Adaptive Weighting Multi-Field-Of-View CNN for Semantic Segmentation in Pathology

By Hiroki Tokunaga, Yuki Teramoto, Akihiko Yoshizawa, Ryoma Bise. 2019

142 citations recorded by the source corpus.

Automated digital histopathology image segmentation is an important task to help pathologists diagnose tumors and cancer subtypes. For pathological diagnosis of cancer subtypes, pathologists usually change the magnification of whole-slide images (WSI) viewers. A key assumption is that the importance of the magnifications depends on the characteristics of the input image, such as cancer subtypes. In this paper, we propose a novel semantic segmentation method, called Adaptive-Weighting-Multi-Field-of-View-CNN (AWMF-CNN), that can adaptively use image features from images with different magnifications to segment multiple cancer subtype regions in the input image. The proposed method aggregates several expert CNNs for images of different magnifications by adaptively changing the weight of each expert depending on the input image. It leverages information in the images with different magnifications that might be useful for identifying the subtypes. It outperformed other state-of-the-art methods in experiments.

Subfields: Artificial Intelligence, Biophysics, Computer Vision and Pattern Recognition, Medical Imaging

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. Shiji Song
2. Yulin Wang
3. Gao Huang
4. Ming-Ming Cheng
5. Le Yang
6. Yizeng Han
7. Yin Li
8. Honghui Wang

In Biophysics

1. Ozan Ciga
2. Chunle Guo
3. Bin Li
4. Kevin W. Eliceiri
5. Anne L. Martel
6. Ming-Ming Cheng
7. Yin Li
8. Chetan L. Srinidhi

In Computer Vision and Pattern Recognition

1. Ming-Ming Cheng
2. Shiji Song
3. Gao Huang
4. Yulin Wang
5. Luc Van Gool
6. Ming-Hsuan Yang
7. Yin Li
8. Dacheng Tao

In Medical Imaging

1. Ozan Ciga
2. Bin Li
3. Kevin W. Eliceiri
4. Anne L. Martel
5. Yin Li
6. Chetan L. Srinidhi

Citations in the corpus

• Dynamic Neural Networks: A Survey (2021, background): “Image I Object detection (face [38], [181], [182], facial point [183], pedestrian [184], general [31], [185], [186], [187], [188]) Image segmentation [101], [189], Super resolution [190], Style transfer [191], Coarse-to-fine classification [192]”
• Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning (2020, methodology): “The assessment of WSIs by pathologists is done in multiscale [2, 16, 43] and it is common to consider multiscale features in WSI analysis.”
• Deep neural network models for computational histopathology: A survey (2019, methodology): “Such mechanisms are integrated into the FCN model by designing multi-magnification networks (Ho et al., 2019; Tokunaga et al., 2019), each trained on different field-ofview image patches to obtain a better discriminative feature representation compared to a single-magnification model.”
• FocusCut: Diving into a Focus View in Interactive Segmentation (2022, methodology): “Moreover, for semantic segmentation, AWMF-CNN [44] assigns weights to different magnification of local patches separately.”