Enhancing decision-making in glioblastoma surgery through an explainable human-AI collaboration: an international multicenter model development and external validation study

Kernbach JM, Schroeder U, Hakvoort K, Ort J, Hamou H, Bzdok D, Temel Y, Kubben P, Weyland C, Wiesmann M, Staartjes V, Akeret K, Vieli M, Serra C, Regli L, Grau S, Dührsen L, Ricklefs F, Schnell O, Ormond DR, Grote A, Simon M, Meredig H, Schell M, Bendszus M, Neuloh G, Clusmann H, Heiland DH, Delev D (2025)

Publication Type: Journal article

Publication year: 2025

Journal

npj Precision Oncology Springer Nature

Book Volume: 9

Article Number: 387

Journal Issue: 1

DOI: 10.1038/s41698-025-01183-2

Abstract

Surgical resection improves survival in glioblastoma, yet predicting the extent of resection (EOR) remains highly challenging. We developed and externally validated an explainable AI model to generate personalized EOR estimates in 811 glioblastoma patients undergoing microsurgical resection. EOR was categorized into gross-total (GTR), near-total (NTR), and subtotal resections (STR). An interpretable framework provided model explanations and sensitivity analyses to assess the model’s strengths and limitations. To demonstrate clinical impact, we compared the performance of the human expert (gold standard) with our AI model and a combined human-AI approach. External validation confirmed generalizability (AUC 0.78, CI 0.73-0.82). Class-specific AUCs were 0.75 (0.67-0.82) for GTR, 0.59 (0.50-0.69) for NTR, and 0.69 (0.53-0.85) for STR. Key predictors included KPS and NANO scores, age, tumor volume, and unfavorable anatomical locations. A combined human-AI collaboration outperformed human experts, with higher overall accuracies (0.53 to 0.94), F1 scores (0.30 to 0.92), and Cohen’s κ (0.41 to 0.84). Enhancing predictive performance through the clinician-AI collaboration, our explainable model supports preoperative planning and highlights the value of integrating machine intelligence into surgical decision-making.

Authors with CRIS profile

Oliver Schnell Lehrstuhl für Neurochirurgie Dieter Henrik Heiland Department of Neurosurgery

Involved external institutions

Universitätsspital Zürich (USZ) Universität zu Köln Universitätsklinikum Hamburg-Eppendorf (UKE) Universitätsklinikum Aachen (UKA) Quebec AI Institute / Quebec Artificial Intelligence Institute / Montreal Institute for Learning Algorithms (MILA) Maastricht University Ruprecht-Karls-Universität Heidelberg Philipps-Universität Marburg University of Colorado Anschutz Medical Campus Evangelisches Klinikum Bethel Klinikum Bremerhaven-Reinkenheide

How to cite

APA:

Kernbach, J.M., Schroeder, U., Hakvoort, K., Ort, J., Hamou, H., Bzdok, D.,... Delev, D. (2025). Enhancing decision-making in glioblastoma surgery through an explainable human-AI collaboration: an international multicenter model development and external validation study. npj Precision Oncology, 9(1). https://doi.org/10.1038/s41698-025-01183-2

MLA:

Kernbach, Julius M., et al. "Enhancing decision-making in glioblastoma surgery through an explainable human-AI collaboration: an international multicenter model development and external validation study." npj Precision Oncology 9.1 (2025).

BibTeX: Download