{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T02:51:42Z","timestamp":1781664702164,"version":"3.54.5"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T00:00:00Z","timestamp":1781568000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T00:00:00Z","timestamp":1781654400000},"content-version":"vor","delay-in-days":1,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Universit\u00e4tsklinikum Bonn"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Langenbecks Arch Surg"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Machine learning (ML) may support decision-making for acute abdominal pain (AAP), but limited interpretability hinders adoption. We evaluated a random-forest classifier for predicting urgent abdominal surgery and examined how predictive value and feature importance evolve across clinical timepoints.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>In this retrospective single-center study, we included adults presenting with AAP to the emergency department. The outcome was urgent abdominal surgery within 24\u00a0h. Features were grouped into stepwise sets. For each feature set, models were trained in 20 random 80\/20 train\u2013test splits with randomized hyperparameter search. Performance was summarized by AUC ROC and AUC PR, and interpretability by permutation importance and SHapley Additive exPlanations (SHAP).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Among 1,350 patients (median age 43 years, 682 (50.5%) females), 276 (20.4%) underwent urgent surgery. The final model achieved a median AUC ROC of 0.83. Discrimination increased stepwise from basic data (0.53), symptoms (0.61), pain history (0.66), vital signs (0.68), laboratory values (0.76), and physical examination (0.83). Computed tomography added only marginal improvement (0.83). Feature importance shifted from symptoms to vital signs and subsequently laboratory markers, particularly c-reactive Protein, white blood cell count, and prothrombin time, complemented by guarding. SHAP analyses confirmed these trends by consistently linking abnormal laboratory or clinical values to operative outcomes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Interpretable ML enables pre-imaging risk stratification for urgent surgery in AAP. Near-final discrimination is achieved after laboratory testing and physical examination, while computed tomography adds limited value at population level. Timepoint-specific feature contribution may facilitate integration of ML into surgical decision-making.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00423-026-04098-z","type":"journal-article","created":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T18:24:53Z","timestamp":1781634293000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Towards clinically interpretable machine learning in emergency surgery: feature importance and insights across clinical time points in abdominal pain cases"],"prefix":"10.1007","volume":"411","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6608-8381","authenticated-orcid":false,"given":"Jonas","family":"Henn","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Simon","family":"Hatterscheidt","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3098-3928","authenticated-orcid":false,"given":"Svetozar","family":"Nesic","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5175-5559","authenticated-orcid":false,"given":"Sebastian","family":"Nowak","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3095-6356","authenticated-orcid":false,"given":"Wolfgang","family":"Block","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Johannes","family":"R\u00f6ttgen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6976-1735","authenticated-orcid":false,"given":"Ingo","family":"Gr\u00e4ff","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5160-8671","authenticated-orcid":false,"given":"J\u00f6rg C.","family":"Kalff","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1435-9562","authenticated-orcid":false,"given":"Alois M.","family":"Sprinkart","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1889-8632","authenticated-orcid":false,"given":"Andreas","family":"Buness","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hanno","family":"Matthaei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,16]]},"reference":[{"key":"4098_CR1","doi-asserted-by":"publisher","unstructured":"Kirubarajan A, Taher A, Khan S, Masood S (2020) Artificial