PreCare ML Project

Project Aims

This project has three main aims. First, we aim to validate and improve our ML models across different hospital networks and populations. Second, we aim to integrate ML models in different hospital information systems and evaluate their impact on daily hospital routine. Third, building upon these validated models, we aim to address effective risk communication strategies in order to effect behavioural changes in patients. Therefore, our project makes a fundamental contribution towards employing innovative personalized risk prediction and to assess its clinical implementation in a transnational context.

Aim-1

Validation and Improvement of ML models across different hospital networks

WP-1 Data management & quality
WP-2 Model validation & adaptation, different cohorts
WP-3 Federated learning for model improvement

Aim-2

Integration of ML models in HIS and evaluation in daily hospital routine

WP-4 Integration of best performing models in HIS
WP-5 Evaluation of the models in prospective clinical settings

Aim-3

Effective risk communication and fair predictions

WP-6 Evaluation of effective/personalised risk communication strategies
WP-7 Personalised health services
WP-8 Evaluation of biases in machine learning models