Accurate Detection of Out of Body Segments in Surgical Video using Semi-Supervised Learning

Maya Zohar, Omri Bar, Daniel Neimark, Gregory D. Hager, Dotan Asselmann
Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:923-936, 2020.

Abstract

Large labeled datasets are an important precondition for deep learning models to achieve state-of-the-art results in computer vision tasks. In the medical imaging domain, privacy concerns have limited the rate of adoption of artificial intelligence methodologies into clinical practice. To alleviate such concerns, and increase comfort levels while sharing and storing surgical video data, we propose a high accuracy method for rapid removal and anonymization of out-of-body and non-relevant surgery segments. Training a deep model to detect out-of-body and non-relevant segments in surgical videos requires suitable labeling. Since annotating surgical videos with per-second relevancy labeling is a tedious task, our proposed framework initiates the learning process from a weakly labeled noisy dataset and iteratively applies Semi-Supervised Learning (SSL) to re-annotate the training data samples. Evaluating our model, on an independent test set, shows a mean detection accuracy of above $97%$ after several training-annotating iterations. Since our final goal is achieving out-of-body segments detection for anonymization, we evaluate our ability to detect these segments at a high demanding recall of $97%$, which leads to a precision of $83.5%$. We believe this approach can be applied to similar related medical problems, in which only a coarse set of relevancy labels exists, currently limiting the possibility for supervision training.

Cite this Paper

BibTeX
@InProceedings{pmlr-v121-zohar20a, title = {Accurate Detection of Out of Body Segments in Surgical Video using Semi-Supervised Learning}, author = {Zohar, Maya and Bar, Omri and Neimark, Daniel and Hager, Gregory D. and Asselmann, Dotan}, booktitle = {Proceedings of the Third Conference on Medical Imaging with Deep Learning}, pages = {923--936}, year = {2020}, editor = {Arbel, Tal and Ben Ayed, Ismail and de Bruijne, Marleen and Descoteaux, Maxime and Lombaert, Herve and Pal, Christopher}, volume = {121}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/zohar20a/zohar20a.pdf}, url = {https://proceedings.mlr.press/v121/zohar20a.html}, abstract = {Large labeled datasets are an important precondition for deep learning models to achieve state-of-the-art results in computer vision tasks. In the medical imaging domain, privacy concerns have limited the rate of adoption of artificial intelligence methodologies into clinical practice. To alleviate such concerns, and increase comfort levels while sharing and storing surgical video data, we propose a high accuracy method for rapid removal and anonymization of out-of-body and non-relevant surgery segments. Training a deep model to detect out-of-body and non-relevant segments in surgical videos requires suitable labeling. Since annotating surgical videos with per-second relevancy labeling is a tedious task, our proposed framework initiates the learning process from a weakly labeled noisy dataset and iteratively applies Semi-Supervised Learning (SSL) to re-annotate the training data samples. Evaluating our model, on an independent test set, shows a mean detection accuracy of above $97%$ after several training-annotating iterations. Since our final goal is achieving out-of-body segments detection for anonymization, we evaluate our ability to detect these segments at a high demanding recall of $97%$, which leads to a precision of $83.5%$. We believe this approach can be applied to similar related medical problems, in which only a coarse set of relevancy labels exists, currently limiting the possibility for supervision training.} }
Endnote
%0 Conference Paper %T Accurate Detection of Out of Body Segments in Surgical Video using Semi-Supervised Learning %A Maya Zohar %A Omri Bar %A Daniel Neimark %A Gregory D. Hager %A Dotan Asselmann %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-zohar20a %I PMLR %P 923--936 %U https://proceedings.mlr.press/v121/zohar20a.html %V 121 %X Large labeled datasets are an important precondition for deep learning models to achieve state-of-the-art results in computer vision tasks. In the medical imaging domain, privacy concerns have limited the rate of adoption of artificial intelligence methodologies into clinical practice. To alleviate such concerns, and increase comfort levels while sharing and storing surgical video data, we propose a high accuracy method for rapid removal and anonymization of out-of-body and non-relevant surgery segments. Training a deep model to detect out-of-body and non-relevant segments in surgical videos requires suitable labeling. Since annotating surgical videos with per-second relevancy labeling is a tedious task, our proposed framework initiates the learning process from a weakly labeled noisy dataset and iteratively applies Semi-Supervised Learning (SSL) to re-annotate the training data samples. Evaluating our model, on an independent test set, shows a mean detection accuracy of above $97%$ after several training-annotating iterations. Since our final goal is achieving out-of-body segments detection for anonymization, we evaluate our ability to detect these segments at a high demanding recall of $97%$, which leads to a precision of $83.5%$. We believe this approach can be applied to similar related medical problems, in which only a coarse set of relevancy labels exists, currently limiting the possibility for supervision training.
APA
Zohar, M., Bar, O., Neimark, D., Hager, G.D. & Asselmann, D.. (2020). Accurate Detection of Out of Body Segments in Surgical Video using Semi-Supervised Learning. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 121:923-936 Available from https://proceedings.mlr.press/v121/zohar20a.html.

Related Material