Generalisation of automatic tumour segmentation in histopathological whole-slide images across multiple cancer types
This work addresses the problem of automating tumor segmentation for pathologists across various cancers, showing it is possible with a single model, though it is incremental as it builds on existing segmentation methods.
The researchers developed a universal deep learning model for tumor segmentation in histopathological whole-slide images and validated it across multiple cancer types, achieving an average Dice coefficient over 80% in all validation cohorts without performance loss compared to specialized models.
Deep learning is expected to aid pathologists by automating tasks such as tumour segmentation. We aimed to develop one universal tumour segmentation model for histopathological images and examine its performance in different cancer types. The model was developed using over 20 000 whole-slide images from over 4 000 patients with colorectal, endometrial, lung, or prostate carcinoma. Performance was validated in pre-planned analyses on external cohorts with over 3 000 patients across six cancer types. Exploratory analyses included over 1 500 additional patients from The Cancer Genome Atlas. Average Dice coefficient was over 80% in all validation cohorts with en bloc resection specimens and in The Cancer Genome Atlas cohorts. No loss of performance was observed when comparing the universal model with models specialised on single cancer types. In conclusion, extensive and rigorous evaluations demonstrate that generic tumour segmentation by a single model is possible across cancer types, patient populations, sample preparations, and slide scanners.