{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,28]],"date-time":"2026-01-28T03:28:35Z","timestamp":1769570915778,"version":"3.49.0"},"reference-count":32,"publisher":"BMJ","issue":"7","license":[{"start":{"date-parts":[[2024,7,5]],"date-time":"2024-07-05T00:00:00Z","timestamp":1720137600000},"content-version":"unspecified","delay-in-days":4,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"Special Foundation for Postdoctoral Research Projects of Chongqing","award":["2021XM3080"],"award-info":[{"award-number":["2021XM3080"]}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Open"],"accepted":{"date-parts":[[2024,6,25]]},"published-print":{"date-parts":[[2024,7]]},"abstract":"<jats:sec><jats:title>Objective<\/jats:title><jats:p>Cellulitis is the most common cause of skin-related hospitalisations, and the mortality of patients with sepsis remains high. Some stratification models have been developed, but their performance in external validation has been unsatisfactory. This study was designed to develop and compare different models for predicting patients with cellulitis developing sepsis during hospitalisation.<\/jats:p><\/jats:sec><jats:sec><jats:title>Design<\/jats:title><jats:p>This is a retrospective cohort study.<\/jats:p><\/jats:sec><jats:sec><jats:title>Setting<\/jats:title><jats:p>This study included both the development and the external-validation phases from two independent large cohorts internationally.<\/jats:p><\/jats:sec><jats:sec><jats:title>Participants and methods<\/jats:title><jats:p>A total of 6695 patients with cellulitis in the Medical Information Mart for Intensive care (MIMIC)-IV database were used to develop models with different machine-learning algorithms. The best models were selected and then externally validated in 2506 patients with cellulitis from the YiduCloud database of our university. The performances and robustness of selected models were further compared in the external-validation group by area under the curve (AUC), diagnostic accuracy, sensitivity, specificity and diagnostic OR.<\/jats:p><\/jats:sec><jats:sec><jats:title>Primary outcome measures<\/jats:title><jats:p>The primary outcome of interest in this study was the development based on the Sepsis-3.0 criteria during hospitalisation.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>Patient characteristics were significantly different between the two groups. In internal validation, XGBoost was the best model, with an AUC of 0.780, and AdaBoost was the worst model, with an AUC of 0.585. In external validation, the AUC of the artificial neural network (ANN) model was the highest, 0.830, while the AUC of the logistic regression (LR) model was the lowest, 0.792. The AUC values changed less in the boosting and ANN models than in the LR model when variables were deleted.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>Boosting and neural network models performed slightly better than the LR model and were more robust in complex clinical situations. The results could provide a tool for clinicians to detect hospitalised patients with cellulitis developing sepsis early.<\/jats:p><\/jats:sec>","DOI":"10.1136\/bmjopen-2024-084183","type":"journal-article","created":{"date-parts":[[2024,7,5]],"date-time":"2024-07-05T14:57:05Z","timestamp":1720191425000},"page":"e084183","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":7,"title":["Development and external validation of machine learning-based models to predict patients with cellulitis developing sepsis during hospitalisation"],"prefix":"10.1136","volume":"14","author":[{"given":"Xilingyuan","family":"Chen","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Li","family":"Hu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4952-9274","authenticated-orcid":false,"given":"Rentao","family":"Yu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"239","published-online":{"date-parts":[[2024,7,5]]},"reference":[{"key":"2024070507401471000_14.7.e084183.1","doi-asserted-by":"crossref","first-page":"ITC17","DOI":"10.7326\/AITC201802060","article-title":"Cellulitis