SagaNet: A Small Sample Gated Network for Pediatric Cancer Diagnosis

Yuhan Liu, Shiliang Sun
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:6947-6956, 2021.

Abstract

The scarcity of available samples and the high annotation cost of medical data cause a bottleneck in many digital diagnosis tasks based on deep learning. This problem is especially severe in pediatric tumor tasks, due to the small population base of children and high sample diversity caused by the high metastasis rate of related tumors. Targeted research on pediatric tumors is urgently needed but lacks sufficient attention. In this work, we propose a novel model to solve the diagnosis task of small round blue cell tumors (SRBCTs). To solve the problem of high noise and high diversity in the small sample scenario, the model is constrained to pay attention to the valid areas in the pathological image with a masking mechanism, and a length-aware loss is proposed to improve the tolerance to feature diversity. We evaluate this framework on a challenging small sample SRBCTs dataset, whose classification is difficult even for professional pathologists. The proposed model shows the best performance compared with state-of-the-art deep models and generalization on another pathological dataset, which illustrates the potentiality of deep learning applications in difficult small sample medical tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-liu21u, title = {SagaNet: A Small Sample Gated Network for Pediatric Cancer Diagnosis}, author = {Liu, Yuhan and Sun, Shiliang}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {6947--6956}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/liu21u/liu21u.pdf}, url = {https://proceedings.mlr.press/v139/liu21u.html}, abstract = {The scarcity of available samples and the high annotation cost of medical data cause a bottleneck in many digital diagnosis tasks based on deep learning. This problem is especially severe in pediatric tumor tasks, due to the small population base of children and high sample diversity caused by the high metastasis rate of related tumors. Targeted research on pediatric tumors is urgently needed but lacks sufficient attention. In this work, we propose a novel model to solve the diagnosis task of small round blue cell tumors (SRBCTs). To solve the problem of high noise and high diversity in the small sample scenario, the model is constrained to pay attention to the valid areas in the pathological image with a masking mechanism, and a length-aware loss is proposed to improve the tolerance to feature diversity. We evaluate this framework on a challenging small sample SRBCTs dataset, whose classification is difficult even for professional pathologists. The proposed model shows the best performance compared with state-of-the-art deep models and generalization on another pathological dataset, which illustrates the potentiality of deep learning applications in difficult small sample medical tasks.} }
Endnote
%0 Conference Paper %T SagaNet: A Small Sample Gated Network for Pediatric Cancer Diagnosis %A Yuhan Liu %A Shiliang Sun %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-liu21u %I PMLR %P 6947--6956 %U https://proceedings.mlr.press/v139/liu21u.html %V 139 %X The scarcity of available samples and the high annotation cost of medical data cause a bottleneck in many digital diagnosis tasks based on deep learning. This problem is especially severe in pediatric tumor tasks, due to the small population base of children and high sample diversity caused by the high metastasis rate of related tumors. Targeted research on pediatric tumors is urgently needed but lacks sufficient attention. In this work, we propose a novel model to solve the diagnosis task of small round blue cell tumors (SRBCTs). To solve the problem of high noise and high diversity in the small sample scenario, the model is constrained to pay attention to the valid areas in the pathological image with a masking mechanism, and a length-aware loss is proposed to improve the tolerance to feature diversity. We evaluate this framework on a challenging small sample SRBCTs dataset, whose classification is difficult even for professional pathologists. The proposed model shows the best performance compared with state-of-the-art deep models and generalization on another pathological dataset, which illustrates the potentiality of deep learning applications in difficult small sample medical tasks.
APA
Liu, Y. & Sun, S.. (2021). SagaNet: A Small Sample Gated Network for Pediatric Cancer Diagnosis. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:6947-6956 Available from https://proceedings.mlr.press/v139/liu21u.html.

Related Material