{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T17:40:38Z","timestamp":1785606038963,"version":"3.56.0"},"reference-count":48,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR AI"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec sec-type=\"background\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>The early detection of clinical deterioration and timely intervention for hospitalized patients can improve patient outcomes. The currently existing early warning systems rely on variables from structured data, such as vital signs and laboratory values, and do not incorporate other potentially predictive data modalities. Because respiratory failure is a common cause of deterioration, chest radiographs are often acquired in patients with clinical deterioration, which may be informative for predicting their risk of intensive care unit (ICU) transfer.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"objective\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study aimed to compare and validate different computer vision models and data augmentation approaches with chest radiographs for predicting clinical deterioration.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"methods\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This retrospective observational study included adult patients hospitalized at the University of Wisconsin Health System between 2009 and 2020 with an elevated electronic cardiac arrest risk triage (eCART) score, a validated clinical deterioration early warning score, on the medical-surgical wards. Patients with a chest radiograph obtained within 48 hours prior to the elevated score were included in this study. Five computer vision model architectures (VGG16, DenseNet121, Vision Transformer, ResNet50, and Inception V3) and four data augmentation methods (histogram normalization, random flip, random Gaussian noise, and random rotate) were compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC) for predicting clinical deterioration (ie, ICU transfer or ward death in the following 24 hours).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"results\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The study included 21,817 patient admissions, of which 1655 (7.6%) experienced clinical deterioration. The DenseNet121 model pretrained on chest radiograph datasets with histogram normalization and random Gaussian noise augmentation had the highest discrimination (AUROC 0.734 and AUPRC 0.414), while the vision transformer having 24 transformer blocks with random rotate augmentation had the lowest discrimination (AUROC 0.598).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec sec-type=\"conclusions\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The study shows the potential of chest radiographs in deep learning models for predicting clinical deterioration. The DenseNet121 architecture pretrained with chest radiographs performed better than other architectures in most experiments, and the addition of histogram normalization with random Gaussian noise data augmentation may enhance the performance of DenseNet121 and pretrained VGG16 architectures.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/67144","type":"journal-article","created":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T00:50:07Z","timestamp":1741567807000},"page":"e67144-e67144","source":"Crossref","is-referenced-by-count":7,"title":["Comparison of Deep Learning Approaches Using Chest Radiographs for Predicting Clinical Deterioration: Retrospective Observational Study"],"prefix":"10.2196","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8546-2970","authenticated-orcid":false,"given":"Mahmudur","family":"Rahman","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6414-130X","authenticated-orcid":false,"given":"Jifan","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5038-6074","authenticated-orcid":false,"given":"Kyle A","family":"Carey","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-0867-3834","authenticated-orcid":false,"given":"Dana P","family":"Edelson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8542-0471","authenticated-orcid":false,"given":"Askar","family":"Afshar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8152-736X","authenticated-orcid":false,"given":"John W","family":"Garrett","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9314-2037","authenticated-orcid":false,"given":"Guanhua","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6368-4652","authenticated-orcid":false,"given":"Majid","family":"Afshar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4030-5250","authenticated-orcid":false,"given":"Matthew