Predicting Textbook Oncological Outcomes with Machine Learning: A Retrospective Review of Patients Who Underwent Resection for Malignant Tumours
Vorhersage von Textbook oncological outcome mit Machine Learning: retrospektive Untersuchung von Patient:innen nach Resektion maligner Tumoren
Journal of Clinical Medicine · 02.11.2026 · 15 (21) · 8527
DOI: 10.3390/jcm15218527 ↗Auf einen Blick
Ein Machine-Learning-Modell sagte das Textbook oncological outcome nach Leberresektion für maligne Tumoren mit guter Diskriminierung voraus. Nach Re-Kalibrierung zeigten sich gute Kalibrierungsmaße.
Fragestellung
Kann ein Machine-Learning-Modell das Risiko für ein Textbook oncological outcome (TOO) nach Leberresektion wegen maligner Tumoren vorhersagen?
Methode
Retrospektive Kohortenstudie mit Daten aus 1999 bis 2023; ML-Modelle mit präoperativen Variablen; Bewertung von Diskriminierung, Kalibrierung und Interpretierbarkeit.
Ergebnisse
Trainingskohorte n = 583, Testkohorte n = 146. RF-Modell: AUROC 0,813 (95 %-KI 0,731–0,877). Nach Re-Kalibrierung: Brier-Score 0,162, Kalibrierungsinterzept 0,06, Kalibrierungsslope 0,92. TOO: 91 % vs. 80 % vs. 38 % in den Risikogruppen (p < 0,001).
Einschränkungen
Explorativer Charakter der Studie; externe Validierung vor klinischer Anwendung erforderlich.
Schlussfolgerung der Autor:innen
Das ML-basierte Modell zur Vorhersage von TOO nach Leberresektion zeigte nach Re-Kalibrierung gute Diskriminierung und Kalibrierung und könnte für präoperative Risikobewertung und Aufklärung nutzbar sein.
Kernaussagen
- RF-Modell mit AUROC 0,813 (95 %-KI 0,731–0,877)
- TOO-Raten: 91 %, 80 % und 38 % in den Risikogruppen
- Wichtige Merkmale: Resektionsausmaß, Tumorort und größte Tumorgröße
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift Journal of Clinical Medicine, online erschienen am 02.11.2026. Verfasst von Meet Patel und 7 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Medicine and Health, The University of Sydney, Camperdown, NSW 2050, Australia. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Australian Government, Commonwealth.
Abstract (Originalsprache)
Introduction
Textbook oncological outcome (TOO) is a composite measure used to assess the quality of care in surgical oncology. Specifically, in liver surgery, it assesses several domains including intraoperative events, postoperative recovery, liver-specific complications, early mortality, and resection margins. Historically, one-third of patients undergoing liver resection have had a non-ideal outcome that has been difficult to predict. The aim of this study is to predict the risk of achieving textbook oncological outcome (TOO) using machine learning (ML) models for patients undergoing liver resection for malignant tumours.
Methods
A retrospective cohort study of patients undergoing liver resection for malignant tumours was undertaken between 1999 and 2023 from a prospectively collected database. ML models were developed using preoperative variables as candidate predictors of TOO. Feature selection was performed using stability selection with LASSO regularisation. Model performance was evaluated using discrimination and calibration metrics. Model interpretability was assessed using SHapley Additive exPlanations. Predicted probabilities were stratified into tertiles to form risk groups (low, moderate, and high).
Results
From the training cohort (n = 583), the RF model demonstrated the highest discrimination (AUROC = 0.813, 95% confidence interval [CI]= 0.731–0.877) for predicting TOO in the test cohort (n = 146). Post-recalibration, the Brier score was 0.162, calibration intercept was 0.06 and calibration slope was 0.92. The top three features were the extent of the resection, the site of the tumour, and the size of the largest tumour. TOO rates decreased in low-, moderate-, and high-risk groups (91%, 80% and 38% respectively, p < 0.001).
Conclusions
An ML-based model for predicting TOO following liver resection demonstrated good discrimination and calibration metrics after recalibration. Risk stratification showed a clear gradient in TOO achievement, supporting its potential use in preoperative risk assessment and patient counselling. The current exploratory nature of this study necessitates further external validation before clinical application.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Faculty of Medicine and Health, The University of Sydney, Camperdown, NSW 2050, Australia.
- Upper Gastrointestinal Surgical Unit, Royal North Shore Hospital, St Leonards, NSW 2065, Australia.
- Faculty of Health and Medical Sciences, University of Adelaide, Adelaide, SA 5000, Australia.
ORCID-Profile: Meet Patel, Jonathan Ben Daniel, Nazim Bhimani, Anthony R Glover, Thomas J Hugh
Förderung
- Australian Government
- Commonwealth
Zitieren
APA
Patel, M., Daniel, J. B., Bhimani, N., Kovoor, J. G., Fadia, A., Brown, K., Glover, A. R., & Hugh, T. J. (2026). Predicting Textbook Oncological Outcomes with Machine Learning: A Retrospective Review of Patients Who Underwent Resection for Malignant Tumours. Journal of Clinical Medicine, 15(21), 8527. https://doi.org/10.3390/jcm15218527
Vancouver
Patel M, Daniel JB, Bhimani N, Kovoor JG, Fadia A, Brown K, et al. Predicting Textbook Oncological Outcomes with Machine Learning: A Retrospective Review of Patients Who Underwent Resection for Malignant Tumours. Journal of Clinical Medicine. 2026;15(21):8527. doi:10.3390/jcm15218527
BibTeX
@article{Patel2026,
author = {Patel, Meet and Daniel, Jonathan Ben and Bhimani, Nazim and Kovoor, Joshua G and Fadia, Arjun and Brown, Kai and Glover, Anthony R and Hugh, Thomas J},
title = {Predicting Textbook Oncological Outcomes with Machine Learning: A Retrospective Review of Patients Who Underwent Resection for Malignant Tumours},
journal = {Journal of Clinical Medicine},
year = {2026},
volume = {15},
number = {21},
pages = {8527},
doi = {10.3390/jcm15218527}
}
EUVIMED