Explainable Machine Learning Predictive Models for Surgical Site Infections: Scoping Review
Erklärbare Machine-Learning-Prädiktionsmodelle für chirurgische Wundinfektionen: Scoping Review
Journal of Medical Internet Research · 30.09.2026 · 28 · e94617-e94617
DOI: 10.2196/94617 ↗Auf einen Blick
Die Scoping-Review kartierte erklärbare ML-Modelle zur Vorhersage chirurgischer Wundinfektionen bei Erwachsenen. Eingeschlossen waren 77 Studien mit 98 Modellen; die Methoden und wichtigen Prädiktoren waren heterogen.
Fragestellung
Welche erklärbaren ML-Modelle zur Vorhersage chirurgischer Wundinfektionen gibt es aus klinischer Perspektive, und wie werden strukturierte und unstrukturierte Daten genutzt?
Methode
Scoping Review nach PRISMA-ScR und JBI; 6 Datenbanken, Suche ab Januar 2010, letzte Suche am 14. Juli 2026; Qualitätsbewertung mit PROBAST+AI.
Ergebnisse
77 Studien mit 98 ML-Modellen. 72/98 (73,5 %) waren prognostisch, 26 diagnostisch. 51,9 % (40/77) untersuchten zusammengesetzte SSI-Vorhersagen; 40/98 waren Black-Box-Modelle mit Post-hoc-Erklärungen.
Einschränkungen
Unvollständige Berichterstattung zu Merkmalsdefinitionen und Erklärungslogik; begrenzte klinische Relevanz schränkte die Interpretation und Anwendbarkeit ein.
Schlussfolgerung der Autor:innen
Die Übersicht zeigte wiederkehrende Prädiktormuster und erhebliche methodische Heterogenität. Gefordert werden clinician-informierte Berichtsrahmen sowie Validierung von Erklärungstreue und klinischer Relevanz.
Kernaussagen
- 77 Studien mit 98 Modellen wurden eingeschlossen
- SHAP plus Ensemble-Lernen war der häufigste Post-hoc-Ansatz
- Berichterstattung und klinische Relevanz waren oft unzureichend
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Scoping Review in der Zeitschrift Journal of Medical Internet Research, online erschienen am 30.09.2026. Verfasst von Rui Sun und 6 weiteren Autor:innen. Beteiligte Einrichtung: Public Health Emergency Center, Chinese Center for Disease Control and Prevention, Beijing, China. Hauptthemen laut MEDLINE: Machine Learning, Surgical Wound Infection. Die Arbeit nennt 195 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Background
Surgical site infections (SSIs) remain a major cause of health care–associated infections, and early prediction is essential for improving patient outcomes. Machine learning (ML) has shown potential for SSI prediction; however, clinical implementation requires models that are both accurate and explainable. Despite recent progress in explainable ML, its clinical application to SSI prediction remains limited.
Objective
This study aimed to map explainable ML models for SSI prediction from a clinical perspective and examine their use of structured and unstructured data across the dimensions of data, methodology, and explanation output.
Methods
We conducted a scoping review following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) and Joanna Briggs Institute (JBI) guidance, and registered the protocol in PROSPERO. Six databases were searched for eligible studies published from January 2010 onward, without language restrictions. The search was conducted on August 9, 2025, and updated on July 14, 2026. We included studies that developed or validated an explainable ML model for predicting SSI in adults. Two reviewers (RS and YL) independently screened studies and extracted data. Findings were narratively synthesized and presented in evidence maps. Methodological quality was assessed using the PROBAST+AI (Prediction model Risk Of Bias Assessment Tool for prediction models using regression or artificial intelligence) methods.
Results
Overall, 77 studies reporting 98 ML models were included. Most models were prognostic (72/98, 73.5%), whereas 26 focused on postoperative SSI diagnosis. A total of 81.8% (63/77) of studies addressed a single surgical specialty, most commonly gastrointestinal surgery (27/63, 42.9%). Overall, 51.9% (40/77) of studies addressed composite SSI predictions. Among all models, % (40/98) were black-box models explained by post hoc methods; SHAP combined with ensemble learning was the leading approach (18/40, 45%). Regression models accounted for half of the inherently interpretable models, interpreted using coefficients. Prognostic models commonly included health and lifestyle (60/72, 83.3%), individual characteristics, and surgical process details (both 57/72, 79.2%); health and lifestyle factors were most frequently important across SSI types. Diagnostic models commonly included surgical process details (13/26, 50%), administrative codes, and individual characteristics (both 10/26, 38.5%). Key diagnostic predictors varied by SSI types: postoperative clinical interventions predominated for composite SSI; vital signs, postoperative interventions, and administrative codes for superficial SSI; postoperative recovery status for deep SSI; and vital signs for organ-space SSI.
Conclusions
Extending previous reviews focusing on model performance, this review mapped explainability methods and important features in SSI prediction, identifying recurring predictor patterns and substantial methodological heterogeneity across prognostic and diagnostic settings. Incomplete reporting of feature definitions and explanatory rationale, together with limited clinical relevance, constrained clinical interpretation and actionability. Clinician-informed reporting frameworks and validation of explanation fidelity and clinical relevance are needed to improve the trustworthiness and utility of SSI prediction models.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Public Health Emergency Center, Chinese Center for Disease Control and Prevention, Beijing, China.
- School of Nursing, The Hong Kong Polytechnic University, 11 Yuk Choi Rd, Hong Kong Special Administrative Region, China (Hong Kong), +85227666398.
- Faculty of Health Sciences, University of Malta, Msida, Malta.
- Cibles et Médicaments des Infections et de l' Immunité - UR 1155, Nantes Université, Nantes, France.
- Centre d' Appui à la Prévention des Infections Associées aux Soins des Pays de la Loire, Centre Hospitalier Universitaire (CHU) - Le Tourville, Nantes, France.
- National Institute for Health Research Health Protection Research Unit in Healthcare Associated Infections and Antimicrobial Resistance, Imperial College London, London, United Kingdom.
Zitieren
APA
Sun, R., Liu, Y., Lu, S., Tartari, E., Birgand, G., Yang, L., & Zhou, L. (2026). Explainable Machine Learning Predictive Models for Surgical Site Infections: Scoping Review. Journal of Medical Internet Research, 28, e94617-e94617. https://doi.org/10.2196/94617
Vancouver
Sun R, Liu Y, Lu S, Tartari E, Birgand G, Yang L, et al. Explainable Machine Learning Predictive Models for Surgical Site Infections: Scoping Review. Journal of Medical Internet Research. 2026;28:e94617-e94617. doi:10.2196/94617
BibTeX
@article{Sun2026,
author = {Sun, Rui and Liu, Yao and Lu, Shuya and Tartari, Ermira and Birgand, Gabriel and Yang, Lin and Zhou, Lei},
title = {Explainable Machine Learning Predictive Models for Surgical Site Infections: Scoping Review},
journal = {Journal of Medical Internet Research},
year = {2026},
volume = {28},
pages = {e94617--e94617},
doi = {10.2196/94617}
}
EUVIMED