A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia
IntroductionAccurate identification of the myocardial texture features of fat around the coronary artery on coronary computed tomography angiography (CCTA) images are crucial to improve clinical diagnostic efficiency of myocardial ischemia (MI). However, current coronary CT examination is difficult...
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-02-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2024.1327912/full |
| _version_ | 1797301670206504960 |
|---|---|
| author | Jing Wang Jing Xu Jing Xu Jingsong Mao Suzhong Fu Suzhong Fu Haowei Gu Haowei Gu Naiming Wu Guoqing Su Zhiping Lin Kaiyue Zhang Yuetong Lin Yang Zhao Gang Liu Hengyu Zhao Qingliang Zhao Qingliang Zhao Qingliang Zhao |
| author_facet | Jing Wang Jing Xu Jing Xu Jingsong Mao Suzhong Fu Suzhong Fu Haowei Gu Haowei Gu Naiming Wu Guoqing Su Zhiping Lin Kaiyue Zhang Yuetong Lin Yang Zhao Gang Liu Hengyu Zhao Qingliang Zhao Qingliang Zhao Qingliang Zhao |
| author_sort | Jing Wang |
| collection | DOAJ |
| description | IntroductionAccurate identification of the myocardial texture features of fat around the coronary artery on coronary computed tomography angiography (CCTA) images are crucial to improve clinical diagnostic efficiency of myocardial ischemia (MI). However, current coronary CT examination is difficult to recognize and segment the MI characteristics accurately during earlier period of inflammation.Materials and methodsWe proposed a random forest model to automatically segment myocardium and extract peripheral fat features. This hybrid machine learning (HML) model is integrated by CCTA images and clinical data. A total of 1,316 radiomics features were extracted from CCTA images. To further obtain the features that contribute the most to the diagnostic model, dimensionality reduction was applied to filter features to three: LNS, GFE, and WLGM. Moreover, statistical hypothesis tests were applied to improve the ability of discriminating and screening clinical features between the ischemic and non-ischemic groups.ResultsBy comparing the accuracy, recall, specificity and AUC of the three models, it can be found that HML had the best performance, with the value of 0.848, 0.762, 0.704 and 0.729.ConclusionIn sum, this study demonstrates that ML-based radiomics model showed good predictive value in MI, and offer an enhanced tool for predicting prognosis with greater accuracy. |
| first_indexed | 2024-03-07T23:24:55Z |
| format | Article |
| id | doaj.art-963b39aa3998456481d124eede29e74b |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-03-07T23:24:55Z |
| publishDate | 2024-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-963b39aa3998456481d124eede29e74b2024-02-21T05:02:49ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2024-02-011110.3389/fcvm.2024.13279121327912A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemiaJing Wang0Jing Xu1Jing Xu2Jingsong Mao3Suzhong Fu4Suzhong Fu5Haowei Gu6Haowei Gu7Naiming Wu8Guoqing Su9Zhiping Lin10Kaiyue Zhang11Yuetong Lin12Yang Zhao13Gang Liu14Hengyu Zhao15Qingliang Zhao16Qingliang Zhao17Qingliang Zhao18Department of Imaging, School of Medicine, Xiamen Cardiovascular Hospital of Xiamen University, Xiamen University, Xiamen, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaState Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, National Innovation Platform for Industry-Education Integration in Vaccine Research, Xiamen University, Xiamen, ChinaDepartment of Vascular Intervention, Affiliated Hospital, Guilin Medical University, Guilin, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaState Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, National Innovation Platform for Industry-Education Integration in Vaccine Research, Xiamen University, Xiamen, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaState Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, National Innovation Platform for Industry-Education Integration in Vaccine Research, Xiamen University, Xiamen, ChinaDepartment of Imaging, School of Medicine, Xiamen Cardiovascular Hospital of Xiamen University, Xiamen University, Xiamen, ChinaDepartment of Radiology, Xiang’an Hospital of Xiamen University, Xiamen, ChinaDepartment of Pharmaceutical Diagnosis, GE Healthcare, Guangzhou, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaDepartment of Mechanical and Electrical Engineering, Xiamen University, Xiamen, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaDepartment of Imaging, School of Medicine, Xiamen Cardiovascular Hospital of Xiamen University, Xiamen University, Xiamen, ChinaState Key Laboratory of Vaccines for Infectious Diseases, Center for Molecular Imaging and Translational Medicine, Xiang An Biomedicine Laboratory, Institute of Artificial Intelligence, School of Public Health, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, ChinaState Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, National Innovation Platform for Industry-Education Integration in Vaccine Research, Xiamen University, Xiamen, ChinaShenzhen Research Institute of Xiamen University, Shenzhen, ChinaIntroductionAccurate identification