Concr modelling accurately identifies cell types from histopathology images

05 May 2023

In this study, presented at AACR Annual Meeting 2023, a deep learning model was developed to reliably identify 3 cell types relevant for treatment of triple negative breast cancers (TNBC) using H&E-stained slides, with performance validated by an expert pathologist.

Histopathology assessments in cancer are currently greatly labour-intensive, needing many hours of highly skilled pathologists’ time to make a diagnosis. This task is getting more specialised and complex with evolution of oncology treatments, particularly immunotherapy.

Computational predictions that use computer vision and machine learning can assist pathologists with their assessment. Furthermore, digital image analysis can be incorporated into more complex predictive modelling for accurate outcome predictions, enabling precision treatment for cancer patients.

Download poster

Abstract for this study was published in Cancer Research.