Residual Conditional Variational Autoencoder for Multi-Center PET/CT Radiomic Feature Harmonization with Integrated Modeling of Batch Effects and Clinical Covariates.
Authors
Affiliations (10)
Affiliations (10)
- Department of Electrical and Electronic Engineering, Faculty of Engineering, Universiti Putra Malaysia, Serdang, Malaysia.
- Department of Biomedical Engineering, Chengde Medical University, Chengde City, Hebei Province, China.
- Department of Nursing, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, Serdang, Malaysia.
- Department of Nursing, Chengde Central Hospital, Chengde City, Hebei Province, China.
- Department of Radiology, The Affiliated Hospital of Chengde Medical University, Chengde City, Hebei Province, China.
- Institute of Artificial Intelligence, University of Science and Technology Beijing, Beijing, China.
- Department of Electrical and Electronic Engineering, Faculty of Engineering, Universiti Putra Malaysia, Serdang, Malaysia. [email protected].
- Department of Biomedical Engineering, Chengde Medical University, Chengde City, Hebei Province, China. [email protected].
- Hebei Key Laboratory of Nerve Injury and Repair, Chengde Medical University, Chengde City, Hebei, China. [email protected].
- Hebei International Research Center of Medical Engineering, Chengde Medical University, Hebei, China. [email protected].
Abstract
This study proposes a Residual Conditional Variational Autoencoder model (ResCVAE-Harmonizer) that integrates batch information and clinical covariates for multi-center feature harmonization and systematically and comprehensively evaluates its harmonization performance. This study collected 806 cases from 9 different centers. After preprocessing, three types of features were extracted from PET and CT images: low-dimensional radiomic features, high-dimensional radiomic features, and deep learning features based on 3D-DenseNet-121. Each feature type was harmonized using ComBat, CovBat, and the proposed ResCVAE-Harmonizer. Both harmonized and original features were included in a comprehensive evaluation framework comprising variance homogeneity analysis, multi-center classification test, and downstream task effectiveness evaluation. The ResCVAE-Harmonizer significantly improved cross-center feature consistency. Levene's test results showed a general reduction in - log<sub>10</sub>(p) values after harmonization, with more pronounced improvements observed in low- and high-dimensional radiomic features. In center classification tasks, ResCVAE-harmonized features demonstrated greater stability across four classifiers and outperformed the original features. For the downstream survival prediction task, PET deep learning features processed by ResCVAE achieved the highest C-index (0.8920, 95% CI 0.8514-0.9325), surpassing those of the original features (0.8765), ComBat (0.8909), and CovBat (0.8455). Similarly, the C-index for CT deep features improved to 0.8296 (95% CI 0.7715-0.8877). Kaplan-Meier survival stratification based on ResCVAE features showed clearer separation between high- and low-risk groups, with statistically significant log-rank test results. While slightly inferior to ComBat in linear variance consistency, ResCVAE-Harmonizer effectively eliminated both linear and nonlinear batch effects and significantly enhanced survival prediction performance, demonstrating strong research potential.