Development and validation
Prognostic AI model identifies different tissue characteristics in colorectal cancer patient samples, and combined with two other clinical parameters, it produces a colorectal cancer recurrence risk score.
- Dr. Rish Pai from Mayo Clinic trained a highly advanced AI model for detecting multiple pathologic features of colorectal carcinoma. He used Aiforia® Create end to end by annotating all the training data himself.
- QuantCRC was verified and validated in Aiforia® Platform using whole slide images and against reviews by independent pathologists from eight different hospitals.
- The effectiveness of the AI model has been demonstrated through retrospective analysis in multiple independent patient sample cohorts¹-³.
1. Pai et al. (2022, December). Quantitative Pathologic Analysis of Digitized Images of Colorectal Carcinoma Improves Prediction of Recurrence-Free Survival. Gastroenterology, 163(6), 1531–1546. https://doi.org/10.1053/j.gastro.2022.08.025
2. Pai et al. (2021, September). Development and initial validation of a deep learning algorithm to quantify histological features in colorectal carcinoma including tumour budding/poorly differentiated clusters. Histopathology, 79(3), 391–405. https://doi.org/10.1111/his.14353
3. Wu et al. (2024, May 1). Improved Risk-Stratification Scheme for Mismatch-Repair Proficient Stage II Colorectal Cancers Using the Digital Pathology Biomarker QuantCRC. Clinical Cancer Research, 30(9), 1811-1821. https://doi.org/10.1158/1078-0432.ccr-23-3211
