{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T18:05:03Z","timestamp":1785175503761,"version":"3.55.0"},"reference-count":35,"publisher":"Public Library of Science (PLoS)","issue":"7","license":[{"start":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T00:00:00Z","timestamp":1785110400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100002341","name":"Academy of Finland","doi-asserted-by":"publisher","award":["339614"],"award-info":[{"award-number":["339614"]}],"id":[{"id":"10.13039\/501100002341","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Missing data in medical datasets poses significant challenges for developing effective AI\/ML pipelines. Inaccurate imputation can lead to biased results, reduced model performance, and compromised clinical insights. Understanding how different imputation methods affect AI\/ML model performance is crucial for ensuring accurate clinical findings.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study systematically investigates the effects of different imputation methods on AI\/ML model performance and their clinical implications.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We investigate the impact of six different missing data strategies on the performance of common classification algorithms for analyzing medical data. The performance was evaluated based on sensitivity and specificity metrics for the tasks of predicting COVID-19 diagnosis and patient deterioration. We also perform feature analysis to understand the clinical implications that the choice of imputation method has.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec004\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The findings reveal that the effects of imputation depend on the clinical setting. For a general screening cohort (Einstein Data4u), multivariate imputation by chained equations (MICE) yielded the best performance in clinical settings, resulting in a 26% improvement in sensitivity compared to baseline methods and unmasking critical viral coinfections. Conversely, in an intensive care cohort (MIMIC-IV), complete-case analysis initially showed higher raw predictive metrics. However, further analysis demonstrates that this stems from selection bias driven by informative missingness (MNAR), where testing patterns are intrinsically tied to patient severity. Thus, while imputation recovers diagnostic signals in sparse screening data, it serves as a crucial tool for reducing bias in high-acuity settings.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec005\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This study demonstrates the critical impact of missing data imputation on AI\/ML model performance and the resulting clinical insights. Our findings underscore the importance of selecting appropriate imputation techniques tailored to the specific characteristics of medical data to ensure accurate and reliable AI\/ML predictions. By utilizing a rigorous cross-validation pipeline and a systematic comparison, we provide insights for selecting the most appropriate imputation methods for clinical decision-making applications.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0320105","type":"journal-article","created":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T17:41:18Z","timestamp":1785174078000},"page":"e0320105","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Mission imputable: Effects of missing data processing on infectious disease detection and prognosis"],"prefix":"10.1371","volume":"21","author":[{"given":"Suravi Saha","family":"Roy","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ngoc Thi","family":"Nguyen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Agustin","family":"Zuniga","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5750-5793","authenticated-orcid":true,"given":"Fatemeh","family":"Sarhaddi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eemil","family":"Lagerspetz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huber","family":"Flores","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Petteri","family":"Nurmi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,7,27]]},"reference":[{"issue":"1","key":"pone.0320105.ref001","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s12935-021-01981-1","article-title":"Clinical Applications of Artificial Intelligence and Machine Learning in Cancer Diagnosis: Looking into the Future","volume":"21","author":"MJ Iqbal","year":"2021","journal-title":"Cancer Cell International"},{"issue":"9","key":"pone.0320105.ref002","article-title":"The Role of Artificial Intelligence and Machine Learning in Cardiovascular Imaging and Diagnosis","volume":"16","author":"S