Health2047 portfolio company HOPPR has introduced the multimodal foundation model, Grace, for medical imaging.

Powered by Amazon Web Services (AWS), the new solution is available through private beta to AI companies, developers, and radiology PACS for fine-tuning and application development.

HOPPR built its foundation model specifically on AWS using Amazon SageMaker.

The company intends to expand its use of AWS services, including AWS HealthImaging and Amazon Bedrock, for tasks such as data storage, inferencing, and model development.

Claimed to be the first-of-its-kind B2B foundation model, Grace facilitates text-to-image and image-to-image learning across all medical imaging modalities, including echocardiograms, MRIs, CTs, and X-rays.

The new model, which is available through an API service, allows application developers to rapidly develop AI solutions to help physicians, technicians, and support staff engage interactively with medical images.

A company can utilise its own data to enhance the model securely. This improved model can be utilised in applications that enable users to interact with medical imaging studies, providing information about findings, suggested surgical interventions, and alternative imaging views and treatment protocols.

The model can also be used for non-clinical use cases, including billing and coding review, workflow, and QA, serving as a one-API shop for all the data required for the imaging sector.

Grace has been developed by using over a petabyte of permission-based, anonymised medical imaging study results.

HOPPR CEO Dr Khan Siddiqui said: “We are thrilled to launch the beta HOPPR foundation model to trusted PACS vendors and developers to fine-tune models and provide feedback to prepare us for commercial expansion in Q1 [first quarter] of 2024.

“Grace represents a game-changing advance for HOPPR and the broader medical imaging space, which stands to benefit enormously from the transformative potential of AI to improve the efficiency and quality of clinical care.”