Academic Lead: Dr Marc Modat
Clinical Lead: Dr Teresa Guerrero Urbano & Dr Tony Greener
Clinical Area: Radiology, Radiotherapy
Partner: Mirada Medical
Medical imaging contouring allows clinicians to see easily areas of interest and suspected disease. The performance of AI based contouring solutions depends on the quality of the data provided, and access to high-quality contoured data is difficult.
Variation in image acquisition and contouring protocols, and a lack of precise definitions for organs at risk contribute to this problem. Additionally, data governance policies often lead to the generation of site-specific models that cannot be widely deployed within the NHS and therefore do not harness the benefits of AI in healthcare at scale.
Mirada’s DLCExpert™ is a CE marked deep learning auto-contouring solution for radiotherapy treatment planning. It seeks to address the above issues and produces computer generated contouring outputs that match human performance with a greater level of clinical acceptance.
We are training DLCExpert™ on high-quality clinical data collected and curated within Guy’s and St Thomas’ Hospital to achieve greater contour accuracy and improved generalised contouring models.
Our goal is to create an off-the-shelf contouring model for the prostate and organs at risk, suitable for wide-scale deployment in the NHS.
Embedding this solution within the NHS will improve current contouring practices used for treatment planning in prostate radiotherapy and will reduce variability among radiologists performing assessments, ensuring patients receive more consistent and accurate treatment.