New AI cancer pathology research platform launched

Octopath is a new UCL research platform that emerged from the UCL Department of Medical Physics and Biomedical Engineering, spearheaded by UKRI Future Leaders Fellow, Dr Charles-Antoine Collins-Fekete, and Senior Research Fellows, Dr Mikael Simard and Dr. Zhuoyan Shen.

The core vision of Octopath revolves around harnessing advanced artificial intelligence (AI) algorithms to revolutionise cancer pathology diagnostics, with a focus on diagnosis, mitosis detection, immune cell identification, tumour stroma ratio assessment, and angiogenesis analysis.

The Challenge

The vision and innovative solutions that Octopath are advancing have been developed in response to the pathology crisis. Over the last two decades, cancer treatment has undergone significant transformation, driven by the detection of targetable genetic alterations and advancements in genomic sequencing and immuno-oncology.

These developments have facilitated the creation of targeted therapies tailored to specific cancer subtypes. However, this progress has also led to a concurrent increase in the complexity of cancer classification, heightening the need for timely diagnostic services. Clinicians require more specialised knowledge to guide treatment decisions and an increasing workload is being delivered by a shrinking workforce, and a crisis point is being reached, with a 25% shortfall of staff to report results already evident. Thus, a sustainable solution is essential to support the diagnostic pathway amidst these challenges.

The Solution

To address this, we’ve developed state-of-the-art AI tools to support pathology, focusing on accelerating cancer diagnosis and prognostication by enhancing state-of-the-art computer vision with domain-specific technology. Seeing the growing interest in our research, and to facilitate its adoption, we created Octopath, a research platform to empower pathologists to use the latest AI tools to help progress their clinical research.

The platform consists of a research web interface to simplify the analysis of digitised tissue specimens, allowing research clinicians direct interaction with AI tools. Our long-term vision is to integrate this technology into healthcare systems for widespread clinical use, thereby improving the accuracy and efficiency of cancer diagnostics and contributing to the advancement of personalised medicine. Octopath has already been used in early-lung cancer, sarcoma, oral cancer, and inflammatory bowel disease, and the community feedback has been spectacular.

Octopath’s development was generously supported by the UKRI Future Leaders Fellowship, No. MR/T040785/1, the Radiation Research Unit at the Cancer Research UK City of London Centre Award C7893/A28990, the UKRI AI for Health Award EP/Y020030/1., and the UCL CDI Impact Accelerator program.

Visit the Octopath website