{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T09:13:07Z","timestamp":1785402787767,"version":"3.56.0"},"reference-count":26,"publisher":"MDPI AG","issue":"15","license":[{"start":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T00:00:00Z","timestamp":1785369600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: This study investigated the impact of feature type, feature dimensionality, and feature selection strategies on intensive care unit (ICU) mortality prediction using first-day portable frontal chest radiographs and demographic data. In addition, the study evaluated whether comparable predictive performance could be achieved using reduced, clinically interpretable feature sets. Methods: A total of 500 patients were included, comprising 400 cases for model development and 100 independent cases for testing. Two complementary experimental frameworks were evaluated. In Experiment Set-1, a clinically interpretable feature set consisting of 12 radiographic and demographic variables was analyzed. In Experiment Set-2, a high-dimensional feature space consisting of 74 radiographic, radiological, radiomic, and image-derived features was investigated using six feature selection methods: ANOVA, Chi-Square, Kruskal\u2013Wallis, MRMR, ReliefF, and Shapley-based importance analysis. Machine learning models were evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, and F1-score. Results: In Experiment Set-1, feature selection reduced the number of predictors from 12 to 4. Using the complete 12-feature set, the Bagged Trees classifier achieved the highest performance (AUC = 0.95). Following feature reduction, the Subspace KNN classifier achieved an AUC of 0.96 with an accuracy of 0.82, indicating that satisfactory predictive performance could be achieved using a substantially smaller feature set. In Experiment Set-2, comparison of six feature selection strategies demonstrated that MRMR and Kruskal\u2013Wallis produced the highest-performing feature subsets. Cobb angle, bilateral infiltrates, bilateral pleural effusion, and unilateral pleural effusion were consistently identified across all feature selection methods as stable predictors associated with ICU mortality. These predictors were considered stable because they were consistently identified across multiple independent feature selection methods despite differences in the underlying selection algorithms. Conclusions: The findings suggest that appropriate feature selection strategies may reduce model complexity while retaining clinically relevant predictive information. The consistent identification of a small group of radiographic features across multiple feature selection methods in the 74-feature experimental framework suggests that these variables represent robust and clinically meaningful predictors of ICU mortality. Further external validation using larger multicenter datasets is required before clinical implementation.<\/jats:p>","DOI":"10.3390\/diagnostics16152391","type":"journal-article","created":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T08:09:42Z","timestamp":1785398982000},"page":"2391","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative Evaluation of Feature Selection Strategies for ICU Mortality Prediction Using Chest Radiographic, Radiomic, and Demographic Features"],"prefix":"10.3390","volume":"16","author":[{"given":"Orhan","family":"Gok","sequence":"first","affiliation":[{"name":"Department of Electrical and Electronics Engineering, Faculty of Engineering,  Sakarya University, Sakarya 54050, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1399-8044","authenticated-orcid":false,"given":"T\u00fcrker Fedai","family":"\u00c7avu\u015f","sequence":"additional","affiliation":[{"name":"Department of Electrical and Electronics Engineering, Faculty of Engineering,  Sakarya University, Sakarya 54050, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Omer","family":"Ozdemir","sequence":"additional","affiliation":[{"name":"Department of Electrical and Electronics Engineering, Faculty of Engineering, Sakarya University of Applied Sciences, Sakarya 54187, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7725-707X","authenticated-orcid":false,"given":"Ahmed Cihad","family":"Gen\u00e7","sequence":"additional","affiliation":[{"name":"Ahmed Cihad Genc Private Clinic, Istanbul 34480, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6768-7973","authenticated-orcid":false,"given":"Selcuk","family":"Yaylac\u0131","sequence":"additional","affiliation":[{"name":"Department