EUVIMEDEUVIMED entdecken
VergleichsstudieSystematische ÜbersichtOpen Access · CC BY 4.0

Transparency in Externally Validated Models: A systematic review of machine learning vs. logistic regression for predicting colorectal anastomotic leakage

Sara Ben Hmido, Houssam Abder Rahim, Erik W Ingwersen, George Burchell, Roel Hompes, Donald van der Peet, Eric Sonneveld, Geert Kazemier, Freek Daams

European Journal of Surgical Oncology · 2025-10 · 51 (10) · 110367

DOI: 10.1016/j.ejso.2025.110367 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift European Journal of Surgical Oncology, 2025-10. Verfasst von Sara Ben Hmido und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Surgery, Amsterdam University Medical Centers, De Boelelaan 1117, 1081, HV Amsterdam, the Netherlands; Cancer Center Amsterdam, De Boel…. Hauptthemen laut MEDLINE: Anastomotic Leak, Colorectal Neoplasms, Machine Learning, Rectum. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Medtronic International Trading Sàrl.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION: Colorectal resection carries a 2.8 %-30 % risk of anastomotic leakage. Machine learning can estimate risks and guide decisions, but clinical implementation remains inadequate due to transparency issues. This review assesses the performance and transparency of machine learning models compared to logistic regression. METHODS: A systematic review followed PRISMA guidelines. Medline, Embase, Web of Science, and Cochrane databases were searched for studies using Logistic Regression or Machine Learning with external validation for colorectal anastomotic leakage prediction. Data were extracted using CHARMS, risk of bias assessed with PROBAST, and transparency with TRIPOD + AI. RESULTS: Ten studies were included. Machine learning models were validated on smaller cohorts than logistic regression. Transparency scores ranged from 29 % to 63 %, averaging 45 % for logistic regression and 43 % for machine learning. Reporting of missing data was inconsistent, and external validation was limited. Most studies had a high risk of bias due to small sample sizes and low event counts. CONCLUSION: In comparison to Logistic regression studies, machine learning studies are limited by small cohorts, low outcome numbers, and a lower level of transparency. Future research should prioritise transparency, adhere to TRIPOD + AI standards, and develop LR and ML models in parallel using the same datasets while ensuring separate models for colon and rectal surgery. Currently, these models are not yet suitable for clinical implementation; more robust and transparent models must be developed based on these recommendations before they can be applied in clinical practice.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Anastomotic LeakColorectal NeoplasmsMachine LearningRectumHumansLogistic ModelsRisk Assessment

Stichwörter der Autor:innen

Anastomotic leakageColorectal surgeryLogistic regressionMachine learningPrediction

Autor:innen und Institutionen

  • Department of Surgery, Amsterdam University Medical Centers, De Boelelaan 1117, 1081, HV Amsterdam, the Netherlands; Cancer Center Amsterdam, De Boelelaan 1118, 1081, HV Amsterdam, the Netherlands. Electronic address: s…
  • Biomedical Engineering, TU Delft, Mekelweg 2, 2628 CD, Delft, the Netherlands.
  • Department of Surgery, Amsterdam University Medical Centers, De Boelelaan 1117, 1081, HV Amsterdam, the Netherlands; Cancer Center Amsterdam, De Boelelaan 1118, 1081, HV Amsterdam, the Netherlands.
  • Medical Library, Vrije Universiteit, De Boelelaan 1117, 1081, HV Amsterdam, the Netherlands.
  • Department of Surgery, Dijklander Hospital, Hoorn, the Netherlands.

Förderung

  • Medtronic International Trading Sàrl

Zitieren

APA

Ben Hmido, S., Abder Rahim, H., Ingwersen, E. W., Burchell, G., Hompes, R., van der Peet, D., Sonneveld, E., Kazemier, G., & Daams, F. (2025). Transparency in Externally Validated Models: A systematic review of machine learning vs. logistic regression for predicting colorectal anastomotic leakage. European Journal of Surgical Oncology, 51(10), 110367. https://doi.org/10.1016/j.ejso.2025.110367

Vancouver

Ben Hmido S, Abder Rahim H, Ingwersen EW, Burchell G, Hompes R, van der Peet D, et al. Transparency in Externally Validated Models: A systematic review of machine learning vs. logistic regression for predicting colorectal anastomotic leakage. European Journal of Surgical Oncology. 2025;51(10):110367. doi:10.1016/j.ejso.2025.110367

BibTeX
@article{BenHmido2025,
  author = {Ben Hmido, Sara and Abder Rahim, Houssam and Ingwersen, Erik W and Burchell, George and Hompes, Roel and van der Peet, Donald and Sonneveld, Eric and Kazemier, Geert and Daams, Freek},
  title = {Transparency in Externally Validated Models: A systematic review of machine learning vs. logistic regression for predicting colorectal anastomotic leakage},
  journal = {European Journal of Surgical Oncology},
  year = {2025},
  volume = {51},
  number = {10},
  pages = {110367},
  doi = {10.1016/j.ejso.2025.110367}
}
Kostenlos starten →