Details

Autor(en) / Beteiligte

• Graham, Simon
• Chen, Hao
• Gamper, Jevgenij
• Dou, Qi
• Heng, Pheng-Ann
• Snead, David
• Tsang, Yee Wah
• Rajpoot, Nasir

Titel

MILD-Net: Minimal information loss dilated network for gland instance segmentation in colon histology images

Ist Teil von

• Medical image analysis, 2019-02, Vol.52, p.199-211

Ort / Verlag

Netherlands: Elsevier B.V

Erscheinungsjahr

2019

Quelle

MEDLINE

Beschreibungen/Notizen

• •Accurate gland instance segmentation in colon histology images using a novel minimal information loss dilated network.•Uncertainty quantification using the concept of random transformation sampling.•A variant of the proposed framework for simultaneous gland and lumen segmentation.•Detailed experimentation and comparative analysis on two independent colorectal adenocarcinoma datasets.•Perform gland instance segmentation on two further datasets containing whole-slide images to demonstrate the extensibility of the proposed method. [Display omitted] The analysis of glandular morphology within colon histopathology images is an important step in determining the grade of colon cancer. Despite the importance of this task, manual segmentation is laborious, time-consuming and can suffer from subjectivity among pathologists. The rise of computational pathology has led to the development of automated methods for gland segmentation that aim to overcome the challenges of manual segmentation. However, this task is non-trivial due to the large variability in glandular appearance and the difficulty in differentiating between certain glandular and non-glandular histological structures. Furthermore, a measure of uncertainty is essential for diagnostic decision making. To address these challenges, we propose a fully convolutional neural network that counters the loss of information caused by max-pooling by re-introducing the original image at multiple points within the network. We also use atrous spatial pyramid pooling with varying dilation rates for preserving the resolution and multi-level aggregation. To incorporate uncertainty, we introduce random transformations during test time for an enhanced segmentation result that simultaneously generates an uncertainty map, highlighting areas of ambiguity. We show that this map can be used to define a metric for disregarding predictions with high uncertainty. The proposed network achieves state-of-the-art performance on the GlaS challenge dataset and on a second independent colorectal adenocarcinoma dataset. In addition, we perform gland instance segmentation on whole-slide images from two further datasets to highlight the generalisability of our method. As an extension, we introduce MILD-Net+ for simultaneous gland and lumen segmentation, to increase the diagnostic power of the network.