NcIEMIL: Rethinking Decoupled Multiple Instance Learning Framework for Histopathological Slide Classification

Sun Qiehe, Doukou Jiang, Jiawen Li, Renao Yan, Yonghong He, Tian Guan, Zhiqiang Cheng
Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, PMLR 250:1166-1178, 2024.

Abstract

On account of superiority in annotation efficiency, multiple instance learning (MIL) has proved to be a promising framework for the whole slide image (WSI) classification in pathological diagnosis. However, current methods employ fully- or semi-decoupled frameworks to address the trade-off between billions of pixels and limited computational resources. This exacerbates the information bottleneck, leading to instance representations in a high-rank space that contains semantic redundancy compared to the potential low-rank category space of instances. Additionally, most negative instances are also independent of the positive properties of the bag. To address this, we introduce a weakly annotation-supervised filtering network, aiming to restore the low-rank nature of the slide-level representations. We then design a parallel aggregation structure that utilizes spatial attention mechanisms to model inter-correlation between instances and simultaneously assigns corresponding weights to channel dimensions to alleviate the redundant information introduced by feature extraction. Extensive experiments on the private gastrointestinal chemotaxis dataset and CAMELYON16 breast dataset show that our proposed framework is capable of handling both binary and multivariate classification problems and outperforms state-of-the-art MIL-based methods. The code is available at: https://github.com/polyethylene16/NcIEMIL.

Cite this Paper

BibTeX
@InProceedings{pmlr-v250-qiehe24a, title = {NcIEMIL: Rethinking Decoupled Multiple Instance Learning Framework for Histopathological Slide Classification}, author = {Qiehe, Sun and Jiang, Doukou and Li, Jiawen and Yan, Renao and He, Yonghong and Guan, Tian and Cheng, Zhiqiang}, booktitle = {Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning}, pages = {1166--1178}, year = {2024}, editor = {Burgos, Ninon and Petitjean, Caroline and Vakalopoulou, Maria and Christodoulidis, Stergios and Coupe, Pierrick and Delingette, Hervé and Lartizien, Carole and Mateus, Diana}, volume = {250}, series = {Proceedings of Machine Learning Research}, month = {03--05 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v250/main/assets/qiehe24a/qiehe24a.pdf}, url = {https://proceedings.mlr.press/v250/qiehe24a.html}, abstract = {On account of superiority in annotation efficiency, multiple instance learning (MIL) has proved to be a promising framework for the whole slide image (WSI) classification in pathological diagnosis. However, current methods employ fully- or semi-decoupled frameworks to address the trade-off between billions of pixels and limited computational resources. This exacerbates the information bottleneck, leading to instance representations in a high-rank space that contains semantic redundancy compared to the potential low-rank category space of instances. Additionally, most negative instances are also independent of the positive properties of the bag. To address this, we introduce a weakly annotation-supervised filtering network, aiming to restore the low-rank nature of the slide-level representations. We then design a parallel aggregation structure that utilizes spatial attention mechanisms to model inter-correlation between instances and simultaneously assigns corresponding weights to channel dimensions to alleviate the redundant information introduced by feature extraction. Extensive experiments on the private gastrointestinal chemotaxis dataset and CAMELYON16 breast dataset show that our proposed framework is capable of handling both binary and multivariate classification problems and outperforms state-of-the-art MIL-based methods. The code is available at: https://github.com/polyethylene16/NcIEMIL.} }
Endnote
%0 Conference Paper %T NcIEMIL: Rethinking Decoupled Multiple Instance Learning Framework for Histopathological Slide Classification %A Sun Qiehe %A Doukou Jiang %A Jiawen Li %A Renao Yan %A Yonghong He %A Tian Guan %A Zhiqiang Cheng %B Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ninon Burgos %E Caroline Petitjean %E Maria Vakalopoulou %E Stergios Christodoulidis %E Pierrick Coupe %E Hervé Delingette %E Carole Lartizien %E Diana Mateus %F pmlr-v250-qiehe24a %I PMLR %P 1166--1178 %U https://proceedings.mlr.press/v250/qiehe24a.html %V 250 %X On account of superiority in annotation efficiency, multiple instance learning (MIL) has proved to be a promising framework for the whole slide image (WSI) classification in pathological diagnosis. However, current methods employ fully- or semi-decoupled frameworks to address the trade-off between billions of pixels and limited computational resources. This exacerbates the information bottleneck, leading to instance representations in a high-rank space that contains semantic redundancy compared to the potential low-rank category space of instances. Additionally, most negative instances are also independent of the positive properties of the bag. To address this, we introduce a weakly annotation-supervised filtering network, aiming to restore the low-rank nature of the slide-level representations. We then design a parallel aggregation structure that utilizes spatial attention mechanisms to model inter-correlation between instances and simultaneously assigns corresponding weights to channel dimensions to alleviate the redundant information introduced by feature extraction. Extensive experiments on the private gastrointestinal chemotaxis dataset and CAMELYON16 breast dataset show that our proposed framework is capable of handling both binary and multivariate classification problems and outperforms state-of-the-art MIL-based methods. The code is available at: https://github.com/polyethylene16/NcIEMIL.
APA
Qiehe, S., Jiang, D., Li, J., Yan, R., He, Y., Guan, T. & Cheng, Z.. (2024). NcIEMIL: Rethinking Decoupled Multiple Instance Learning Framework for Histopathological Slide Classification. Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 250:1166-1178 Available from https://proceedings.mlr.press/v250/qiehe24a.html.

Related Material