A new publication from the AI-PROGNOSIS project has been published in the proceedings of the 2025 IEEE International Conference on E-health Networking, Application & Services (HealthCom).

The study presents a deep learning–based framework for detecting Parkinsonian tremor and estimating tremor amplitude using data from wrist-worn wearable devices. The proposed approach combines machine learning techniques to identify tremor patterns in accelerometer data and model how tremor evolves over time. Tremor amplitude is estimated only when tremor is detected, supporting more focused and objective tremor assessment.

Read the full publication: https://zenodo.org/records/18469204