Doyoung Park, Jinsoo Kim, Qi Chang, Shuang Leng, Liang Zhong, Lohendran Baskaran
{"title":"RICAU-Net:用于心脏 CT 中细小稀疏钙化病变分割的残余阻滞启发坐标注意 U 网","authors":"Doyoung Park, Jinsoo Kim, Qi Chang, Shuang Leng, Liang Zhong, Lohendran Baskaran","doi":"arxiv-2409.06993","DOIUrl":null,"url":null,"abstract":"The Agatston score, which is the sum of the calcification in the four main\ncoronary arteries, has been widely used in the diagnosis of coronary artery\ndisease (CAD). However, many studies have emphasized the importance of the\nvessel-specific Agatston score, as calcification in a specific vessel is\nsignificantly correlated with the occurrence of coronary heart disease (CHD).\nIn this paper, we propose the Residual-block Inspired Coordinate Attention\nU-Net (RICAU-Net), which incorporates coordinate attention in two distinct\nmanners and a customized combo loss function for lesion-specific coronary\nartery calcium (CAC) segmentation. This approach aims to tackle the high\nclass-imbalance issue associated with small and sparse lesions, particularly\nfor CAC in the left main coronary artery (LM) which is generally small and the\nscarcest in the dataset due to its anatomical structure. The proposed method\nwas compared with six different methods using Dice score, precision, and\nrecall. Our approach achieved the highest per-lesion Dice scores for all four\nlesions, especially for CAC in LM compared to other methods. The ablation\nstudies demonstrated the significance of positional information from the\ncoordinate attention and the customized loss function in segmenting small and\nsparse lesions with a high class-imbalance problem.","PeriodicalId":501289,"journal":{"name":"arXiv - EE - Image and Video Processing","volume":"26 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"RICAU-Net: Residual-block Inspired Coordinate Attention U-Net for Segmentation of Small and Sparse Calcium Lesions in Cardiac CT\",\"authors\":\"Doyoung Park, Jinsoo Kim, Qi Chang, Shuang Leng, Liang Zhong, Lohendran Baskaran\",\"doi\":\"arxiv-2409.06993\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The Agatston score, which is the sum of the calcification in the four main\\ncoronary arteries, has been widely used in the diagnosis of coronary artery\\ndisease (CAD). However, many studies have emphasized the importance of the\\nvessel-specific Agatston score, as calcification in a specific vessel is\\nsignificantly correlated with the occurrence of coronary heart disease (CHD).\\nIn this paper, we propose the Residual-block Inspired Coordinate Attention\\nU-Net (RICAU-Net), which incorporates coordinate attention in two distinct\\nmanners and a customized combo loss function for lesion-specific coronary\\nartery calcium (CAC) segmentation. This approach aims to tackle the high\\nclass-imbalance issue associated with small and sparse lesions, particularly\\nfor CAC in the left main coronary artery (LM) which is generally small and the\\nscarcest in the dataset due to its anatomical structure. The proposed method\\nwas compared with six different methods using Dice score, precision, and\\nrecall. Our approach achieved the highest per-lesion Dice scores for all four\\nlesions, especially for CAC in LM compared to other methods. The ablation\\nstudies demonstrated the significance of positional information from the\\ncoordinate attention and the customized loss function in segmenting small and\\nsparse lesions with a high class-imbalance problem.\",\"PeriodicalId\":501289,\"journal\":{\"name\":\"arXiv - EE - Image and Video Processing\",\"volume\":\"26 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - EE - Image and Video Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.06993\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Image and Video Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.06993","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
RICAU-Net: Residual-block Inspired Coordinate Attention U-Net for Segmentation of Small and Sparse Calcium Lesions in Cardiac CT
The Agatston score, which is the sum of the calcification in the four main
coronary arteries, has been widely used in the diagnosis of coronary artery
disease (CAD). However, many studies have emphasized the importance of the
vessel-specific Agatston score, as calcification in a specific vessel is
significantly correlated with the occurrence of coronary heart disease (CHD).
In this paper, we propose the Residual-block Inspired Coordinate Attention
U-Net (RICAU-Net), which incorporates coordinate attention in two distinct
manners and a customized combo loss function for lesion-specific coronary
artery calcium (CAC) segmentation. This approach aims to tackle the high
class-imbalance issue associated with small and sparse lesions, particularly
for CAC in the left main coronary artery (LM) which is generally small and the
scarcest in the dataset due to its anatomical structure. The proposed method
was compared with six different methods using Dice score, precision, and
recall. Our approach achieved the highest per-lesion Dice scores for all four
lesions, especially for CAC in LM compared to other methods. The ablation
studies demonstrated the significance of positional information from the
coordinate attention and the customized loss function in segmenting small and
sparse lesions with a high class-imbalance problem.