DCAN: Deep Contour-Aware Networks for Accurate Gland Segmentation

By Hao Chen, Xiaojuan Qi, Lequan Yu, P. Heng. Computer Vision and Pattern Recognition, 2016

636 citations recorded by the source corpus.

The morphology of glands has been used routinely by pathologists to assess the malignancy degree of adenocarcinomas. Accurate segmentation of glands from histology images is a crucial step to obtain reliable morphological statistics for quantitative diagnosis. In this paper, we proposed an efficient deep contour-aware network (DCAN) to solve this challenging problem under a unified multi-task learning framework. In the proposed network, multi-level contextual features from the hierarchical architecture are explored with auxiliary supervision for accurate gland segmentation. When incorporated with multi-task regularization during the training, the discriminative capability of intermediate features can be further improved. Moreover, our network can not only output accurate probability maps of glands, but also depict clear contours simultaneously for separating clustered objects, which further boosts the gland segmentation performance. This unified framework can be efficient when applied to large-scale histopathological data without resorting to additional steps to generate contours based on low-level cues for post-separating. Our method won the 2015 MICCAI Gland Segmentation Challenge out of 13 competitive teams, surpassing all the other methods by a significant margin.

Subfields: Artificial Intelligence, Biomedical Engineering, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/ba124bbde8d114a9103eb5751036e2e4710e2fff

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. P. Heng
2. Qi-Yan Dou
3. Jing Qin
4. Tarek M. Taha
5. Md. Zahangir Alom
6. Vijayan K. Asari
7. Lequan Yu
8. Hao Chen

In Biomedical Engineering

1. P. Heng
2. Qi-Yan Dou
3. Lequan Yu
4. Hao Chen
5. Nasir M. Rajpoot
6. Simon Graham
7. Daisuke Komura
8. Shumpei Ishikawa

In Radiology, Nuclear Medicine and Imaging

1. P. Heng
2. Hao Chen
3. Qi-Yan Dou
4. Lequan Yu
5. Jing Qin
6. B. van Ginneken
7. Simon Graham
8. Nasir M. Rajpoot

Citations in the corpus

• Semi-supervised Learning for Network-Based Cardiac MR Image Segmentation (2017, background): “1 Related Works Many works have proposed using FCN for medical image segmentation [3,4,12].”
• Adaptive Weighting Multi-Field-Of-View CNN for Semantic Segmentation in Pathology (2019, background): “The state-of-the-art methods accurately distinguish regions (normal and tumor) in digital pathology images [7][42][43].”
• Enabling a Single Deep Learning Model for Accurate Gland Instance Segmentation: A Shape-Aware Adversarial Learning Framework (2020, methodology): “[minLength,maxLength] = [12, 24]), we can use a simple curve fitting method (i.”
• Deeply-supervised CNN for prostate segmentation (2017, background): “[25] proposed a deep contour-aware network that integrates multi-level contextual features to segment glands.”
• Recurrent residual U-Net for medical image segmentation (2019, methodology): “The deep contour-aware network had been proposed in 2016, which can extract multilevel contextual features using a hierarchical architecture for accurate gland segmentation of histology images, and it shows very good performance for segmentation.(35) Furthermore, Nabla-Net, a deep dig-like convoluti”
• Medical Image Computing and Computer-Assisted Intervention − MICCAI 2017: 20th International Conference, Quebec City, QC, Canada, September 11-13, 2017, Proceedings, Part II (2017, positive): “Although, VAE have been extended to semi-supervised classification [3, 4,9], it has not been applied on the segmentation task.”
• The impact of pre- and post-image processing techniques on deep learning frameworks: A comprehensive review for digital pathology image analysis (2020, background): “[193] proposed a novel deep contour-aware network that both depicted the gland object contours and output segmentation probability maps.”
• CellViT: Vision Transformers for Precise Cell Segmentation and Classification (2023, methodology): “In contrast, boundary-based methods such as DCAN [45] and TSFD-Net [9] adopt a di ff erent approach, where instead of us-ing distance maps, watershed markers, or polygon predictions, they directly predict the nuclear contour using a prediction map.”
• Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation (2018, background): “A deep contour-aware network called Deep ContourAware Networks (DCAN) was proposed in 2016, which can extract multi-level contextual features using a hierarchical architecture for accurate gland segmentation of histology images and shows very good performance for segmentation [34].”
• A survey on deep learning in medical image analysis (2017, methodology): “…classified with SVM Shen et al. (2015b) Three CNNs at different scales estimate nodule malignancy scores of radiologists (LIDC-IDRI data set) Chen et al. (2017) Combines features from CNN, SDAE and classical features to characterize nodules from LIDC-IDRI data set Ciompi et al. (2016)…”
• Hover-Net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images (2018, background): “Other methods exploit information about the nuclear contour (or boundary) within the network, such as DCAN [25] that utilised a dual architecture that outputs the nuclear cluster and the nuclear contour as two separate prediction maps.”
• Cell Detection with Star-convex Polygons (2018, background): “The purpose of this is to differentiate crowded cells with touching borders (similar to [4,5]).”
• Multilevel Contextual 3-D CNNs for False Positive Reduction in Pulmonary Nodule Detection (2017, background): “From a broader perspective, while 2-D CNNs have witnessed many fruitful applications in medical image analysis field in recent years [18]–[20], 3-D CNN is still in its infant stage in medical applications even though 3-D medical data are quite common and popular in clinical practice.”
• Multimodal Co-Attention Transformer for Survival Prediction in Gigapixel Whole Slide Images (2021, background): “…the entire composition of a WSI and dependent on the interactions of its components ( i.e. - a fine-grained visual recognition problem such as one that involves understanding the complex milieu of stroma, tumor aggregates, immune cells and other visual concepts) [68, 7, 47, 22, 40, 21, 6, 39, 15].”