Explainable Therapy Related Annotations: Patient & Practitioner Oriented Learning Assisting Trust & Engagement
It takes a lot of manual effort to quality-assure therapy sessions. Automated rating systems can provide a useful function, saving time and scarce professional resources to check whether the therapist is following good practice, how well that fits with the individual patient, and the impact on their health.
This project will examine the key components of trust in algorithm driven digital mental health through a participatory design and dissemination study. A co-create collaborative machine learning decision support tool will be developed to help mental health practitioners and patients classify key processes in therapy transcripts. The tool will speed up rating of therapist fidelity and assessment of patient activation, thereby providing evidence for improving practice.
Underpinned by stakeholder participation and inclusive social dialogue this project partners with patients and carers to include their perspectives on the approach taken, to gain a more robust understanding of the context and the social impact of the techniques used.
This project runs from: 1/3/2020 – 15/10/2020
Project Partners: University of Nottingham, Nottinghamshire Healthcare NHS Trust, University of Essex, Academy for Recovery Coaching, University of Lincoln, Virtual Health Labs, University of Glasgow
Funded by UKRI, EPSRC
Horizon Digital Economy Research, University of
Nottingham Innovation Park, Triumph Road, Nottingham,