M","family":"Churpek","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,4,10]]},"reference":[{"issue":"7-8","key":"R1","doi-asserted-by":"publisher","first-page":"1360","DOI":"10.1111\/jocn.14238","article-title":"Clinical deterioration: a concept analysis","volume":"27","author":"Padilla","journal-title":"J Clin Nurs"},{"issue":"5","key":"R2","doi-asserted-by":"publisher","first-page":"325","DOI":"10.1097\/EJA.0000000000000798","article-title":"Improving detection of patient deterioration in the general hospital ward environment","volume":"35","author":"Vincent","journal-title":"Eur J Anaesthesiol"},{"key":"R3","unstructured":"U.S. food and drug administration. ECARTv5 clinical deterioration suite. URL: https:\/\/www.accessdata.fda.gov\/cdrh_docs\/pdf23\/K233253.pdf [Accessed 25-03-2025]"},{"issue":"10","key":"R4","doi-asserted-by":"publisher","first-page":"521","DOI":"10.1093\/qjmed\/94.10.521","article-title":"Validation of a modified Early Warning Score in medical admissions","volume":"94","author":"Subbe","journal-title":"QJM"},{"issue":"8","key":"R5","doi-asserted-by":"publisher","first-page":"932","DOI":"10.1016\/j.resuscitation.2010.04.014","article-title":"ViEWS--Towards a national early warning score for detecting adult inpatient deterioration","volume":"81","author":"Prytherch","journal-title":"Resuscitation"},{"issue":"4","key":"R6","doi-asserted-by":"publisher","first-page":"465","DOI":"10.1016\/j.resuscitation.2012.12.016","article-title":"The ability of the National Early Warning Score (NEWS) to discriminate patients at risk of early cardiac arrest, unanticipated intensive care unit admission, and death","volume":"84","author":"Smith","journal-title":"Resuscitation"},{"issue":"1","key":"R7","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-020-01144-8","article-title":"Early warning score validation methodologies and performance metrics: a systematic review","volume":"20","author":"Fang","journal-title":"BMC Med Inform Decis Mak"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-32245-8_64","article-title":"A multi-modality network for cardiomyopathy death risk prediction with CMR images and clinical information","volume":"11765","author":"Xia","journal-title":"Medical Image Computing and Computer Assisted Intervention \u2013 MICCAI 2019"},{"key":"R9","doi-asserted-by":"crossref","unstructured":"Rawal K Sethi G Walia GK . Impact of machine learning and deep learning in medical image analysis. In: Rabie K Karthik C Chowdhury S Dutta PK , editors. Deep Learning in Medical Image Processing and Analysis. 2023:187-199. [doi: 10.1049\/PBHE059E_ch11]","DOI":"10.1049\/PBHE059E_ch11"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2021.115141","article-title":"Determination of COVID-19 pneumonia based on generalized convolutional neural network model from chest X-ray images","volume":"180","author":"Alhudhaif","journal-title":"Expert Syst Appl"},{"issue":"5","key":"R11","doi-asserted-by":"publisher","first-page":"2850","DOI":"10.1007\/s10489-020-02055-x","article-title":"Attention-based VGG-16 model for COVID-19 chest X-ray image classification","volume":"51","author":"Sitaula","journal-title":"Appl Intell (Dordr)"},{"issue":"4","key":"R12","doi-asserted-by":"publisher","first-page":"460","DOI":"10.1007\/s10278-017-9980-7","article-title":"Deep convolutional neural networks for endotracheal tube position and X-ray image classification: challenges and opportunities","volume":"30","author":"Lakhani","journal-title":"J Digit Imaging"},{"issue":"6","key":"R13","doi-asserted-by":"publisher","first-page":"3267","DOI":"10.1007\/s11831-024-10081-y","article-title":"Chest X-ray images for lung disease detection using deep learning techniques: a comprehensive survey","volume":"31","author":"Al-qaness","journal-title":"Arch Computat Methods Eng"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.aej.2022.10.053","article-title":"A deep learning architecture for multi-class lung diseases classification using chest X-ray (CXR) images","volume":"64","author":"Alshmrani","journal-title":"Alexandria Engineering Journal"},{"issue":"6","key":"R15","doi-asserted-by":"publisher","first-page":"e340","DOI":"10.1016\/S2589-7500(21)00056-X","article-title":"Deep learning to detect acute respiratory distress syndrome on chest radiographs: a retrospective study with external validation","volume":"3","author":"Sjoding","journal-title":"Lancet Digit Health"},{"issue":"8","key":"R16","doi-asserted-by":"publisher","first-page":"24101","DOI":"10.1007\/s11042-023-16419-1","article-title":"A systematic literature review on deep learning approaches for pneumonia detection using chest X-ray images","volume":"83","author":"Sharma","journal-title":"Multimed