TY - GEN
T1 - A Multi-modality Network for Cardiomyopathy Death Risk Prediction with CMR Images and Clinical Information
AU - Xia, Chaoyang
AU - Li, Xiaojie
AU - Wang, Xin
AU - Kong, Bin
AU - Chen, Yucheng
AU - Yin, Youbing
AU - Cao, Kunlin
AU - Song, Qi
AU - Lyu, Siwei
AU - Wu, Xi
N1 - Publisher Copyright: © 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Dilated Cardiomyopathy (DCM) is one of the main worldwide causes of sudden cardiac death (SCD). Early diagnostics significantly increases the chances of correct treatment and survival. However, there are no efficient methods for mortality risk prediction from learning cardiac magnetic resonance (CMR) image and clinical data due to the poor image quality and extreme imbalanced datasets. To solve this problem, we proposed an effective multi-modality network (MMNet) for mortality risk prediction in DCM, and we firstly directly optimize the AUC to train the multimodal deep learning classifier by maximizing the WMW statistic. This can achieve significant improvements in AUC, especially under the imbalanced learning problem. MMNet consists of two branches: clinical data branch and T1 mapping CMR images branch, which allows the model to learn more comprehensive features and makes a more accurate prediction. We validated our approach on a DCM dataset, which contains 450 CMR images that only holds 34 positive samples. Experimental results show that our approach archived accuracy of 98.89%, AUC of 99.61, sensitivity of 100% and specificity of 98.8%, demonstrating the effectiveness of the proposed method.
AB - Dilated Cardiomyopathy (DCM) is one of the main worldwide causes of sudden cardiac death (SCD). Early diagnostics significantly increases the chances of correct treatment and survival. However, there are no efficient methods for mortality risk prediction from learning cardiac magnetic resonance (CMR) image and clinical data due to the poor image quality and extreme imbalanced datasets. To solve this problem, we proposed an effective multi-modality network (MMNet) for mortality risk prediction in DCM, and we firstly directly optimize the AUC to train the multimodal deep learning classifier by maximizing the WMW statistic. This can achieve significant improvements in AUC, especially under the imbalanced learning problem. MMNet consists of two branches: clinical data branch and T1 mapping CMR images branch, which allows the model to learn more comprehensive features and makes a more accurate prediction. We validated our approach on a DCM dataset, which contains 450 CMR images that only holds 34 positive samples. Experimental results show that our approach archived accuracy of 98.89%, AUC of 99.61, sensitivity of 100% and specificity of 98.8%, demonstrating the effectiveness of the proposed method.
KW - AUC optimization
KW - Cross-modality medical data
KW - Dilated cardiomyopathy
UR - https://www.scopus.com/pages/publications/85075662784
U2 - 10.1007/978-3-030-32245-8_64
DO - 10.1007/978-3-030-32245-8_64
M3 - Conference contribution
SN - 9783030322441
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 577
EP - 585
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2019 - 22nd International Conference, Proceedings
A2 - Shen, Dinggang
A2 - Yap, Pew-Thian
A2 - Liu, Tianming
A2 - Peters, Terry M.
A2 - Khan, Ali
A2 - Staib, Lawrence H.
A2 - Essert, Caroline
A2 - Zhou, Sean
PB - Springer Science and Business Media Deutschland GmbH
T2 - 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Y2 - 13 October 2019 through 17 October 2019
ER -