Download the PDF poster:
Watch the presentation on-demand:
Presented at Digital Pathology & AI Congress USA 2025 | In support of the INCISE Project
As part of the INCISE project, an ambitious collaboration between the University of Glasgow, NHS Greater Glasgow and Clyde, and a network of industry partners, OracleBio led the Digital Pathology work package, helping to advance bowel cancer surveillance through scalable image analysis.
Our role focused on developing and deploying an automated pipeline to quantify Ki67 biomarker expression across a large cohort of colorectal polyp samples. This data forms part of a wider effort to create a multi-omics risk model that can help stratify patients and personalise future screening pathways.
What you’ll learn:
- A four-stage AI-driven image analysis workflow – Built in Visiopharm and validated by expert pathologists, our pipeline included tissue QC, adenoma classification, gland segmentation, and single-cell phenotyping.
- Execution at scale – The final pipeline was used to process over 2,200 Ki67-stained whole slide images, executed in parallel across 75 GPU-enabled cloud nodes.
- Fast, reliable performance – All 2,211 samples were analysed in just 45 hours—demonstrating the pipeline’s robustness, scalability, and future potential to support high-throughput diagnostic workflows.
Click the thumbnail above to download the full poster and see how OracleBio’s image analysis expertise is helping power the next generation of bowel cancer surveillance.
Stay up to date with OracleBio
Sign up to our mailing list to stay informed about news, webinars, events and more
You might also like these Posters:
High ICOS/FOXP3 Tregs Content in the Tumor Microenvironment is Associated with Poorer Survival in Patients with Hepatocellular Carcinoma
OracleBio data quantifying ICOS and FoxP3 dual IHC in clinical HCC samples is presented on a collaborative poster at the AACR 2020 Virtual Annual Meeting. The poster, produced in collaboration with Kymab and the National Taiwan University Hospital, explores the...
Applying deep learning classification for tumor identification across immunohistochemical markers on serial sections to eliminate the need for image co-registration
OracleBio presented a poster at the Society for Immunotherapy of Cancer Annual Meeting 2019 on the 6th – 10th November. The poster is entitled ‘Applying deep learning classification for tumor identification across immunohistochemical markers on serial sections to...
Characterization of Anti-CTLA-4 Effects on Tumor Growth and Immune Microenvironment in a Syngeneic CT26.WT Colon Carcinoma Mouse Model
In collaboration with Pharmatest, OracleBio presented a poster at the AACR Annual Meeting 2019, which took place from the 29th of March to the 3rd of April 2019. The poster is entitled ‘Characterization of Anti-CTLA-4 Effects on Tumor Growth and Immune...



