Innovating the next generation of AI health diagnostic tools

Across the UK, there is a significant shortage of trained occupational health (OH) professionals, which has led to long waiting lists and increasing costs to providers of the service. With the vision of providing world class, cost effective OH services, the Heales Medical Group sought cutting edge technology to innovate their current offering to reach more people, enhance user experience, and minimise staff absence while maximising staff wellbeing.

To address chronic staff shortages, Heales embarked on a radical approach, leveraging AI diagnostics and machine learning in an accessible, patient-focused recommendation platform. This approach automated many of the repetitive, but necessary, rote assessment and reporting activities involved in consultations, thus, offering OH practitioners more time to focus on providing a better service to a wider number of clients. Consequently, the recommendation platform opened a whole new global market for Heales through remote access to OH practitioners from the comfort of a user’s home.

Through the Knowledge Transfer Partnership (KTP) programme, Heales collaborated with the University of Hertfordshire (UH) in June 2021, successfully completing a research project developing novel artificial intelligent diagnostic health recommendation tools. Over 24 months, the project team, led by the Associate Alok Sahu and supervised by Eamonn Swanton (Heales), Prof Farshid Amirabdollahian, and Dr Yi Sun (UH), worked to identify transformer-based clinical AI models for triaging as well as automatic questions and management advice report generation. In addition to integrating the models into existing systems, Heales staff were trained in these new areas of expertise, ultimately, achieving the perfect triangle of Knowledge Exchange between the company and the University of Hertfordshire via the project coordination and technical skills of the Associate.

By the project’s completion in June 2023, it had delivered positive impacts across several areas of Heales, including improved customer satisfaction, increased efficiencies and productivity, and a rise in the number of people benefiting from the new service. While Heales are still in the midst of embedding the new technology in their systems and processes, with more efficient OH processes and a higher margin on licensing, they anticipate a projected annual profit increase in the region of £1.17 million from an additional turner of £2.44 million by June 2026 (3 years post project end), resulting from an increased profit margin of 30% or more and additional licensing, whilst achieving reduced costs for clients. Recently, Heales has embarked on another KTP with the University of Hertfordshire, with the previous Associate Alok Sahu now line managing a new Associate project lead, Akhila Anil. When asked about his experience of the KTP programme, Eamonn Swanton said:

“We’ve found working with our second Associate and the UH team to be a very pleasurable and productive experience. We’ve developed a greater knowledge of machine learning and AI and have realised just how much we don’t know and how much more we can do, hence, the next KTP.”