and soft tissue infections","volume":"168","author":"Bystritsky","year":"2018","journal-title":"Ann Intern Med"},{"key":"2024070507401471000_14.7.e084183.2","doi-asserted-by":"publisher","DOI":"10.1001\/archinte.168.14.1585"},{"key":"2024070507401471000_14.7.e084183.3","doi-asserted-by":"crossref","first-page":"1656","DOI":"10.1016\/j.joms.2017.01.032","article-title":"The burden of facial cellulitis leading to inpatient hospitalization","volume":"75","author":"Abramowicz","year":"2017","journal-title":"J Oral Maxillofac Surg"},{"key":"2024070507401471000_14.7.e084183.4","doi-asserted-by":"publisher","DOI":"10.2310\/7750.2013.13075"},{"key":"2024070507401471000_14.7.e084183.5","doi-asserted-by":"crossref","first-page":"1553","DOI":"10.1007\/s11606-018-4546-z","article-title":"Do patients with cellulitis need to be hospitalized? A systematic review and meta-analysis of mortality rates of inpatients with cellulitis","volume":"33","author":"Gunderson","year":"2018","journal-title":"J Gen Intern Med"},{"key":"2024070507401471000_14.7.e084183.6","doi-asserted-by":"crossref","first-page":"578","DOI":"10.1001\/jamadermatol.2017.0159","article-title":"Clinical characteristics and outcomes of patients with cellulitis requiring intensive care","volume":"153","author":"Cranendonk","year":"2017","journal-title":"JAMA Dermatol"},{"key":"2024070507401471000_14.7.e084183.7","unstructured":"Fulton R , Doherty L , Gill D , et al . Guidelines on the management of cellulitis in adults. Northern Ireland: Crest, 2005."},{"key":"2024070507401471000_14.7.e084183.8","doi-asserted-by":"crossref","first-page":"160","DOI":"10.7861\/clinmedicine.18-2-160","article-title":"Diagnosis and management of cellulitis","volume":"18","author":"Sullivan","year":"2018","journal-title":"Clin Med (Lond)"},{"key":"2024070507401471000_14.7.e084183.9","doi-asserted-by":"publisher","DOI":"10.1016\/j.jaad.2016.12.044"},{"key":"2024070507401471000_14.7.e084183.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.jaad.2018.06.062"},{"key":"2024070507401471000_14.7.e084183.11","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMc1906060"},{"key":"2024070507401471000_14.7.e084183.12","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbx044"},{"key":"2024070507401471000_14.7.e084183.13","doi-asserted-by":"publisher","DOI":"10.7326\/M14-0698"},{"key":"2024070507401471000_14.7.e084183.14","unstructured":"Johnson ABL , Pollard T , Horng S , et al . MIMIC-IV (version 2.0). Physionet. 2022. doi:10.13026\/7vcr-e114"},{"key":"2024070507401471000_14.7.e084183.15","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2016.0287"},{"key":"2024070507401471000_14.7.e084183.16","doi-asserted-by":"publisher","DOI":"10.1159\/000521288"},{"key":"2024070507401471000_14.7.e084183.17","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-019-2663-7"},{"key":"2024070507401471000_14.7.e084183.18","doi-asserted-by":"crossref","unstructured":"Ibrahim LF , Hopper SM , Donath S , et al . Development and validation of a cellulitis risk score: the Melbourne ASSET score. Pediatrics 2019;143. doi:10.1542\/peds.2018-1420","DOI":"10.1542\/peds.2018-1420"},{"key":"2024070507401471000_14.7.e084183.19","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1080\/23744235.2020.1726447","article-title":"Non-purulent skin and soft tissue infections: predictive power of a severity score and the appropriateness of treatment in a prospective cohort","volume":"52","author":"Rath","year":"2020","journal-title":"Infect Dis (Lond)"},{"key":"2024070507401471000_14.7.e084183.20","doi-asserted-by":"crossref","first-page":"200","DOI":"10.1093\/jac\/dky400","article-title":"Association of the Dundee severity classification with mortality, length of stay and readmission in adult inpatients with cellulitis","volume":"74","author":"Cutfield","year":"2019","journal-title":"J Antimicrob