Deep Learning Morphological Analysis of Chest Radiographs as Preliminary Predictor for Hospitalization in Patients with Chronic Cardiac Disease
DOI:
https://doi.org/10.22317/jcms.v12i2.2142Keywords:
Chronic Disease, Heart Diseases, Radiography, Thoracic, Signal Processing, Computer-Assisted, Forecasting, Hospitalization, Deep LearningAbstract
Objectives: This study uses deep learning to analyze morphological features of chest radiographs to predict the need for hospitalization in patients with cardiac problems.
Methods: A retrospective case-control approach was designed. A public dataset of chest x rays was used and validated. The multimodal AI model was designed to extract x-ray features at different CNN levels to classify patients into 'admission' and 'No admission' groups.
Results: Several important morphological features from patients attributes and chest radiographs were obtained to predict the need for hospitalization in patients with chronic cardiac disease. These preliminary predictors, belonging to different CNN layers: L3 (layer 3) Texton Density 61, L4 Reflection Symmetry 40, Relative Size Context may be linked to structural changes and severity.
Conclusion: There was a preliminary association between multimodal DL features and personalized hospitalization decisions in patients with chronic cardiac disease.
References
Fine NM, Howlett JG. Chronic Heart Failure. In: Merck Manual Professional Version [Internet]. Kenilworth (NJ): Merck & Co., Inc.; 2025 Oct [cited 2026 Mar 16]. Available from: https://www.merckmanuals.com/professional/cardiovascular-disorders/heart-failure/chronic-heart-failure
Clarivate. Acute Heart Failure – Epidemiology – Epidemiology Dashboard [Internet]. London: Clarivate Plc; 2026 Jan [cited 2026 Mar 16]. Available from: https://clarivate.com/life-sciences-healthcare/report/epidcv0041-biopharma-acute-heart-failure-epidemiology-epidemiology-dashboard/
Goyal P, Kim M, Krishnaswamy A, et al. Noninvasive Cardiac Imaging in Older Adults: Diagnostic Challenges and Prognostic Implications Across Modalities. JACC: Advances. 2025;4(12_Part_1):102284. doi:10.1016/j.jacadv.2025.102284
Mukherjee M, Kizer JR, Ambale-Venkatesh B, et al. Age-Related Changes in Cardiac Structure and Function and the Risk of Heart Failure in Older Adults. JAMA Cardiol. 2025;10(3):245-254. doi:10.1001/jamacardio.2024.5291
Shah AM, Claggett B, Prasad N, et al. Association of Abnormal Cardiac Morphology on Routine Imaging with Incident Heart Failure and Hospitalization in Community-Dwelling Older Adults. Circulation. 2025;151(2):110-122. doi:10.1161/CIRCULATIONAHA.124.071234
He Z, Zhu Y, Li P, Xie J. [A study on the proportion of medical expenses and the rationality of hospitalization cost structure in chronic heart failure]. West China Medical Journal. 2024;39(9):1413-1418. Chinese. doi:10.7507/1002-0179.202406148
Ohuchi H, Hayama Y, Iwasa T, et al. Unplanned Hospitalization Due to All‐Cause Morbidity and Its Real‐World Management Practices in Patients With Fontan Circulation. J Am Heart Assoc. 2025;24(14):e044374. doi:10.1161/JAHA.125.044374
Gorbachev A, Lee S, Park J, et al. Hybrid deep learning framework MedFusionNet assists multilabel biomedical risk stratification from imaging and tabular data. Commun Med. 2026;6:47. doi:10.1038/s43856-025-01334-3
Mahin MIS, Arefin MS, Hasan MT. A Hybrid Machine Learning Approach for Synthetic Data Generation with Post Hoc Calibration for Clinical Tabular Datasets. arXiv [preprint]. 2025. arXiv:2510.10513. Available from: https://arxiv.org/abs/2510.10513
Cho Y, Kim J, Lee SM, et al. Multimodal Deep Learning for Predicting Acute Decompensation in Chronic Heart Failure Using Chest Radiographs and Clinical Data. Nat Commun. 2025;16:2123. doi:10.1038/s41467-025-57344-7
European Society of Cardiology. 2024 ESC Guidelines for the management of chronic coronary syndromes. Eur Heart J. 2024;45(35):3415-3537. doi:10.1093/eurheartj/ehae556
