10102111Encoding Empathy: Solving the Healthcare Workforce Crisis Through the Safe Deployment of an AI-Driven Voice-Based Clinical Conversational Assistant for Long Term MonitoringActiveCollaborative R&DInnovate UKThis project aims to develop, evaluate and deploy a new technology that will improve the way patients with long-term health conditions receive care. The project is led by Ufonia, a UK company that specialises in delivering automated clinical conversations over the telephone. Ufonia has already developed an artificial intelligence (AI) software called Dora that is used at 13 NHS sites to automate appointments for patients undergoing routine surgery. Currently, Dora relies on a form of AI that can only follow set rules. Recent advancements in AI technology mean that large language models (LLM) can now be trained to have **more empathetic, personalised clinical conversations** with patients. This project will work with doctors, researchers and patients to build on Ufonia's existing technology platform for voice conversations and replace Dora's existing core language model with a **LLM model trained specifically for empathetic, personalised conversations.** This will be particularly helpful for patients with long-term conditions, as it will allow them to have regular check-ins and consultations without needing to visit the hospital in person. This will improve outcomes for patients and reduce the strain on healthcare systems. In addition to the words used during a clinical conversation, the voice style and flow is vital to **show empathy**. We will also work with the University of Newcastle to develop a linguistic approach called the CADENCE model, which will improve the **voice cues of empathy in autonomous conversation**s. One of the main challenges in using AI in healthcare is **ensuring that it is safe and effective**. To address this, the project will collaborate with the University of York's Centre for the Safety Assurance of AI-enabled Autonomous Systems (SAINTS) to develop a framework to ensure the clinical safety of the AI conversations. Specifically, this work will enable Dora to screen for patients at high risk of repeat bone fractures, and long term monitoring of patients with osteoporosis. The project will culminate in a clinical study at University College London Hospitals NHS Foundation Trust's Rheumatology Department. The results from this study will provide evidence for wider use of the technology and its potential expansion to other long-term conditions. This project has the potential to greatly improve the way patients with long-term conditions receive care. By automating regular interactions and implementing AI technology, patients will have more convenient access to the care they need, while healthcare professionals can focus on those who require their expertise the most.