WP1: Patient monitoring and support system
WP1 is concerned with developing a patient monitoring and support system consisting of collecting data from the patient for modelling behaviour including emotional state to be applied for prediction of these to reduce mental health challenges. The developed technology will provide a support system for patients and clinicians by recommending appropriate possible actions and treatments.
(1) Research what specific features are being significant for human behaviour and mental health.
(2) Research and prototype a flexible system for modelling and prediction of behaviour.
(3) Research and develop a support system for effective mental health improvement.
Methods and approaches
Within the work package, we follow an approach of applied based research using rapid prototyping and patient oriented evaluation. We focus on a wide range of machine learning approaches combining methods such as Random Forest or Regression with advanced Deep Learning methods (LSTM, GRU, RNN, CNN). Most of the data that has to be analysed within the INTROMAT cases (disorders) are time series related multimodal data.
WP 1 milestones
Enrique Garcia-Ceja and Michael Riegler and Petter Jakobsen and Jim Tørresen and Tine Nordgreen and Ketil J. Oedegaard and Ole Bernt Fasmer, Depresjon: A Motor Activity Database of Depression Episodes in Unipolar and Bipolar Patients, Proceedings of the 9th ACM on Multimedia Systems Conference, 2018.
Enrique Garcia-Ceja and Michael Riegler and Petter Jakobsen and Jim Tørresen and Tine Nordgreen and Ketil J. Oedegaard and Ole Bernt Fasmer, Motor Activity Based Classification of Depression in Unipolar and Bipolar Patients, Proceedings of the 31st IEEE CBMS International Symposium on Computer-Based Medical Systems, 2018.
Open source contributions
Depresjon: A Motor Activity Database of Depression Episodes in Unipolar and Bipolar Patients. Datasetlink