Chest-X-ray based deep feature-driven framework for anemia detection insights from a large hospital cohort.
Authors
Affiliations (9)
Affiliations (9)
- Department of Internal Medicine, Tri-Service General Hospital, and School of Medicine, National Defense Medical University, Taipei, Taiwan, Republic of China.
- Department of Artificial Intelligence and Internet of Things, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan, Republic of China.
- Medical Technology Education Center, School of Medicine, National Defense Medical University, Taipei, Taiwan, Republic of China.
- Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Tri-Service General Hospital, and School of Medicine, National Defense Medical University, Taipei, Taiwan, Republic of China.
- Department of Education and Research, Taipei City Hospital, Taipei, Taiwan, Republic of China.
- Division of Nephrology, Department of Internal Medicine, Tri-Service General Hospital, Taipei, Taiwan, Republic of China.
- Department of Family and Community Medicine, Tri-Service General Hospital, and School of Medicine, National Defense Medical University, Taipei, Taiwan, Republic of China. [email protected].
- Department of Artificial Intelligence and Internet of Things, Tri-Service General Hospital, National Defense Medical University, Taipei, Taiwan, Republic of China. [email protected].
- Department of Education and Research, Taipei City Hospital, Taipei, Taiwan, Republic of China. [email protected].
Abstract
Anemia is a common global health problem traditionally diagnosed through laboratory hemoglobin testing, which requires invasive blood sampling. This study evaluated the feasibility of using a chest X-ray (CXR)-based deep feature-driven framework for noninvasive anemia prediction. We conducted a retrospective cohort study using data from Tri-Service General Hospital, Taipei, Taiwan, collected between June 2016 and February 2022. A total of 305,793 adults aged ≥ 20 years who had at least one CXR were included. The dataset was divided into development, tuning, internal validation, and external validation sets. A two-stage framework leveraging a pretrained Vision Transformer (ViT-B/32) encoder for feature extraction, followed by a downstream logistic regression classifier, was developed to predict anemia (Hb ≤ 10 g/dL) from CXR images without model fine-tuning. The framework for predicting anemia showed strong performance, with an AUC of 0.845 in the internal validation set (sensitivity 68.5%, specificity 84.7%) and 0.852 in the external set (sensitivity 71.5%, specificity 83.2%). Subgroup analysis revealed better diagnostic accuracy with computed radiography, posteroanterior view, and in males aged 55 to 64, while younger females had lower accuracy. The framework, especially when combined with clinical variables, outperformed other clinical models. This study presents a deep feature extraction approach for opportunistic moderate-to-severe anemia anemia screening using chest X-rays. With further validation, this model may facilitate early identification of anemia in routine clinical practice.