Tools Appl"},{"issue":"4","key":"R17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s12559-020-09787-5","article-title":"Pneumonia classification using deep learning from chest X-ray images during COVID-19","volume":"16","author":"Ibrahim","journal-title":"Cognit Comput"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.122129","article-title":"Incremental learning-based cascaded model for detection and localization of tuberculosis from chest x-ray images","volume":"238","author":"Vats","journal-title":"Expert Syst Appl"},{"issue":"2","key":"R19","doi-asserted-by":"publisher","first-page":"104","DOI":"10.1016\/j.imed.2023.06.001","article-title":"Deep learning models for tuberculosis detection and infected region visualization in chest X-ray images","volume":"4","author":"Sharma","journal-title":"Intelligent Medicine"},{"key":"R20","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2022.119430","article-title":"Biphasic majority voting-based comparative COVID-19 diagnosis using chest X-ray images","volume":"216","author":"Sunnetci","journal-title":"Expert Syst Appl"},{"key":"R21","unstructured":"Chen Z et al. A vision-language foundation model to enhance efficiency of chest X-ray interpretation. arXiv. Preprint posted online on2024. URL: http:\/\/arxiv.org\/abs\/2401.12208 [Accessed 25-09-2024] [doi: 10.48550\/arXiv.2401.12208]"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","first-page":"e44","DOI":"10.1016\/S2589-7500(23)00218-2","article-title":"Development and validation of open-source deep neural networks for comprehensive chest x-ray reading: a retrospective, multicentre study","volume":"6","author":"Cid","journal-title":"Lancet Digit Health"},{"key":"R23","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2022.109319","article-title":"Chest X-ray analysis empowered with deep learning: a systematic review","volume":"126","author":"Meedeniya","journal-title":"Appl Soft Comput"},{"key":"R24","unstructured":"Simonyan K Zisserman A . Very deep convolutional networks for large-scale image recognition. arXiv. Preprint posted online on2015. URL: http:\/\/arxiv.org\/abs\/1409.1556 [Accessed 24-09-2024] [doi: 10.48550\/arXiv.1409.1556]"},{"key":"R25","doi-asserted-by":"crossref","unstructured":"Huang G Liu Z Van Der Maaten L Weinberger KQ . Densely connected convolutional networks. Presented at: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Jul 21-26, 2017:2261-2269; Honolulu, HI. [doi: 10.1109\/CVPR.2017.243]","DOI":"10.1109\/CVPR.2017.243"},{"key":"R26","unstructured":"Dosovitskiy A et al. An image is worth 16x16 words: transformers for image recognition at scale. arXiv. Preprint posted online on2021. URL: http:\/\/arxiv.org\/abs\/2010.11929 [Accessed 25-03-2025] [doi: 10.48550\/arXiv.2010.11929]"},{"key":"R27","unstructured":"He K Zhang X Ren S Sun J . Deep residual learning for image recognition. arXiv. Preprint posted online on2015. URL: https:\/\/arxiv.org\/abs\/1512.03385 [Accessed 25-03-2025]"},{"key":"R28","doi-asserted-by":"crossref","unstructured":"Szegedy C Vanhoucke V Ioffe S Shlens J Wojna Z . Rethinking the inception architecture for computer vision. Presented at: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Jun 26 to Jul 1, 2016; Las Vegas, NV, USA. 2016.[doi: 10.1109\/CVPR.2016.308]","DOI":"10.1109\/CVPR.2016.308"},{"issue":"1","key":"R29","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-56478-4","article-title":"Identification of kidney stones in KUB X-ray images using VGG16 empowered with explainable artificial intelligence","volume":"14","author":"Ahmed","journal-title":"Sci Rep"},{"key":"R30","doi-asserted-by":"crossref","unstructured":"Islam M Hannan T Sarker L Ahmed Z . COVID-densenet: A deep learning architecture to detect COVID-19 from chest radiology images. In: Saraswat M Chowdhury C Mandal CK Gandomi AH , editors. Presented at: Proceedings of International Conference on Data Science and Applications; Feb 7, 2023:397-415; [doi: 10.1007\/978-981-19-6634-7_28]","DOI":"10.1007\/978-981-19-6634-7_28"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2021.115519","article-title":"Detection of tuberculosis from chest X-ray images: Boosting the performance with vision transformer and transfer learning","volume":"184","author":"Duong","journal-title":"Expert Syst Appl"},{"key":"R32","doi-asserted-by":"crossref","unstructured":"Deng J Dong W Socher R Li LJ . ImageNet: A large-scale hierarchical image database. Presented at: 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops); Jun 20-25, 2005:248-255; Miami, FL. [doi: 10.1109\/CVPR.2009.5206848]","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"R33","unstructured":"Cohen JP et al. TorchXRayVision: