{"id":314,"date":"2020-05-20T10:24:41","date_gmt":"2020-05-20T08:24:41","guid":{"rendered":"http:\/\/www.itec.aau.at\/~mt\/?p=314"},"modified":"2020-07-24T10:26:49","modified_gmt":"2020-07-24T08:26:49","slug":"tool-segmentation-in-cataract-surgery-videos","status":"publish","type":"post","link":"http:\/\/www.itec.aau.at\/~mt\/2020\/05\/tool-segmentation-in-cataract-surgery-videos\/","title":{"rendered":"Tool Segmentation in Cataract Surgery Videos"},"content":{"rendered":"\n<p>Our conference paper on instrument segmentation in cataract surgery videos has been accepted for presentation at the <a href=\"http:\/\/www.cbms2020.org\">CBMS 2020 conference<\/a>.<\/p>\n\n\n\n<p><strong>Title:<\/strong> Pixel-Based Tool Segmentation in Cataract Surgery Videos with Mask R-CNN<\/p>\n\n\n\n<p><strong>Authors:<\/strong> Markus Fox, Mario Taschwer, and Klaus Schoeffmann<\/p>\n\n\n\n<p><strong>Abstract:<\/strong> Automatically detecting surgical tools in \nrecorded surgery videos is an important building block of further \ncontent-based video analysis. In ophthalmology, the results of such \nmethods can support training and teaching of operation techniques and \nenable investigation of medical research questions on a dataset of \nrecorded surgery videos. Our work applies a recent deep-learning \nsegmentation method (Mask R-CNN) to localize and segment surgical tools \nused in ophthalmic cataract surgery. We add ground-truth annotations for\n multi-class instance segmentation to two existing datasets of cataract \nsurgery videos and make resulting datasets publicly available for \nresearch purposes. In the absence of comparable results from literature,\n we tune and evaluate Mask R-CNN on these datasets for instrument \nsegmentation\/localization and achieve promising results (61% mean \naverage precision on 50% intersection over union for instance \nsegmentation, working even better for bounding box detection or binary \nsegmentation), establishing a reasonable baseline for further research. \nMoreover, we experiment with common data augmentation techniques and \nanalyze the achieved segmentation performance with respect to each class\n (instrument), providing evidence for future improvements of this \napproach.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Our conference paper on instrument segmentation in cataract surgery videos has been accepted for presentation at the CBMS 2020 conference. Title: Pixel-Based Tool Segmentation in Cataract Surgery Videos with Mask R-CNN Authors: Markus Fox, Mario Taschwer, and Klaus Schoeffmann Abstract: Automatically detecting surgical tools in recorded surgery videos is an important building block of further [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[4],"tags":[],"class_list":["post-314","post","type-post","status-publish","format-standard","hentry","category-research"],"_links":{"self":[{"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/posts\/314","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/comments?post=314"}],"version-history":[{"count":1,"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/posts\/314\/revisions"}],"predecessor-version":[{"id":315,"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/posts\/314\/revisions\/315"}],"wp:attachment":[{"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/media?parent=314"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/categories?post=314"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.itec.aau.at\/~mt\/wp-json\/wp\/v2\/tags?post=314"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}