Reza-Soltani","year":"2024","journal-title":"Cureus"},{"issue":"9","key":"pone.0320105.ref003","doi-asserted-by":"crossref","first-page":"749","DOI":"10.1016\/S2213-2600(16)30200-4","article-title":"Lung Cancer Screening: Latest Developments and Unanswered Questions","volume":"4","author":"CM Van der Aalst","year":"2016","journal-title":"The Lancet Respiratory Medicine"},{"issue":"5","key":"pone.0320105.ref004","doi-asserted-by":"crossref","first-page":"7","DOI":"10.5120\/ijca2017916020","article-title":"Prediction of Onset Diabetes using Machine Learning Techniques","volume":"180","author":"Md Aminul","year":"2017","journal-title":"IJCA"},{"issue":"5","key":"pone.0320105.ref005","doi-asserted-by":"crossref","DOI":"10.1016\/S1470-2045(19)30149-4","article-title":"Big data and machine learning algorithms for health-care delivery","volume":"20","author":"KY Ngiam","year":"2019","journal-title":"Lancet Oncol"},{"issue":"1","key":"pone.0320105.ref006","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1016\/j.artmed.2013.01.003","article-title":"Missing data in medical databases: impute, delete or classify?","volume":"58","author":"F Cismondi","year":"2013","journal-title":"Artif Intell Med"},{"key":"pone.0320105.ref007","first-page":"100","volume-title":"Missing Data Imputation: Do Advanced ML\/DL Techniques Outperform Traditional Approaches? In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases","author":"Y Zhou","year":"2024"},{"issue":"2","key":"pone.0320105.ref008","doi-asserted-by":"crossref","first-page":"143","DOI":"10.18196\/jrc.v3i2.13133","article-title":"Systematic Review on Missing Data Imputation Techniques with Machine Learning Algorithms for Healthcare","volume":"3","author":"AR Ismail","year":"2022","journal-title":"Journal of Robotics and Control"},{"issue":"1","key":"pone.0320105.ref009","doi-asserted-by":"crossref","first-page":"188","DOI":"10.1186\/s12874-024-02310-6","article-title":"Identify the most appropriate imputation method for handling missing values in clinical structured datasets: a systematic review","volume":"24","author":"M Afkanpour","year":"2024","journal-title":"BMC Med Res Methodol"},{"key":"pone.0320105.ref010","doi-asserted-by":"crossref","first-page":"100720","DOI":"10.1016\/j.cosrev.2024.100720","article-title":"Multimodal missing data in healthcare: A comprehensive review and future directions","volume":"56","author":"LP Le","year":"2025","journal-title":"Computer Science Review"},{"key":"pone.0320105.ref011","doi-asserted-by":"crossref","DOI":"10.1109\/ACCESS.2020.3010287","article-title":"Can AI help in screening viral and COVID-19 pneumonia","volume":"8","author":"ME Chowdhury","year":"2020","journal-title":"IEEE Access"},{"issue":"7","key":"pone.0320105.ref012","first-page":"13","article-title":"Chest X-Ray Findings in 636 Ambulatory Patients with COVID-19 Presenting to an Urgent Care Center: A Normal Chest X-Ray Is no Guarantee","volume":"14","author":"MB Weinstock","year":"2020","journal-title":"Journal Urgent Care Med"},{"issue":"8","key":"pone.0320105.ref013","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1007\/s10916-020-01597-4","article-title":"Detection of COVID-19 Infection from Routine Blood Exams with Machine Learning: A Feasibility Study","volume":"44","author":"D Brinati","year":"2020","journal-title":"J Med Syst"},{"key":"pone.0320105.ref014","doi-asserted-by":"crossref","first-page":"104502","DOI":"10.1016\/j.jcv.2020.104502","article-title":"A predictive tool for identification of SARS-CoV-2 PCR-negative emergency department patients using routine test results","volume":"129","author":"RP Joshi","year":"2020","journal-title":"J Clin Virol"},{"issue":"1","key":"pone.0320105.ref015","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1038\/s41746-021-00383-x","article-title":"Predicting COVID-19 mortality with electronic medical records","volume":"4","author":"H Estiri","year":"2021","journal-title":"NPJ Digit Med"},{"key":"pone.0320105.ref016","unstructured":"Hospital Israelita Albert Einstein, Sao Paulo, Brazil. Diagnosis of COVID-19 and its clinical spectrum, 3\/2020. Einstein Data4u; 2020. https:\/\/www.kaggle.com\/datasets\/einsteindata4u\/covid19"},{"key":"pone.0320105.ref017","unstructured":"Johnson A, Bulgarelli L, Pollard T, Gow B, Moody B, Horng S, et al. MIMIC-IV. PhysioNet. 