{"title":"TransNUNet:利用注意机制进行全心分割","authors":"Xiaoniu Yang, Xiaolin Tian","doi":"10.1109/ICPECA53709.2022.9719101","DOIUrl":null,"url":null,"abstract":"Cardiac CT segmentation of the whole heart is one of the very complex and key technologies in the auxiliary treatment of cardiovascular diseases. It can help doctors analyze the lesion and other key areas. Because the data of cardiac CT images have the characteristics of small data volume and large size, there are higher requirements for the processing speed and segmentation accuracy of the algorithm. This research proposes a semantic segmentation network based on transformer. This method uses TransUNet as the main architecture, introduces an attention mechanism on its basis, and improves the loss function. Our experiment achieved a Dice score of 0.921 on the dataset from MM-WHS 2017 Challenge, and the results show that the method has good performance.","PeriodicalId":244448,"journal":{"name":"2022 IEEE 2nd International Conference on Power, Electronics and Computer Applications (ICPECA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"TransNUNet: Using Attention Mechanism for Whole Heart Segmentation\",\"authors\":\"Xiaoniu Yang, Xiaolin Tian\",\"doi\":\"10.1109/ICPECA53709.2022.9719101\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Cardiac CT segmentation of the whole heart is one of the very complex and key technologies in the auxiliary treatment of cardiovascular diseases. It can help doctors analyze the lesion and other key areas. Because the data of cardiac CT images have the characteristics of small data volume and large size, there are higher requirements for the processing speed and segmentation accuracy of the algorithm. This research proposes a semantic segmentation network based on transformer. This method uses TransUNet as the main architecture, introduces an attention mechanism on its basis, and improves the loss function. Our experiment achieved a Dice score of 0.921 on the dataset from MM-WHS 2017 Challenge, and the results show that the method has good performance.\",\"PeriodicalId\":244448,\"journal\":{\"name\":\"2022 IEEE 2nd International Conference on Power, Electronics and Computer Applications (ICPECA)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-01-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 IEEE 2nd International Conference on Power, Electronics and Computer Applications (ICPECA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICPECA53709.2022.9719101\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE 2nd International Conference on Power, Electronics and Computer Applications (ICPECA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICPECA53709.2022.9719101","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
TransNUNet: Using Attention Mechanism for Whole Heart Segmentation
Cardiac CT segmentation of the whole heart is one of the very complex and key technologies in the auxiliary treatment of cardiovascular diseases. It can help doctors analyze the lesion and other key areas. Because the data of cardiac CT images have the characteristics of small data volume and large size, there are higher requirements for the processing speed and segmentation accuracy of the algorithm. This research proposes a semantic segmentation network based on transformer. This method uses TransUNet as the main architecture, introduces an attention mechanism on its basis, and improves the loss function. Our experiment achieved a Dice score of 0.921 on the dataset from MM-WHS 2017 Challenge, and the results show that the method has good performance.