Google Health has launched a new open model for health app developers, described as its “most capable open model for multimodal medical text and image comprehension”, built on Gemma 3. Potential use cases include medical image classification, medical image interpretation, and medical text comprehension and clinical reasoning.
In a LinkedIn post, Google Health shared how the model, named MedGemma, would be an “ideal starting point” for developers looking to create health applications, supporting customisation to suit specific needs and to run in different environments, offering cost efficiency thanks to its small size, and granting control over model versioning and lifecycle.
On its website, Google shares how the tool has been evaluated on “a range of clinically relevant benchmarks” to support their baseline performance, including open and curated benchmark datasets. Part of the tool, includes a large language model component for potential use cases of patient interviewing, triaging, clinical decision support, and summarisation.
Further information for developers is also provided, covering running the model locally, deploying an online service, and fine-tuning.
Wider trend: Innovation in healthcare technology
HTN’s AI and Data Awards presented some of the latest innovations utilising AI and data for diagnosis, treatment, communication, system efficiency, and more.
OpenAI, the AI research and deployment company responsible for ChatGPT, has launched a new benchmark for evaluating the capabilities of health AI systems, built in partnership with 262 physicians practicing across 60 countries. The company shares findings that large language models “have improved significantly over time and already outperform experts in writing responses to examples tested in our benchmark”.
NHS Scotland has published a prior information notice for a framework agreement to deliver a national digital diabetes remission programme, relying on digital solutions such as emotional wellbeing tools and app-based systems to aid with self-management activities. The programme aims to provide support to eligible individuals with type 2 diabetes, helping them to achieve remission through a “structured, digitally delivered intervention”.
Guy’s and St Thomas’ NHS Foundation Trust has entered a partnership with Aneira Health to offer AI-enabled healthcare for women working for the NHS, focusing on fitting around busy lives, improving access, and delivering personalised, proactive care. The service is expected to launch this spring, with a small cohort of the trust’s workforce involved in testing the approach.
The West of England Pathology Network has published a prior information notice for the procurement of an end-to-end digital histopathology pathway. The group is looking for a single supplier to provide a solution that comes with an image management system, scanners, monitors and associated hardware, while also offering storage and the ability to “link seamlessly into trust’s LIMs” and with 3rd party AI providers.
