Automatic tumour segmentation in H&E-stained whole-slide images of the pancreas

TL;DR

This paper presents a multi-task convolutional neural network that automatically segments pancreatic tumours in H&E-stained whole-slide images, achieving high accuracy and potentially aiding faster diagnostics.

Contribution

The study introduces a novel multi-task CNN approach for automatic tumour segmentation in pancreatic histopathology images, improving accuracy over single-task models.

Findings

Median Dice score of 0.885 for single-task segmentation.

Multi-task network achieved a median Dice of 0.934.

Validated on 58 slides from 29 patients.

Abstract

Pancreatic cancer will soon be the second leading cause of cancer-related death in Western society. Imaging techniques such as CT, MRI and ultrasound typically help providing the initial diagnosis, but histopathological assessment is still the gold standard for final confirmation of disease presence and prognosis. In recent years machine learning approaches and pathomics pipelines have shown potential in improving diagnostics and prognostics in other cancerous entities, such as breast and prostate cancer. A crucial first step in these pipelines is typically identification and segmentation of the tumour area. Ideally this step is done automatically to prevent time consuming manual annotation. We propose a multi-task convolutional neural network to balance disease detection and segmentation accuracy. We validated our approach on a dataset of 29 patients (for a total of 58 slides) at different resolutions. The best single task segmentation network achieved a median Dice of 0.885 (0.122) IQR at a resolution of 15.56 $\mu$m. Our multi-task network improved on that with a median Dice score of 0.934 (0.077) IQR.