Deep Learning in Prediction of Late Major Bleeding After Transcatheter Aortic Valve Replacement

Yuheng Jia; Gaden Luosang; Yiming Li; Jianyong Wang; Pengyu Li; Tianyuan Xiong; Yijian Li; Yanbiao Liao; Zhengang Zhao; Yong Peng; Yuan Feng; Weili Jiang; Wenjian Li; Xinpei Zhang; Zhang Yi; Mao Chen

doi:10.2147/CLEP.S333147

Deep Learning in Prediction of Late Major Bleeding After Transcatheter Aortic Valve Replacement

Clin Epidemiol. 2022 Jan 12:14:9-20. doi: 10.2147/CLEP.S333147. eCollection 2022.

Authors

Yuheng Jia^#¹, Gaden Luosang^#^{2

3}, Yiming Li¹, Jianyong Wang², Pengyu Li⁴, Tianyuan Xiong¹, Yijian Li¹, Yanbiao Liao¹, Zhengang Zhao¹, Yong Peng¹, Yuan Feng¹, Weili Jiang², Wenjian Li², Xinpei Zhang², Zhang Yi², Mao Chen¹

Affiliations

¹ Department of Cardiology, West China Hospital, Sichuan University, Chengdu, Sichuan, People's Republic of China.
² Machine Intelligence Laboratory, College of Computer Science, Sichuan University, Chengdu, Sichuan, People's Republic of China.
³ Department of Information Science and Technology, Tibet University, Lhasa City, People's Republic of China.
⁴ West China Medical School, Sichuan University, Chengdu, Sichuan, People's Republic of China.

^# Contributed equally.

Abstract

Purpose: Late major bleeding is one of the main complications after transcatheter aortic valve replacement (TAVR). We aimed to develop a risk prediction model based on deep learning to predict major or life-threatening bleeding complications (MLBCs) after TAVR.

Patients and methods: This was a retrospective study including TAVR patients from West China Hospital of Sichuan University Transcatheter Aortic Valve Replacement Registry (ChiCTR2000033419) between April 17, 2012 and May 27, 2020. A deep learning-based model named BLeNet was developed with 56 features covering baseline, procedural, and post-procedural characteristics. The model was validated with the bootstrap method and evaluated using Harrell's concordance index (c-index), receiver operating characteristics (ROC) curve, calibration curve, and Kaplan-Meier estimate. Captum interpretation library was applied to identify feature importance. The BLeNet model was compared with the traditional Cox proportional hazard (Cox-PH) model and the random survival forest model in the metrics mentioned above.

Results: The BLeNet model outperformed the Cox-PH and random survival forest models significantly in discrimination [optimism-corrected c-index of BLeNet vs Cox-PH vs random survival forest: 0.81 (95% CI: 0.79-0.92) vs 0.72 (95% CI: 0.63-0.77) vs 0.70 (95% CI: 0.61-0.74)] and calibration (integrated calibration index of BLeNet vs Cox-PH vs random survival forest: 0.007 vs 0.015 vs 0.019). In Kaplan-Meier analysis, BLeNet model had great performance in stratifying high- and low-bleeding risk patients (p < 0.0001).

Conclusion: Deep learning is a feasible way to build prediction models concerning TAVR prognosis. A dedicated bleeding risk prediction model was developed for TAVR patients to facilitate well-informed clinical decisions.

Keywords: deep learning; major or life-threatening bleeding complications; prediction model; transcatheter aortic valve replacement.

Grants and funding

This work was supported by the National Major Science and Technology Projects (grant number 2018AAA0100201) to ZY; the National Natural Science Foundation of China (grant 81970325 to MC; grant number 61906127 to JW); the 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University to MC; the Science and Technology Achievement Transformation Fund of West China Hospital of Sichuan University (CGZH19009) to MC.