of Internal Medicine, Faculty of Medicine, Sakarya University, Sakarya 54100, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4211-7958","authenticated-orcid":false,"given":"La\u00e7in","family":"Tatl\u0131 Ayhan","sequence":"additional","affiliation":[{"name":"Department of Radiology, Faculty of Medicine, Sakarya University, Sakarya 54100, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,30]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Lau, V.I., Hammal, F., Sligl, W.I., Karvellas, C.J., Kutsogiannis, D.J., Zampieri, F.G., Wilcox, M.E., Fiest, K.M., Niven, D.J., and Stelfox, H.T. (2025). Intensive care unit and hospital mortality for non-COVID critically ill patients before, and during the COVID-19 pandemic in Alberta hospitals: Retrospective, observational cohort study. medRxiv.","DOI":"10.1101\/2025.09.25.25336571"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"e2660","DOI":"10.47243\/jos.2660","article-title":"Assessment of mortality in intensive care: A critical look at risk factors and care practices","volume":"6","author":"Sayar","year":"2025","journal-title":"J. Orig. Stud."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"562","DOI":"10.1097\/MLR.0000000000001878","article-title":"Prediction of in-hospital mortality among intensive care unit patients using modified daily Laboratory-based Acute Physiology Scores, version 2 (LAPS2)","volume":"61","author":"Kohn","year":"2023","journal-title":"Med. Care"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Broder, J. (2011). Imaging the chest: The chest radiograph. Diagnostic Imaging for the Emergency Physician, Elsevier.","DOI":"10.1016\/B978-1-4160-6113-7.10005-5"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"40","DOI":"10.25259\/JCIS_76_2020","article-title":"Performance of radiologists in the evaluation of the chest radiography with the use of a \u201cnew software score\u201d in coronavirus disease 2019 pneumonia suspected patients","volume":"10","author":"Bagnera","year":"2020","journal-title":"J. Clin. Imaging Sci."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Matusik, P.S., Popiela, T.J., and Matusik, P.T. (2025). Novel Indexes in the Assessment of Cardiac Enlargement Using Chest Radiography: A New Look at an Old Problem. J. Clin. Med., 14.","DOI":"10.3390\/jcm14030942"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Polis, B., Zawadzka-Fabijan, A., Fabijan, R., Kosi\u0144ska, R., Nowostawska, E., and Fabijan, A. (2025). Exploring BiomedCLIP\u2019s capabilities in medical image analysis: A focus on scoliosis detection and severity assessment. Appl. Sci., 15.","DOI":"10.3390\/app15010398"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Fabijan, A., Fabijan, R., Zawadzka-Fabijan, A., Nowos\u0142awska, E., Zakrzewski, K., and Polis, B. (2023). Evaluating scoliosis severity based on posturographic X-ray images using a contrastive language\u2013image pretraining model. Diagnostics, 13.","DOI":"10.3390\/diagnostics13132142"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e220158","DOI":"10.1148\/ryai.220158","article-title":"Conquering the Cobb angle: A deep learning algorithm for automated, hardware-invariant measurement of Cobb angle on radiographs in patients with scoliosis","volume":"5","author":"Suri","year":"2023","journal-title":"Radiol. Artif. Intell."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"S1","DOI":"10.3233\/THC-248021","article-title":"Comparative analysis of supervised learning algorithms for prediction of cardiovascular diseases","volume":"32","author":"Dou","year":"2024","journal-title":"Technol. Health Care"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Garc\u00eda, M.V., Bouza-Rodr\u00edguez, J.-B., and Comesa\u00f1a-Campos, A. (2025). Convolutional Neural Network-Based Approach for Cobb Angle Measurement Using Mask R-CNN. Diagnostics, 15.","DOI":"10.20944\/preprints202503.0980.v1"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Zhu, T., Xu, K., Son, W., Linton-Reid, K., Boubnovski-Martell, M., Grech-Sollars, M., Lain, A.D., and Posma, J.M. (PLoS Digit. Health, 2025). Designing a computer-assisted diagnosis system for cardiomegaly detection and radiology report generation, PLoS Digit. Health, Advance online publication.","DOI":"10.1101\/2024.09.02.24311997"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Gok, O., Cavus, T.F., Genc, A.C., Yaylaci, S., and Ayhan, L.T. (2025). Mortality prediction from patient\u2019s first day PAAC radiograph in internal medicine intensive care unit using artificial intelligence methods. Diagnostics, 15.","DOI":"10.3390\/diagnostics15243138"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"460","DOI":"10.1177\/10815589251335327","article-title":"Comprehensive analyses: Using machine learning models for mortality prediction in the intensive care unit of internal medicine","volume":"73","author":"Durmaz","year":"2025","journal-title":"J. Investig. Med."