Result description
Predictive models are implemented for the personalised rehabilitation stage within the PRECISE4Q project. The three dedicated models are i) the cognitive improvement model, ii) the cognitive therapy compliance model, and iii) the motor improvement model. Demographic, cognitive/motor assessments and therapy information are used to train these models. Cognitive improvement and therapy compliance variables are adapted to a binary classification problem (i and iii). These models are able to identify ‘at risk’ patients with low cognitive improvement and therapy compliance. The prediction of the motor capacities at discharge is made using the items from the Fugl-Meyer Assessment Upper Extremity (FMA-UE) adapted to a multiple linear regression problem.
Addressing target audiences and expressing needs
- We are sharing our knowledge
- Collaboration
We are looking for clinical and academic partners interested in stroke cognitive and motor rehabilitation. We wish to develop and validate this solution so that this research can have a real beneficial impact on society.
- Others/ No specific audience
- Research and Technology Organisations
- Academia/ Universities
R&D, Technology and Innovation aspects
The current model is a proof of concept AI predictive model that has been tested in lab conditions on retrospective data. The next step is to stress test the model for bias analysis using external datasets and progress to a prospective data study.
Assuming relevant data is available it is possible to retrain to adapt it to different populations, and even adapt it to different prediction tasks.
The product is an AI prediction model, so it can be easily replicated and deployed on multiple sites as required. However, prior to this happening, its performance should be validated across multiple datasets with patients from multiple ethnicities to test for bias and stability across different demographics and ethnicities.
Given the proliferation of AI tools for personalised predictions, the use of these types of tools to predict the rehabilitation outcomes of patients has the potential to have a significant impact across many healthcare areas, thereby adding value in all these areas for stakeholders and society.
