Jian Shi, Yi Xin, Benlian Xu, Mingli Lu, Jinliang Cong
{"title":"A Deep Framework for Cell Mitosis Detection in Microscopy Images","authors":"Jian Shi, Yi Xin, Benlian Xu, Mingli Lu, Jinliang Cong","doi":"10.1109/CIS52066.2020.00030","DOIUrl":null,"url":null,"abstract":"Detection and tracking of multiple cells is critical in biomedical research and computer vision. Resolving lineage relationships between mitotic cells has been of fundamental interest in this filed recently. Microscopy images with cells at poor imagining conditions are difficult to detect and manual operation still remains standard procedure. This paper proposed a cell detection framework consisting of a convolution neural network (CNN) cell detector and a convolutional long short-term memory (LSTM) model. The detector is modeled by a well-trained Faster RCNN network to learn various cell features, and the convolutional LSTM network is employed to capture cell mitotic events, which utilizes both appearance and motion information from candidate sequences. Experimental results on realistic low contrast cell images are presented to demonstrate the robustness and validation of the proposed method.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 16th International Conference on Computational Intelligence and Security (CIS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CIS52066.2020.00030","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Detection and tracking of multiple cells is critical in biomedical research and computer vision. Resolving lineage relationships between mitotic cells has been of fundamental interest in this filed recently. Microscopy images with cells at poor imagining conditions are difficult to detect and manual operation still remains standard procedure. This paper proposed a cell detection framework consisting of a convolution neural network (CNN) cell detector and a convolutional long short-term memory (LSTM) model. The detector is modeled by a well-trained Faster RCNN network to learn various cell features, and the convolutional LSTM network is employed to capture cell mitotic events, which utilizes both appearance and motion information from candidate sequences. Experimental results on realistic low contrast cell images are presented to demonstrate the robustness and validation of the proposed method.