of the myocardial texture features of fat around the coronary artery on coronary computed tomography angiography (CCTA) images are crucial to improve clinical diagnostic efficiency of myocardial ischemia (MI). However, current coronary CT examination is difficult to recognize and segment the MI characteristics accurately during earlier period of inflammation.Materials and methodsWe proposed a random forest model to automatically segment myocardium and extract peripheral fat features. This hybrid machine learning (HML) model is integrated by CCTA images and clinical data. A total of 1,316 radiomics features were extracted from CCTA images. To further obtain the features that contribute the most to the diagnostic model, dimensionality reduction was applied to filter features to three: LNS, GFE, and WLGM. Moreover, statistical hypothesis tests were applied to improve the ability of discriminating and screening clinical features between the ischemic and non-ischemic groups.ResultsBy comparing the accuracy, recall, specificity and AUC of the three models, it can be found that HML had the best performance, with the value of 0.848, 0.762, 0.704 and 0.729.ConclusionIn sum, this study demonstrates that ML-based radiomics model showed good predictive value in MI, and offer an enhanced tool for predicting prognosis with greater accuracy.https://www.frontiersin.org/articles/10.3389/fcvm.2024.1327912/fullcoronary atherosclerosiscoronary CT angiographyradiomicsrandom forest modelmedical image analysis |
| spellingShingle | Jing Wang Jing Xu Jing Xu Jingsong Mao Suzhong Fu Suzhong Fu Haowei Gu Haowei Gu Naiming Wu Guoqing Su Zhiping Lin Kaiyue Zhang Yuetong Lin Yang Zhao Gang Liu Hengyu Zhao Qingliang Zhao Qingliang Zhao Qingliang Zhao A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia Frontiers in Cardiovascular Medicine coronary atherosclerosis coronary CT angiography radiomics random forest model medical image analysis |
| title | A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia |
| title_full | A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia |
| title_fullStr | A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia |
| title_full_unstemmed | A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia |
| title_short | A novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia |
| title_sort | novel hybrid machine learning model for auxiliary diagnosing myocardial ischemia |
| topic | coronary atherosclerosis coronary CT angiography radiomics random forest model medical image analysis |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2024.1327912/full |
| work_keys_str_mv | AT jingwang anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingxu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingxu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingsongmao anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT suzhongfu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT suzhongfu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT haoweigu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT haoweigu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT naimingwu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT guoqingsu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT zhipinglin anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT kaiyuezhang anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT yuetonglin anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT yangzhao anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT gangliu anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT hengyuzhao anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT qingliangzhao anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT qingliangzhao anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT qingliangzhao anovelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingwang novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingxu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingxu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT jingsongmao novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT suzhongfu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT suzhongfu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT haoweigu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT haoweigu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT naimingwu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT guoqingsu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT zhipinglin novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT kaiyuezhang novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT yuetonglin novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT yangzhao novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT gangliu novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT hengyuzhao novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT qingliangzhao novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT qingliangzhao novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia AT qingliangzhao novelhybridmachinelearningmodelforauxiliarydiagnosingmyocardialischemia |