Prediction of early-phase cytomegalovirus pneumonia in post-stem cell transplantation using a deep learning model

Article type: Research Article

Authors: Zheng, Yanhua^{a; e} | Ren, Ruilin^{b; d} | Zuo, Teng^f | Chen, Xuan^{b; d} | Li, Hanxuan^{b; d} | Xie, Cheng^{b; d} | Weng, Meiling^{b; d} | He, Chunxiao^{a; b} | Xu, Min^{a; b} | Wang, Lili^c | Li, Nainong^{a; b} | Li, Xiaofan^{a; b; *}

Affiliations: [a] Fujian Provincial Key Laboratory on Hematology, Fujian Institute of Hematology, Fujian Medical University Union Hospital, Fuzhou, China | [b] Department of Hematology, Fujian Medical University Union Hospital, Fuzhou, China | [c] Department of Radiology, Fujian Medical University Union Hospital, Fuzhou, China | [d] School of Basic Medical Sciences, Fujian Medical University, Fuzhou, China | [e] Department of Hematology, The First Hospital of China Medical University, Shenyang, China | [f] Urology Department, Fourth Affiliated Hospital of China Medical University, Shenyang, China

Correspondence: [*] Corresponding author: Xiaofan Li, Fujian Institute of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Medical University Union Hospital, Fuzhou, 350001, Fujian, China. E-mail: [email protected].

Abstract: BACKGROUND: Diagnostic challenges exist for CMV pneumonia in post-hematopoietic stem cell transplantation (post-HSCT) patients, despite early-phase radiographic changes. OBJECTIVE:The study aims to employ a deep learning model distinguishing CMV pneumonia from COVID-19 pneumonia, community-acquired pneumonia, and normal lungs post-HSCT. METHODS:Initially, 6 neural network models were pre-trained with COVID-19 pneumonia, community-acquired pneumonia, and normal lung CT images from Kaggle’s COVID multiclass dataset (Dataset A), then Dataset A was combined with the CMV pneumonia images from our center, forming Dataset B. We use a few-shot transfer learning strategy to fine-tune the pre-trained models and evaluate model performance in Dataset B. RESULTS:34 cases of CMV pneumonia were found between January 2018 and December 2022 post-HSCT. Dataset A contained 1681 images of each subgroup from Kaggle. Combined with Dataset A, Dataset B was initially formed by 98 images of CMV pneumonia and normal lung. The optimal model (Xception) achieved an accuracy of 0.9034. Precision, recall, and F1-score all reached 0.9091, with an AUC of 0.9668 in the test set of Dataset B. CONCLUSIONS:This framework demonstrates the deep learning model’s ability to distinguish rare pneumonia types utilizing a small volume of CT images, facilitating early detection of CMV pneumonia post-HSCT.

Keywords: Cytomegalovirus pneumonia, hematopoietic stem cell transplantation, imaging diagnosis, deep learning, transfer learning

DOI: 10.3233/THC-240597

Journal: Technology and Health Care, vol. 32, no. 5, pp. 3557-3568, 2024