ProCAncer-I (RETRò & Prospective)
An AI Platform integrating imaging data and models, supporting precision care through prostate cancer’s continuum.
The Royal Marsden NHS FT
Grant Agreement 952159, Horizon 2020 Research and Innovation Programme
Duration of Study in the UK
3 years, 8 months, 1 days
In Europe, prostate cancer (PCa) is the second most frequent type of cancer in men and the third most lethal. Current clinical practices, often leading to overdiagnosis and overtreatment of indolent tumors, suffer from lack of precision. This calls for advanced Artificial Intelligence (AI) models to decipher non-intuitive, high-level medical image patterns and increase performance in discriminating indolent from aggressive disease early on. This extends to these models also predicting recurrence, detecting metastases and predicting the effectiveness of therapies. To date, efforts in this field are fragmented, based on single–institution, size-limited and vendor-specific datasets while available PCa public datasets are only a few hundred cases, making model generalisability impossible.
The ProCAncer-I project brings together 13 partners (the consortium), including The Royal Marsden NHS Foundation Trust (RMH), PCa centers, world leaders in AI and innovative enterprises with recognised expertise in their respective domains. The objective is to design, develop and sustain a cloud-based, secure European Image Infrastructure with tools and services for data handling. The platform hosts the largest collection of PCa multi-parametric Magnetic Resonance Imaging (mpMRI) scans and anonymised image data worldwide with more than 17,000 cases, based on retrospective and prospective data from the consortium in line with EU legislation (GDPR).
Robust AI models will be developed, based on novel learning methodologies, leading to AI models that will address nine PCa clinical scenarios. To accelerate the clinical adoption of PCa AI models, the project focuses on improving the trust in the AI solutions with respect to fairness, safety, explainability and reproducibility. Metrics to monitor model performance are being developed to further increase clinical trust and inform on possible failures and errors, hopefully validating the effectiveness of AI-based models for clinical decision making.
Yorkshire & The Humber - South Yorkshire Research Ethics Committee
Date of REC Opinion
25 Apr 2022