Introducing a machine learning algorithm for delirium prediction—the Supporting SURgery with GEriatric Co-Management and AI project (SURGE-Ahead)
Einführung eines Machine-Learning-Algorithmus zur Delirvorhersage – Projekt Supporting SURgery with GEriatric Co-Management and AI (SURGE-Ahead)
Age and Ageing · 01.05.2024 · 53 (5)
DOI: 10.1093/ageing/afae101 ↗Auf einen Blick
Ein ML-Modell zur Vorhersage eines postoperativen Delirs (POD) wurde bei Patient:innen ab 70 Jahren entwickelt. Der lineare SVM erzielte die beste Trennschärfe.
Fragestellung
Lässt sich bei älteren Patient:innen ein maschinelles Lernmodell zur Vorhersage eines postoperativen Delirs entwickeln?
Methode
Modellentwicklung mit Daten aus der PAWEL-Studie; 878 Patient:innen ab 70 Jahren, davon 209 mit POD. Vergleich von logistischer Regression und linearem SVM; Bewertung per ROC.
Ergebnisse
Ausgewählt wurden 15 Merkmale, u. a. ASA-Score, Multimorbidität, Operationszeit, eGFR, Polypharmazie und kognitive Testsubscores. Der lineare SVM erreichte eine AUC von 0,82 (95 %-KI 0,78–0,85) im Training, 0,81 (95 %-KI 0,71–0,88) im Test und 0,76 (95 %-KI 0,71–0,79) in der Cross-centre-Validierung.
Schlussfolgerung der Autor:innen
Die Autor:innen beschreiben ein klinisch nützliches und erklärbares ML-Modell zur POD-Vorhersage; es soll im SURGE-Ahead-Projekt eingesetzt werden.
Kernaussagen
- Lineares SVM mit bester ROC-AUC für POD-Vorhersage
- AUC: 0,82 Training; 0,81 Test; 0,76 Cross-centre-Validierung
- Modell basiert auf 15 klinischen Merkmalen
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 Age and Ageing, online erschienen am 22.05.2024. Verfasst von Samuel Benovic und 11 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany. Hauptthemen laut MEDLINE: Machine Learning, Delirium, Geriatric Assessment. Die Arbeit wurde bisher 22-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: German Federal Ministry of Education and Research.
Abstract (Originalsprache)
Introduction
Post-operative delirium (POD) is a common complication in older patients, with an incidence of 14–56%. To implement preventative procedures, it is necessary to identify patients at risk for POD. In the present study, we aimed to develop a machine learning (ML) model for POD prediction in older patients, in close cooperation with the PAWEL (patient safety, cost-effectiveness and quality of life in elective surgery) project.
Methods
The model was trained on the PAWEL study’s dataset of 878 patients (no intervention, age ≥ 70, 209 with POD). Presence of POD was determined by the Confusion Assessment Method and a chart review. We selected 15 features based on domain knowledge, ethical considerations and a recursive feature elimination. A logistic regression and a linear support vector machine (SVM) were trained, and evaluated using receiver operator characteristics (ROC).
Results
The selected features were American Society of Anesthesiologists score, multimorbidity, cut-to-suture time, estimated glomerular filtration rate, polypharmacy, use of cardio-pulmonary bypass, the Montreal cognitive assessment subscores ‘memory’, ‘orientation’ and ‘verbal fluency’, pre-existing dementia, clinical frailty scale, age, recent falls, post-operative isolation and pre-operative benzodiazepines. The linear SVM performed best, with an ROC area under the curve of 0.82 [95% CI 0.78–0.85] in the training set, 0.81 [95% CI 0.71–0.88] in the test set and 0.76 [95% CI 0.71–0.79] in a cross-centre validation.
Conclusion
We present a clinically useful and explainable ML model for POD prediction. The model will be deployed in the Supporting SURgery with GEriatric Co-Management and AI project.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute of Geriatric Research, Ulm University Medical Center, Ulm, Germany.
- Agaplesion Bethesda Clinic Ulm, Ulm, Germany.
- Institute of the History, Philosophy and Ethics of Medicine, Ulm University, Ulm, Germany.
- Department of Sociology with a Focus on Innovation and Digitalization, Institute of Sociology, Johannes Kepler University Linz, Linz, Austria.
- DigiHealth Institute, Neu-Ulm University of Applied Sciences, Neu-Ulm, Germany.
- Institute of Medical Systems Biology, Ulm University, Ulm, Germany.
ORCID-Profile: Thomas D Kocar
Förderung
- German Federal Ministry of Education and Research
Zitieren
APA
Benovic, S., Ajlani, A. H., Leinert, C., Fotteler, M., Wolf, D., Steger, F., Kestler, H., Dallmeier, D., Denkinger, M., Eschweiler, G. W., Thomas, C., & Kocar, T. D. (2024). Introducing a machine learning algorithm for delirium prediction—the Supporting SURgery with GEriatric Co-Management and AI project (SURGE-Ahead). Age and Ageing, 53(5). https://doi.org/10.1093/ageing/afae101
Vancouver
Benovic S, Ajlani AH, Leinert C, Fotteler M, Wolf D, Steger F, et al. Introducing a machine learning algorithm for delirium prediction—the Supporting SURgery with GEriatric Co-Management and AI project (SURGE-Ahead). Age and Ageing. 2024;53(5). doi:10.1093/ageing/afae101
BibTeX
@article{Benovic2024,
author = {Benovic, Samuel and Ajlani, Anna H and Leinert, Christoph and Fotteler, Marina and Wolf, Dennis and Steger, Florian and Kestler, Hans and Dallmeier, Dhayana and Denkinger, Michael and Eschweiler, Gerhard W and Thomas, Christine and Kocar, Thomas D},
title = {Introducing a machine learning algorithm for delirium prediction—the Supporting SURgery with GEriatric Co-Management and AI project (SURGE-Ahead)},
journal = {Age and Ageing},
year = {2024},
volume = {53},
number = {5},
doi = {10.1093/ageing/afae101}
}
EUVIMED