intelligence in emergency medicine: A scoping review. J Am Coll Emerg Physicians Open 1:1691\u20131702. https:\/\/doi.org\/10.1002\/emp2.12277","DOI":"10.1002\/emp2.12277"},{"key":"4098_CR2","doi-asserted-by":"publisher","unstructured":"Reismann J, Romualdi A, Kiss N, et al (2019) Diagnosis and classification of pediatric acute appendicitis by artificial intelligence methods: An investigator-independent approach. PLOS ONE 14:e0222030. https:\/\/doi.org\/10.1371\/journal.pone.0222030","DOI":"10.1371\/journal.pone.0222030"},{"key":"4098_CR3","doi-asserted-by":"publisher","unstructured":"Bertsimas D, Dunn J, Velmahos GC, Kaafarani HMA (2018) Surgical risk is not linear: derivation and validation of a novel, user-friendly, and machine-learning-based predictive optimal trees in emergency surgery risk (POTTER) calculator. Ann Surg 268:574\u2013583. https:\/\/doi.org\/10.1097\/SLA.0000000000002956","DOI":"10.1097\/SLA.0000000000002956"},{"key":"4098_CR4","doi-asserted-by":"publisher","unstructured":"Loftus TJ, Tighe PJ, Filiberto AC, et al (2020) Artificial intelligence and surgical decision-making. JAMA Surg 155:148. https:\/\/doi.org\/10.1001\/jamasurg.2019.4917","DOI":"10.1001\/jamasurg.2019.4917"},{"key":"4098_CR5","doi-asserted-by":"publisher","unstructured":"Gans SL, Pols MA, Stoker J, et al (2015) Guideline for the diagnostic pathway in patients with acute abdominal pain. Dig Surg 32:23\u201331. https:\/\/doi.org\/10.1159\/000371583","DOI":"10.1159\/000371583"},{"key":"4098_CR6","doi-asserted-by":"publisher","unstructured":"Henn J, Hatterscheidt S, Sahu A, et al (2023) Machine learning for decision-support in acute abdominal pain \u2013 proof of concept and central considerations. Zentralblatt F\u00fcr Chir - Z F\u00fcr Allg Visz Thorax- Gef\u00e4\u00dfchirurgie 148:376\u2013383. https:\/\/doi.org\/10.1055\/a-2125-1559","DOI":"10.1055\/a-2125-1559"},{"key":"4098_CR7","doi-asserted-by":"publisher","unstructured":"Schipper A, Belgers P, O\u2019Connor R, et al (2024) Machine-learning based prediction of appendicitis for patients presenting with acute abdominal pain at the emergency department. World J Emerg Surg 19:40. https:\/\/doi.org\/10.1186\/s13017-024-00570-7","DOI":"10.1186\/s13017-024-00570-7"},{"key":"4098_CR8","doi-asserted-by":"publisher","unstructured":"Ma Y, Luo M, Guan G, et al (2025) An explainable predictive machine learning model of gangrenous cholecystitis based on clinical data: a retrospective single center study. World J Emerg Surg 20:1. https:\/\/doi.org\/10.1186\/s13017-024-00571-6","DOI":"10.1186\/s13017-024-00571-6"},{"key":"4098_CR9","doi-asserted-by":"publisher","unstructured":"Henn J, Vandemeulebroucke T, Hatterscheidt S, et al (2025) German surgeons\u2019 perspective on the application of artificial intelligence in clinical decision-making. Int J Comput Assist Radiol Surg. https:\/\/doi.org\/10.1007\/s11548-025-03326-z","DOI":"10.1007\/s11548-025-03326-z"},{"key":"4098_CR10","doi-asserted-by":"publisher","unstructured":"Busch F, Hoffmann L, Xu L, et al (2025) Multinational attitudes toward ai in health care and diagnostics among hospital patients. JAMA Netw Open 8:e2514452. https:\/\/doi.org\/10.1001\/jamanetworkopen.2025.14452","DOI":"10.1001\/jamanetworkopen.2025.14452"},{"key":"4098_CR11","doi-asserted-by":"publisher","unstructured":"Lameris W, van Randen A, van Es HW, et al (2009) Imaging strategies for detection of urgent conditions in patients with acute abdominal pain: diagnostic accuracy study. BMJ 338:b2431\u2013b2431. https:\/\/doi.org\/10.1136\/bmj.b2431","DOI":"10.1136\/bmj.b2431"},{"key":"4098_CR12","doi-asserted-by":"publisher","unstructured":"Bernett J, Blumenthal DB, Grimm DG, et al (2024) Guiding questions to avoid data leakage in biological machine learning applications. Nat Methods 21:1444\u20131453. https:\/\/doi.org\/10.1038\/s41592-024-02362-y","DOI":"10.1038\/s41592-024-02362-y"},{"key":"4098_CR13","unstructured":"H2O.ai (2022) h2o: R Interface for H2O"},{"key":"4098_CR14","doi-asserted-by":"publisher","unstructured":"Faqar-Uz-Zaman SF, Anantharajah L, Baumartz P, et al (2022) The diagnostic efficacy of an app-based diagnostic health care application in the emergency room: eRadaR-Trial. A prospective, double-blinded, observational study. Ann Surg 276:935\u2013942. https:\/\/doi.org\/10.1097\/SLA.0000000000005614","DOI":"10.1097\/SLA.0000000000005614"},{"key":"4098_CR15","unstructured":"B\u00f6rner N, Kappenberger A-S, Weber S, et