Establishment and validation of an AI-aid method in the diagnosis of myocardial perfusion imaging

BMC Medical Imaging

Research

Yanzhu Bian¹  Yujing Hu¹  Xianghe Liao²  Yan Fan²  Jianming Li³  Xiaojie Wang³  Sen Wang⁴  Jie Shen⁵  Momo Sun⁵  Shen Wang⁶  Xuemei Zhang⁶  Yiming Shen⁶  Ning Li⁶  Peng Wang⁶  Qiang Jia⁶  Ruyi Zhang⁶  Lingyun Xu⁶  Miao Wang⁶  Zhaowei Meng⁶  Weiming Wu⁶  Jian Tan⁶  Xuehui Liu⁷  Jianping Zhang⁷  Chengyu Song⁸  Wei Zhang⁸  He Wang⁸  Wangxiao Li⁸ 

[1] Department of Nuclear Medicine, Hebei General Hospital, Shijiazhuang, China;Department of Nuclear Medicine, Peking University First Hospital, Beijing, China;Department of Nuclear Medicine, Teda International Cardiovascular Hospital, Tianjin, China;Department of Nuclear Medicine, The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital, Zhengzhou, China;Department of Nuclear Medicine, Tianjin First Central Hospital, Tianjin, China;Department of Nuclear Medicine, Tianjin Medical University General Hospital, Anshan Road No. 154, Heping District, 300052, Tianjin, China;Department of Nuclear Medicine, Tianjin Third Central Hospital, Tianjin, China;School of Microelectronics, Tianjin University, Weijin Road No. 92, Nankai District, 300072, Tianjin, China;

关键词: Artificial intelligence (AI);    Machine learning;    Coronary artery disease (CAD);    Myocardial perfusion imaging (MPI);    SPECT/CT;   

DOI  :  10.1186/s12880-023-01037-y

 received in 2022-07-20, accepted in 2023-05-29,  发布年份 2023

来源: Springer

【 摘 要 】

BackgroundThis study aimed to develop and validate an AI (artificial intelligence)-aid method in myocardial perfusion imaging (MPI) to differentiate ischemia in coronary artery disease.MethodsWe retrospectively selected 599 patients who had received gated-MPI protocol. Images were acquired using hybrid SPECT-CT systems. A training set was used to train and develop the neural network and a validation set was used to test the predictive ability of the neural network. We used a learning technique named “YOLO” to carry out the training process. We compared the predictive accuracy of AI with that of physician interpreters (beginner, inexperienced, and experienced interpreters).ResultsTraining performance showed that the accuracy ranged from 66.20% to 94.64%, the recall rate ranged from 76.96% to 98.76%, and the average precision ranged from 80.17% to 98.15%. In the ROC analysis of the validation set, the sensitivity range was 88.9 ~ 93.8%, the specificity range was 93.0 ~ 97.6%, and the AUC range was 94.1 ~ 96.1%. In the comparison between AI and different interpreters, AI outperformed the other interpreters (most P-value < 0.05).ConclusionThe AI system of our study showed excellent predictive accuracy in the diagnosis of MPI protocols, and therefore might be potentially helpful to aid radiologists in clinical practice and develop more sophisticated models.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

13690_2023_1130_Article_IEq34.gif

13690_2023_1130_Article_IEq32.gif

Fig. 5

40517_2023_252_Article_IEq141.gif

Fig. 1

12302_2023_754_Article_IEq4.gif

Fig. 3