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"E1683","DOI":"10.1503\/cmaj.210083","article-title":"Diagnostic performance of chest radiography measurements for the assessment of cardiac chamber enlargement","volume":"193","author":"Torres","year":"2021","journal-title":"Can. Med. Assoc. J."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"593","DOI":"10.1007\/s00586-009-0913-0","article-title":"A review of methods for quantitative evaluation of spinal curvature","volume":"18","author":"Vrtovec","year":"2009","journal-title":"Eur. Spine J."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Liu, D., Zhang, L., Yang, J., and Lin, A. (2023). Lenke Classification of Scoliosis Based on Segmentation Network and Adaptive Shape Descriptor. Appl. Sci., 13.","DOI":"10.3390\/app13063905"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"158","DOI":"10.1186\/s13244-021-01097-0","article-title":"Limitations of cardiothoracic ratio derived from chest radiographs to predict real heart size: Comparison with magnetic resonance imaging","volume":"12","author":"Simkus","year":"2021","journal-title":"Insights Into Imaging"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"125","DOI":"10.18100\/ijamec.799363","article-title":"Lung Disease Classification Using Machine Learning Algorithms","volume":"8","author":"Aykanat","year":"2020","journal-title":"Int. J. Appl. Math. Electron. Comput."},{"key":"ref_20","unstructured":"(2022). Performance analysis of machine learning algorithms in predicting pulmonary diseases. 2022 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COM-IT-CON), IEEE."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Vairale, P., and Dubey, V. (2025, January 7\u20138). A comprehensive comparison of machine learning classification techniques for early detection of Alzheimer\u2019s disorder. Proceedings of the 2025 4th International Conference on Computing, Communication, and Intelligent Systems (ICCCIT), Indore, India.","DOI":"10.1109\/ICCCIT62592.2025.10927864"},{"key":"ref_22","first-page":"201","article-title":"Radiographic determination of cardiomegaly using cardiothoracic ratio and transverse cardiac diameter: Can one size fit all? Part one","volume":"27","author":"Brakohiapa","year":"2017","journal-title":"Pan Afr. Med. J."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"e000426","DOI":"10.1136\/bmjresp-2019-000426","article-title":"Impact of blood group on survival following critical illness: A single-centre retrospective observational study","volume":"6","author":"Slade","year":"2019","journal-title":"BMJ Open Respir. Res."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Kander, T., Bjurstr\u00f6m, M.F., Frigyesi, A., J\u00f6ud, M., and Nilsson, C.U. (2022). ABO and RhD blood group are not associated with mortality and morbidity in critically ill patients; a multicentre observational study of 29 512 patients. BMC Anesthesiol., 22.","DOI":"10.1186\/s12871-022-01626-4"},{"key":"ref_25","first-page":"S264","article-title":"Abo Blood Types and Mortality Following Critical Illness: A Single Centre Retrospective Observational Study","volume":"29","author":"Reddy","year":"2025","journal-title":"Indian J. Crit. Care Med. Peer-Rev. Off. Publ. Indian Soc. Crit. Care Med."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"146","DOI":"10.1186\/s12967-020-02312-0","article-title":"Comparison and development of machine learning tools for the prediction of chronic obstructive pulmonary disease in the Chinese population","volume":"18","author":"Ma","year":"2020","journal-title":"J. Transl. Med."}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/15\/2391\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T08:28:21Z","timestamp":1785400101000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/15\/2391"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,30]]},"references-count":26,"journal-issue":{"issue":"15","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["diagnostics16152391"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16152391","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,30]]}}}