al (2025) The Acute Abdomen. Dtsch Arztebl Int 122:137\u2013144"},{"key":"4098_CR16","doi-asserted-by":"publisher","unstructured":"Loftus TJ, Altieri MS, Balch JA, et al (2023) Artificial intelligence\u2013enabled decision support in surgery: state-of-the-art and future directions. Ann Surg 278:51\u201358. https:\/\/doi.org\/10.1097\/SLA.0000000000005853","DOI":"10.1097\/SLA.0000000000005853"},{"key":"4098_CR17","doi-asserted-by":"publisher","unstructured":"Schulwolf S, Brower C, Karam A, et al (2023) Clinical features vs ct findings to estimate need for surgery in small bowel obstruction. JAMA Netw Open 6:e2341376. https:\/\/doi.org\/10.1001\/jamanetworkopen.2023.41376","DOI":"10.1001\/jamanetworkopen.2023.41376"},{"key":"4098_CR18","doi-asserted-by":"publisher","unstructured":"Guo H, Jia Y-J, Li H-L, et al (2026) Impact of CT turnaround time on diagnostic and clinical outcomes in acute abdominal pain: A retrospective cohort study. Am J Emerg Med 99:9\u201316. https:\/\/doi.org\/10.1016\/j.ajem.2025.09.011","DOI":"10.1016\/j.ajem.2025.09.011"},{"key":"4098_CR19","doi-asserted-by":"publisher","unstructured":"Ng CS (2002) Evaluation of early abdominopelvic computed tomography in patients with acute abdominal pain of unknown cause: prospective randomised study. BMJ 325:1387\u20131387. https:\/\/doi.org\/10.1136\/bmj.325.7377.1387","DOI":"10.1136\/bmj.325.7377.1387"},{"key":"4098_CR20","doi-asserted-by":"publisher","unstructured":"Boccatonda A, Brighenti A, Tiraferri V, et al (2025) POCUS for acute abdominal pain: practical scan protocols on gastrointestinal diseases and an evidence review. J Ultrasound 28:851\u2013871. https:\/\/doi.org\/10.1007\/s40477-025-01088-7","DOI":"10.1007\/s40477-025-01088-7"},{"key":"4098_CR21","doi-asserted-by":"publisher","unstructured":"Breidthardt T, Brunner-Schaub N, Balmelli C, et al (2019) Inflammatory biomarkers and clinical judgment in the emergency diagnosis of urgent abdominal pain. Clin Chem 65:302\u2013312. https:\/\/doi.org\/10.1373\/clinchem.2018.296491","DOI":"10.1373\/clinchem.2018.296491"},{"key":"4098_CR22","doi-asserted-by":"publisher","unstructured":"Etzioni DA, Liu JH, Maggard MA, Ko CY (2003) The aging population and its impact on the surgery workforce. Ann Surg 238:170\u2013177. https:\/\/doi.org\/10.1097\/01.SLA.0000081085.98792.3d","DOI":"10.1097\/01.SLA.0000081085.98792.3d"},{"key":"4098_CR23","doi-asserted-by":"publisher","unstructured":"Hoot NR, Aronsky D (2008) Systematic review of emergency department crowding: causes, effects, and solutions. Ann Emerg Med 52:126\u2013136.e1. https:\/\/doi.org\/10.1016\/j.annemergmed.2008.03.014","DOI":"10.1016\/j.annemergmed.2008.03.014"},{"key":"4098_CR24","doi-asserted-by":"publisher","unstructured":"DeCamp M, Tilburt JC (2019) Why we cannot trust artificial intelligence in medicine. Lancet Digit Health 1:e390. https:\/\/doi.org\/10.1016\/S2589-7500(19)30197-9","DOI":"10.1016\/S2589-7500(19)30197-9"},{"key":"4098_CR25","doi-asserted-by":"publisher","unstructured":"Kerasidou C (Xaroula), Kerasidou A, Buscher M, Wilkinson S (2022) Before and beyond trust: reliance in medical AI. J Med Ethics 48:852\u2013856. https:\/\/doi.org\/10.1136\/medethics-2020-107095","DOI":"10.1136\/medethics-2020-107095"}],"container-title":["Langenbeck's Archives of Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00423-026-04098-z","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00423-026-04098-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00423-026-04098-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T01:54:05Z","timestamp":1781661245000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00423-026-04098-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,16]]},"references-count":25,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["4098"],"URL":"https:\/\/doi.org\/10.1007\/s00423-026-04098-z","relation":{},"ISSN":["1435-2451"],"issn-type":[{"value":"1435-2451","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,16]]},"assertion":[{"value":"28 November 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the University Hospital Bonn (163\/23-EP). No dedicated informed consent was required for this retrospective study according to local legislation.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"JCK is an associate Editor of Langenbeck\u2019s Archives of Surgery. Besides this, the authors declare that they have no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"165"}}