Chemother"},{"key":"2024070507401471000_14.7.e084183.21","doi-asserted-by":"publisher","DOI":"10.1093\/jac\/dkq362"},{"key":"2024070507401471000_14.7.e084183.22","doi-asserted-by":"crossref","first-page":"859","DOI":"10.1093\/cid\/ciab181","article-title":"The impact of the Auckland cellulitis pathway on length of hospital stay, mortality readmission rate, and antibiotic stewardship","volume":"73","author":"Ritchie","year":"2021","journal-title":"Clin Infect Dis"},{"key":"2024070507401471000_14.7.e084183.23","doi-asserted-by":"crossref","first-page":"303","DOI":"10.1016\/j.jormas.2021.07.007","article-title":"Biological factors predicting the length of hospital stay in odontogenic cellulitis","volume":"123","author":"B\u00e8gue","year":"2022","journal-title":"J Stomatol Oral Maxillofac Surg"},{"key":"2024070507401471000_14.7.e084183.24","doi-asserted-by":"crossref","first-page":"626","DOI":"10.1016\/j.jaad.2016.11.063","article-title":"Clinical characteristics associated with days to discharge among patients admitted with a primary diagnosis of lower limb cellulitis","volume":"76","author":"Garg","year":"2017","journal-title":"J Am Acad Dermatol"},{"key":"2024070507401471000_14.7.e084183.25","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1016\/S0165-9936(04)00307-3","article-title":"Robustness of models developed by multivariate calibration","volume":"23","author":"Zeaiter","year":"2004","journal-title":"TrAC Trends Anal Chem"},{"key":"2024070507401471000_14.7.e084183.26","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1109\/RBME.2020.3013489","article-title":"Secure and robust machine learning for healthcare: a survey","volume":"14","author":"Qayyum","year":"2021","journal-title":"IEEE Rev Biomed Eng"},{"key":"2024070507401471000_14.7.e084183.27","doi-asserted-by":"crossref","first-page":"656","DOI":"10.1001\/jamapediatrics.2013.1329","article-title":"Missing data and multiple imputation","volume":"167","author":"Cummings","year":"2013","journal-title":"JAMA Pediatr"},{"key":"2024070507401471000_14.7.e084183.28","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.psych.58.110405.085530"},{"key":"2024070507401471000_14.7.e084183.29","doi-asserted-by":"crossref","first-page":"2219","DOI":"10.1021\/acs.chemrestox.2c00189","article-title":"Effects of class imbalance and data scarcity on the performance of binary classification machine learning models developed based on ToxCast\/Tox21 assay data","volume":"35","author":"Kim","year":"2022","journal-title":"Chem Res Toxicol"},{"key":"2024070507401471000_14.7.e084183.30","doi-asserted-by":"crossref","first-page":"1525","DOI":"10.1093\/jamia\/ocac093","article-title":"The harm of class imbalance corrections for risk prediction models: illustration and simulation using logistic regression","volume":"29","author":"van den Goorbergh","year":"2022","journal-title":"J Am Med Inform Assoc"},{"key":"2024070507401471000_14.7.e084183.31","doi-asserted-by":"crossref","first-page":"1299","DOI":"10.1161\/CIRCULATIONAHA.121.053733","article-title":"Trusting magic: interpretability of predictions from machine learning algorithms","volume":"143","author":"Rosenberg","year":"2021","journal-title":"Circulation"},{"key":"2024070507401471000_14.7.e084183.32","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-020-03179-9"}],"container-title":["BMJ Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjopen-2024-084183","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,7,5]],"date-time":"2024-07-05T14:57:20Z","timestamp":1720191440000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjopen.bmj.com\/lookup\/doi\/10.1136\/bmjopen-2024-084183"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7]]},"references-count":32,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2024,7,5]]},"published-print":{"date-parts":[[2024,7]]}},"alternative-id":["10.1136\/bmjopen-2024-084183"],"URL":"https:\/\/doi.org\/10.1136\/bmjopen-2024-084183","relation":{},"ISSN":["2044-6055","2044-6055"],"issn-type":[{"value":"2044-6055","type":"print"},{"value":"2044-6055","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,7]]}}}