United States Food and Drug Administration. Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices [Internet]. Silver Spring (MD): FDA; 2025 Mar [cited 2026 Mar 16]. Available from: https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-aiml-enabled-medical-devices
Zhu T, Xu K, Son W, et al. Designing a computer-assisted diagnosis system for cardiomegaly detection and radiology report generation. bioRxiv [preprint]. 2025 Jun 25. doi:10.1101/2024.09.02.24311997
Zhang CJ, Lu Y, Tang FQ, Zhang SQ, Cai HP. Modal interaction and contrastive learning for heart failure diagnosis combining chest X-rays and clinical text. Biomed Signal Process Control. 2026;112:108436. doi:10.1016/j.bspc.2025.108436
Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Med Image Anal. 2017;42:60-88. doi:10.1016/j.media.2017.07.005
Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25(1):44-56. doi:10.1038/s41591-018-0300-7
Huang SC, Pareek A, Zamanian R, et al. Multimodal fusion of chest radiographs and clinical data for mortality prediction. Radiology: Artificial Intelligence. 2024;6(3):e230195. doi:10.1148/ryai.230195
Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP. SMOTE: synthetic minority over-sampling technique. J Artif Intell Res. 2002;16:321-357. doi:10.1613/jair.953
Kaggle. Heart Failure Prediction Dataset [Internet]. 2025 [cited 2026 Mar 16]. Available from: https://www.kaggle.com/datasets/
Grinberg M. Flask Web Development: Developing Web Applications with Python. 2nd ed. Sebastopol (CA): O'Reilly Media; 2018.
Gorbachev A, Lee S, Park J, et al. Hybrid deep learning framework MedFusionNet assists multilabel biomedical risk stratification from imaging and tabular data. Commun Med. 2026;6:47. doi:10.1038/s43856-025-01334-3
Kingma DP, Ba J. Adam: A Method for Stochastic Optimization. In: Proceedings of the 3rd International Conference on Learning Representations (ICLR); 2015 May 7-9; San Diego, CA.
National Institutes of Health (NIH) Clinical Center. ChestX-ray14 Dataset [Internet]. Bethesda (MD): NIH; [cited 2026 Mar 16]. Available from: https://nihcc.app.box.com/v/ChestXray-NIHCC/
Xiaosong Wang, Yifan Peng, Le Lu, Zhiyong Lu, Mohammadhadi Bagheri, Ronald M. Summers. ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly- Supervised Classification and Localization of Common Thorax Diseases, IEEE CVPR, pp. 3462-3471,2017.
National Institutes of Health (NIH) Clinical Center. ChestX-ray14 Dataset [Internet]. Bethesda (MD): NIH; [cited 2026 Mar 16]. Available from: https://nihcc.app.box.com/v/ChestXray-NIHCC/
Hussein AK. [Multimodal Deep Learning Model for Medical Imaging Analysis]. 2025.
Shorten C, Khoshgoftaar TM. A survey on Image Data Augmentation for Deep Learning. J Big Data. 2019;6(1):60. doi:10.1186/s40537-019-0197-0
Ambale-Venkatesh B, Liu CY, Liu YC, et al. Association of Myocardial Fibrosis and Cardiovascular Events: The Multi-Ethnic Study of Atherosclerosis. Circulation. 2019;140(25):2031-2042. doi:10.1161/CIRCULATIONAHA.119.041550
Ying X. An Overview of Overfitting and its Solutions. J Phys: Conf Ser. 2019;1168:022022. doi:10.1088/1742-6596/1168/2/022022
Matsumoto T, Kodera S, Shinohara H, et al. Predicting Heart Failure Hospitalization from Chest Radiographs Using a Deep Learning Model. Int J Cardiol. 2024;398:131621. doi:10.1016/j.ijcard.2023.131621
Koo HJ, Lee JG, Kim YJ, et al. Conventional and Deep Learning-Based Radiographic Assessment for Predicting Clinical Outcomes in Patients with Heart Failure. Korean J Radiol. 2025;26(1):32-42. doi:10.3348/kjr.2024.0567
Tjoa E, Guan C. A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI. IEEE Trans Neural Netw Learn Syst. 2021;32(11):4793-4813. doi:10.1109/TNNLS.2020.3027314
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Journal of Contemporary Medical Sciences
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
