotis Paraskevas Filntisis, Niki Efthymiou, George Retsinas, Athanasia Zlatintsi, Christos Garoufis, Thomas Sounapoglou, Panayotis Tsanakas, Nikolaos Smyrnis, Petros Maragos

Abstract: The 2nd e-Prevention challenge1 aims to foster innovative research in the prediction and identification of mental health relapses by an-alyzing and processing the digital phenotype of patients within the psychotic spectrum. The challenge offers participants access to extensive, continuous recordings of raw biosignals from wearable sensors such as accelerometers, gyroscopes, and heart rate monitors integrated into smartwatches. Participants are tasked with utilizing the provided data to develop digital phenotypes that accurately capture behavioral patterns and traits in patients. The evaluation focus on two main objectives: first, the detection of non-psychotic relapses, and second, the detection of psychotic relapses.