Differential diagnosis of pulmonary Talaromyces marneffei infection and pulmonary tuberculosis in HIV/AIDS patients using a CT-based multimodal 2.5D deep learning approach
- 1Department of Tuberculosis, The Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
- 2Clinical Medical School, Guangxi Medical University, Nanning, Guangxi, China
- 3Department of Radiology, The Fourth People’s Hospital of Nanning, Nanning, Guangxi, China
Abstract
BackgroundTo investigate the diagnostic value of a computed tomography (CT)-based multimodal 2.5-dimensional (2.5D) deep learning model for differentiating Talaromyces marneffei (TSM) from pulmonary tuberculosis (PTB) in patients with human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/AIDS).MethodsA total of 380 HIV/AIDS patients with pulmonary infections who were admitted to the Fourth People’s Hospital of Nanning from January 2018 to August 2025 were retrospectively enrolled, including 203 patients with PTB, 177 with TSM. The cohort was randomly divided into a training set and a test set at a ratio of 7:3. A clinical model, a radiomics model, and a 2.5D deep learning model were constructed separately.ResultsIn the clinical feature model, Random Forest achieved the optimal performance, with a test set AUC of 0.856. In the radiomics model, XGBoost attained the highest AUC of 0.866. In the 2.5D deep learning model, ResNet18 performed best at the slice level with an AUC of 0.865, and the patient-level support vector machine (SVM) model based on MIL achieved an AUC of 0.885. The multimodal combined model integrating clinical, radiomics, and deep learning features achieved the highest AUC of 0.931 and a specificity of 0.981. It significantly outperformed the clinical model (P = 0.020) and the radiomics model (P = 0.008), but the improvement over DL25D was not statistically significant (ΔAUC = 0.046; P = 0.116). The high specificity of the combined model should be interpreted alongside its sensitivity (0.717) and the selected decision threshold.ConclusionThe CT-based multimodal 2.5D deep learning model effectively integrates clinical indicators, radiomics features, and deep learning representations, and substantially improves the differentiation of TSM from PTB in HIV/AIDS patients.
Related
- Rapid diagnosis of pulmonary tuberculosis on induced sputum in children in a high burden setting for HIV and malnutrition
- Caputo fractional-order differential equations for tuberculosis transmission and relapse dynamics with empirical validation
- Hydropneumothorax as a Diagnostic Clue to Pulmonary Tuberculosis With Pleural Involvement: A Case Report
- Determinants of Anticipated Video-Supported Treatment Acceptability in Drug-Susceptible Pulmonary Tuberculosis
- Prediction Model for Detecting Active Pulmonary Tuberculosis in Patients With Head and Neck Cancer Prior to Concurrent Chemoradiotherapy
BibTeX
@article{song2026differential,
title = {{Differential diagnosis of pulmonary Talaromyces marneffei infection and pulmonary tuberculosis in HIV/AIDS patients using a CT-based multimodal 2.5D deep learning approach}},
author = {Chang Song and Chang Song and Chun-Yan Zhao and Chun-Yan Zhao and Xiao-Shi Lin and Zhen-Tao Huang and Xue-Wen Huang and Xue-Wen Huang and Shu-Lin Song and Zhou-Hua Xie and Qing-Dong Zhu},
journal = {Frontiers in Cellular and Infection Microbiology},
year = {2026},
month = oct,
doi = {10.3389/fcimb.2026.1927286},
url = {https://doi.org/10.3389/fcimb.2026.1927286},
note = {CC BY},
}
RIS
TY - JOUR
TI - Differential diagnosis of pulmonary Talaromyces marneffei infection and pulmonary tuberculosis in HIV/AIDS patients using a CT-based multimodal 2.5D deep learning approach
AU - Chang Song
AU - Chang Song
AU - Chun-Yan Zhao
AU - Chun-Yan Zhao
AU - Xiao-Shi Lin
AU - Zhen-Tao Huang
AU - Xue-Wen Huang
AU - Xue-Wen Huang
AU - Shu-Lin Song
AU - Zhou-Hua Xie
AU - Qing-Dong Zhu
PY - 2026
DA - 2026/10/23
JO - Frontiers in Cellular and Infection Microbiology
DO - 10.3389/fcimb.2026.1927286
UR - https://doi.org/10.3389/fcimb.2026.1927286
AB - BackgroundTo investigate the diagnostic value of a computed tomography (CT)-based multimodal 2.5-dimensional (2.5D) deep learning model for differentiating Talaromyces marneffei (TSM) from pulmonary tuberculosis (PTB) in patients with human immunodeficiency virus infection/acquired immunodeficiency syndrome (HIV/AIDS).MethodsA total of 380 HIV/AIDS patients with pulmonary infections who were admitted to the Fourth People’s Hospital of Nanning from January 2018 to August 2025 were retrospectively enrolled, including 203 patients with PTB, 177 with TSM. The cohort was randomly divided into a training set and a test set at a ratio of 7:3. A clinical model, a radiomics model, and a 2.5D deep learning model were constructed separately.ResultsIn the clinical feature model, Random Forest achieved the optimal performance, with a test set AUC of 0.856. In the radiomics model, XGBoost attained the highest AUC of 0.866. In the 2.5D deep learning model, ResNet18 performed best at the slice level with an AUC of 0.865, and the patient-level support vector machine (SVM) model based on MIL achieved an AUC of 0.885. The multimodal combined model integrating clinical, radiomics, and deep learning features achieved the highest AUC of 0.931 and a specificity of 0.981. It significantly outperformed the clinical model (P = 0.020) and the radiomics model (P = 0.008), but the improvement over DL25D was not statistically significant (ΔAUC = 0.046; P = 0.116). The high specificity of the combined model should be interpreted alongside its sensitivity (0.717) and the selected decision threshold.ConclusionThe CT-based multimodal 2.5D deep learning model effectively integrates clinical indicators, radiomics features, and deep learning representations, and substantially improves the differentiation of TSM from PTB in HIV/AIDS patients.
SN - 2235-2988
ER -