A library of chest X-ray datasets and models. arXiv. Preprint posted online on2021. [doi: 10.48550\/ARXIV.2111.00595]"},{"key":"R34","doi-asserted-by":"crossref","unstructured":"Wang X Peng Y Lu L Lu Z Bagheri M Summers RM . ChestX-ray8: hospital-scale chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. Presented at: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Jul 21-26, 2017:3462-3471; Honolulu, HI. [doi: 10.1109\/CVPR.2017.369]","DOI":"10.1109\/CVPR.2017.369"},{"key":"R35","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101797","article-title":"PadChest: A large chest x-ray image dataset with multi-label annotated reports","volume":"66","author":"Bustos","journal-title":"Med Image Anal"},{"key":"R36","doi-asserted-by":"crossref","unstructured":"Irvin J Rajpurkar P Ko M et al. CheXpert: a large chest radiograph dataset with uncertainty labels and expert comparison. 2019;33(1):590-597. [doi: 10.1609\/aaai.v33i01.3301590]","DOI":"10.1609\/aaai.v33i01.3301590"},{"issue":"1","key":"R37","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-019-0322-0","article-title":"MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports","volume":"6","author":"Johnson","journal-title":"Sci Data"},{"issue":"2","key":"R38","doi-asserted-by":"publisher","first-page":"304","DOI":"10.1093\/jamia\/ocv080","article-title":"Preparing a collection of radiology examinations for distribution and retrieval","volume":"23","author":"Demner-Fushman","journal-title":"J Am Med Inform Assoc"},{"issue":"2","key":"R39","doi-asserted-by":"publisher","first-page":"421","DOI":"10.1148\/radiol.2019191293","article-title":"Chest radiograph interpretation with deep learning models: assessment with radiologist-adjudicated reference standards and population-adjusted evaluation","volume":"294","author":"Majkowska","journal-title":"Radiology"},{"key":"R40","unstructured":"RSNA pneumonia detection challenge. URL: https:\/\/www.kaggle.com\/competitions\/rsna-pneumonia-detection-challenge [Accessed 25-03-2025]"},{"key":"R41","unstructured":"Liaw R Liang E Nishihara R Moritz P Gonzalez JE Stoica I . Tune: A research platform for distributed model selection and training. arXiv. Preprint posted online on2018. [doi: 10.48550\/arXiv.1807.05118]"},{"issue":"3","key":"R42","doi-asserted-by":"publisher","first-page":"837","DOI":"10.2307\/2531595","article-title":"Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach","volume":"44","author":"DeLong","journal-title":"Biometrics"},{"key":"R43","unstructured":"Netflix. GitHub - Netflix\/vmaf: Perceptual video quality assessment based on multi-method fusion. URL: https:\/\/github.com\/Netflix\/vmaf\/ [Accessed 25-03-2025]"},{"key":"R44","doi-asserted-by":"crossref","unstructured":"He K Girshick R Dollar P . Rethinking imagenet pre-training. Presented at: 2019 IEEE\/CVF International Conference on Computer Vision (ICCV); Oct 27 to Nov 2, 2019:4917-4926; Seoul, Korea (South). URL: https:\/\/ieeexplore.ieee.org\/xpl\/mostRecentIssue.jsp?punumber=8972782 [doi: 10.1109\/ICCV.2019.00502]","DOI":"10.1109\/ICCV.2019.00502"},{"key":"R45","doi-asserted-by":"crossref","unstructured":"Lee J Lee EJ . Self-supervised pre-training improves fundus image classification for diabetic retinopathy. In: Kehtarnavaz N Carlsohn MF , editors. Presented at: Real-Time Image Processing and Deep Learning 2022; Apr 3-7, 2022; Orlando, United States. [doi: 10.1117\/12.2632901]","DOI":"10.1117\/12.2632901"},{"issue":"10s","key":"R46","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3505244","article-title":"Transformers in vision: a survey","volume":"54","author":"Khan","journal-title":"ACM Comput Surv"},{"key":"R47","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2406.00237","article-title":"A comparative study of CNN, ResNet, and Vision Transformers for multi-classification of chest diseases","author":"Jain","journal-title":"arXiv"},{"issue":"4","key":"R48","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0265949","article-title":"Pre-processing methods in chest X-ray image classification","volume":"17","author":"Gie\u0142czyk","journal-title":"PLoS ONE"}],"container-title":["JMIR AI"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/ai.jmir.org\/2025\/1\/e67144","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:28:10Z","timestamp":1785536890000},"score":1,"resource":{"primary":{"URL":"https:\/\/ai.jmir.org\/2025\/1\/e67144"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,10]]},"references-count":48,"URL":"https:\/\/doi.org\/10.2196\/67144","relation":{},"ISSN":["2817-1705"],"issn-type":[{"value":"2817-1705","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,10]]},"article-number":"v4i7-8e67144"}}