A Self-improving Skin Lesions Diagnosis Framework Via Pseudo-labeling and Self-distillation

Shaochang Deng, Mengxiao Yin, Feng Yang
Proceedings of The 14th Asian Conference on Machine Learning, PMLR 189:296-310, 2023.

Abstract

In the past few years, supervised-based deep learning methods has yielded good results in skin lesions diagnosis tasks. Unfortunately, obtaining large of labels for medical images is expensive and time consuming. In this paper, we propose a self-improving skin lesions diagnosis (SISLD) framework to explore useful information in unlabeled data. We first propose a semi-supervised model ${f}$, which combining consistency and class-balanced pseudo-labeling to make full use of unlabeled data in scenarios with sparse manually labeled samples, and obtain a teacher model ${f_{t}}$ by semi-supervised self-training. Then, we introduce self-distillation method to enable knowledge distillation for the diagnosis of skin lesions. Finally, we measure diagnostic effectiveness in the context of label sparsity and class imbalance. The experiments on skin lesion images dataset ISIC2018 shows that SISLD achieves significant improvements in AUC, Accuracy, Specificity and Sensitivity.

Cite this Paper


BibTeX
@InProceedings{pmlr-v189-deng23a, title = {A Self-improving Skin Lesions Diagnosis Framework Via Pseudo-labeling and Self-distillation}, author = {Deng, Shaochang and Yin, Mengxiao and Yang, Feng}, booktitle = {Proceedings of The 14th Asian Conference on Machine Learning}, pages = {296--310}, year = {2023}, editor = {Khan, Emtiyaz and Gonen, Mehmet}, volume = {189}, series = {Proceedings of Machine Learning Research}, month = {12--14 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v189/deng23a/deng23a.pdf}, url = {https://proceedings.mlr.press/v189/deng23a.html}, abstract = {In the past few years, supervised-based deep learning methods has yielded good results in skin lesions diagnosis tasks. Unfortunately, obtaining large of labels for medical images is expensive and time consuming. In this paper, we propose a self-improving skin lesions diagnosis (SISLD) framework to explore useful information in unlabeled data. We first propose a semi-supervised model ${f}$, which combining consistency and class-balanced pseudo-labeling to make full use of unlabeled data in scenarios with sparse manually labeled samples, and obtain a teacher model ${f_{t}}$ by semi-supervised self-training. Then, we introduce self-distillation method to enable knowledge distillation for the diagnosis of skin lesions. Finally, we measure diagnostic effectiveness in the context of label sparsity and class imbalance. The experiments on skin lesion images dataset ISIC2018 shows that SISLD achieves significant improvements in AUC, Accuracy, Specificity and Sensitivity.} }
Endnote
%0 Conference Paper %T A Self-improving Skin Lesions Diagnosis Framework Via Pseudo-labeling and Self-distillation %A Shaochang Deng %A Mengxiao Yin %A Feng Yang %B Proceedings of The 14th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Emtiyaz Khan %E Mehmet Gonen %F pmlr-v189-deng23a %I PMLR %P 296--310 %U https://proceedings.mlr.press/v189/deng23a.html %V 189 %X In the past few years, supervised-based deep learning methods has yielded good results in skin lesions diagnosis tasks. Unfortunately, obtaining large of labels for medical images is expensive and time consuming. In this paper, we propose a self-improving skin lesions diagnosis (SISLD) framework to explore useful information in unlabeled data. We first propose a semi-supervised model ${f}$, which combining consistency and class-balanced pseudo-labeling to make full use of unlabeled data in scenarios with sparse manually labeled samples, and obtain a teacher model ${f_{t}}$ by semi-supervised self-training. Then, we introduce self-distillation method to enable knowledge distillation for the diagnosis of skin lesions. Finally, we measure diagnostic effectiveness in the context of label sparsity and class imbalance. The experiments on skin lesion images dataset ISIC2018 shows that SISLD achieves significant improvements in AUC, Accuracy, Specificity and Sensitivity.
APA
Deng, S., Yin, M. & Yang, F.. (2023). A Self-improving Skin Lesions Diagnosis Framework Via Pseudo-labeling and Self-distillation. Proceedings of The 14th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 189:296-310 Available from https://proceedings.mlr.press/v189/deng23a.html.

Related Material