AI & Lymphoma

Unmet need

PET/CT is used in many cancer types, to detect and determine the spread of disease, monitor treatment response and detect cancer relapse. State-of-the-art clinical PET/CT reading software supports clinicians reading and deriving measurements from the images that inform the doctor’s impression and report.

Mirada Medical, in close collaboration with partners at Leeds Teaching Hospitals and the Alliance Medical Group, is investigating the role that Artificial Intelligence and Deep Learning can play in diagnostic nuclear medicine. The partners are working in concert to develop and deploy the next generation of intelligent and integrated imaging solutions for diagnostics applications in the clinic. The solution will build on Mirada’s vendor-neutral XD diagnostics software application which is used to read more than half of the PET/CT scans in England & Wales through its partnership with the Alliance Medical Group and the NHS PET scanning program.

Project aim and objectives

The aim of this project is to develop software technology that will yield benefits to cancer patients through improved efficiency and accuracy of cancer detection and diagnosis, enabling more comprehensive and personalized diagnostics decisions, and faster cancer treatment delivery for improved patient outcomes.

The partners aim to accelerate the development and the deployment of AI-based solutions that seamlessly integrate within the existing PET/CT diagnostics clinical workflow through a breadth of combined expertise in academic and industrial medical imaging research, clinical translation and validation, as well as in deployment and PET-CT operations. This will help the NHS deliver the highest standards of care using cutting edge technology.

This project encompasses a broad range of activities to deliver those end-user improvements, ranging from scientific research, prototyping and validation, to user-centred design, and real-world clinical evaluation of the technology.

Achievements

A large volume of high-quality clinically annotated datasets have been collated and are currently being used in technical feasibility studies. The team is also working on designing a clinical validation study.

An early proof-of-concept has demonstrated the technical feasibility.