2024; doi: 10.13026\/kpb9-mt58"},{"issue":"7","key":"pone.0320105.ref018","doi-asserted-by":"crossref","first-page":"1095","DOI":"10.1515\/cclm-2020-0398","article-title":"Routine blood tests as a potential diagnostic tool for COVID-19","volume":"58","author":"D Ferrari","year":"2020","journal-title":"Clin Chem Lab Med"},{"issue":"4","key":"pone.0320105.ref019","first-page":"707","article-title":"Comparison of hematocrit\/hemoglobin ratios in subjects with alpha-thalassemia, with subjects having chronic kidney disease and normal subjects","volume":"44","author":"S Insiripong","year":"2013","journal-title":"Southeast Asian J Trop Med Public Health"},{"key":"pone.0320105.ref020","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1016\/j.amsu.2020.09.044","article-title":"Mortality prediction model for the triage of COVID-19, pneumonia, and mechanically ventilated ICU patients: A retrospective study","volume":"59","author":"L Ryan","year":"2020","journal-title":"Ann Med Surg (Lond)"},{"key":"pone.0320105.ref021","unstructured":"Norazian RMN, Yahaya AS, Ramli NA, Yusof NFFM, Abdullah MMA. Roles of Imputation Methods for Filling the Missing Values: A Review. Advances in Environmental Biology. 2013;7(SPEC. ISSUE 12):3861\u20139."},{"key":"pone.0320105.ref022","article-title":"Statistical analysis with missing data","author":"RJ Little","year":"2019","edition":"3"},{"issue":"3","key":"pone.0320105.ref023","first-page":"343","article-title":"Strategies for dealing with missing data in clinical trials: from design to analysis","volume":"86","author":"JD Dziura","year":"2013","journal-title":"Yale J Biol Med"},{"issue":"12","key":"pone.0320105.ref024","doi-asserted-by":"crossref","first-page":"1255","DOI":"10.1093\/oxfordjournals.aje.a117592","article-title":"A critical look at methods for handling missing covariates in epidemiologic regression analyses","volume":"142","author":"S Greenland","year":"1995","journal-title":"Am J Epidemiol"},{"issue":"12","key":"pone.0320105.ref025","doi-asserted-by":"crossref","first-page":"486","DOI":"10.3390\/metabo10120486","article-title":"A Workflow for Missing Values Imputation of Untargeted Metabolomics Data","volume":"10","author":"T Faquih","year":"2020","journal-title":"Metabolites"},{"issue":"4","key":"pone.0320105.ref026","doi-asserted-by":"crossref","first-page":"372","DOI":"10.1177\/1094428114548590","article-title":"Missing data: Five practical guidelines","volume":"17","author":"DA Newman","year":"2014","journal-title":"Organizational research methods"},{"key":"pone.0320105.ref027","unstructured":"Yoon J, Jordon J, Schaar M. Gain: Missing data imputation using generative adversarial nets. In: International conference on machine learning. PMLR; 2018. p. 5689\u201398."},{"key":"pone.0320105.ref028","doi-asserted-by":"crossref","unstructured":"Caruana R, Karampatziakis N, Yessenalina A. An empirical evaluation of supervised learning in high dimensions. In: Proceedings of the 25th international conference on Machine learning - ICML \u201908, 2008. 96\u2013103. https:\/\/doi.org\/10.1145\/1390156.1390169","DOI":"10.1145\/1390156.1390169"},{"key":"pone.0320105.ref029","doi-asserted-by":"crossref","unstructured":"Du M, Liu N, Hu X. Techniques for Interpretable Machine Learning; 2019.","DOI":"10.1145\/3359786"},{"key":"pone.0320105.ref030","doi-asserted-by":"crossref","first-page":"307","DOI":"10.3389\/fpubh.2017.00307","article-title":"Sensitivity, specificity, and predictive values: foundations, pliabilities, and pitfalls in research and practice","volume":"5","author":"R Trevethan","year":"2017","journal-title":"Frontiers in Public Health"},{"key":"pone.0320105.ref031","first-page":"4768","article-title":"A Unified Approach to Interpreting Model Predictions","volume-title":"Advances in Neural Information Processing Systems","author":"SM Lundberg","year":"2017"},{"issue":"7825","key":"pone.0320105.ref032","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1038\/s41586-020-2649-2","article-title":"Array programming with NumPy","volume":"585","author":"CR Harris","year":"2020","journal-title":"Nature"},{"key":"pone.0320105.ref033","unstructured":"McKinney W. Pandas: A Foundational Python Library for Data Analysis and Statistics; 2011."},{"key":"pone.0320105.ref034","unstructured":"Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research. 2012;12."},{"issue":"3","key":"pone.0320105.ref035","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1109\/MCSE.2007.55","article-title":"Matplotlib: A 2D Graphics Environment","volume":"9","author":"JD Hunter","year":"2007","journal-title":"Comput Sci Eng"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0320105","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T17:41:25Z","timestamp":1785174085000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0320105"}},"subtitle":[],"editor":[{"given":"Bilal","family":"Alatas","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,7,27]]},"references-count":35,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,27]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0320105","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,27]]}}}