Performance of intraoperative artificial intelligence-assisted label-free stimulated Raman histology in recurrent brain tumors
Neuro-Oncology Advances · 12.01.2026 · 8 (1)
DOI: 10.1093/noajnl/vdag267 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neuro-Oncology Advances, online erschienen am 29.10.2026. Verfasst von Gina Fürtjes und 20 weiteren Autor:innen. Beteiligte Einrichtung: Department of General Neurosurgery, Centre of Neurosurgery, University Hospital Cologne, Cologne, Germany. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Joachim Herz Foundation, Köln Fortune, internal research funding of the Faculty of Medicine, University of Cologne.
Abstract (Originalsprache)
Background
Therapy-induced tissue alterations complicate intraoperative diagnosis of recurrent central nervous system (CNS) tumors. We evaluated stimulated Raman histology (SRH) combined with convolutional neural networks (CNNs) for rapid tumor detection and classification.
Methods
In this bicentric observational study, fresh tissue from patients undergoing resection or stereotactic biopsy for suspected recurrent CNS tumors was imaged with fiber-laser-based SRH. CNN1 performed binary tumor/non-tumor prediction and CNN2 histotype classification. CNN1 was evaluated against blinded neuropathologist SRH consensus and CNN2 against the final integrated histopathological diagnosis.
Results
Of 382 acquired SRH images, 262 images from 143 patients were included. CNN1 achieved an AUC of 0.975 (95% CI, 0.95-1.00) for tumor detection. An exploratory cohort-derived tumor probability threshold of ≥59 increased specificity from 70.7% to 92.0% while maintaining 95.2% sensitivity and identifying 71.4% of radiation necrosis samples. Among 212 images in treatment subgroup analysis, CNN1 accuracy did not differ significantly across prior-treatment groups (overall 90.1%; P = .080). CNN2 accuracy declined from 61.8% after surgery alone to 39.0% after surgery, radiotherapy, and medical treatment (χ2 = 6.04, P = .049). Multimodal therapy was independently associated with lower odds of correct CNN2 prediction (adjusted OR, 0.10 [95% CI, 0.02-0.39]; P = .001 and 0.04 [95% CI, 0.01-0.20]; P < .001). 120/382 images (31.4%) were of insufficient diagnostic quality; application of the quality control threshold would have rejected 84/120 (70.0%).
Conclusions
CNN1 maintained high tumor/non-tumor discrimination across treatment subgroups when image quality was sufficient, whereas CNN2 entity-classification performance declined with increasing treatment intensity. Reliable SRH-based diagnosis in recurrent tumors requires validated image-quality control measures and model adaptation to recurrent disease.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of General Neurosurgery, Centre of Neurosurgery, University Hospital Cologne, Cologne, Germany.
- Institute for Neuropathology, University Hospital Cologne, University of Cologne, Cologne, Germany.
- Department of Neurosurgery, Medical Center, University of Freiburg, Freiburg, Germany.
- Center for Advanced Surgical Tissue Analysis (CAST), Medical Center, University of Freiburg, Freiburg, Germany.
- Department of Neurosurgery, West Coast Clinic Heide, Heide, Germany.
- Department of Stereotaxy and Functional Neurosurgery, Centre of Neurosurgery, University Hospital Cologne, Cologne, Germany.
ORCID-Profile: Gina Fürtjes, Nicolas Neidert, Pierre Scheffler, Roland Goldbrunner
Förderung
- Joachim Herz Foundation
- Köln Fortune, internal research funding of the Faculty of Medicine, University of Cologne
Zitieren
APA
Fürtjes, G., Al Shughri, A., Stehle, T., Reinecke, D., Meißner, A. K., Jung, N., Mercan, Z., Neidert, N., Scheffler, P., Leyer, L., Wang, C., Gueorguiev, Z., von Spreckelsen, N., Ruess, D., Freudiger, C., Khalid, F., Ion-Margineanu, A., Goldbrunner, R., Beck, J., …, & Straehle, J. (2026). Performance of intraoperative artificial intelligence-assisted label-free stimulated Raman histology in recurrent brain tumors. Neuro-Oncology Advances, 8(1). https://doi.org/10.1093/noajnl/vdag267
Vancouver
Fürtjes G, Al Shughri A, Stehle T, Reinecke D, Meißner AK, Jung N, et al. Performance of intraoperative artificial intelligence-assisted label-free stimulated Raman histology in recurrent brain tumors. Neuro-Oncology Advances. 2026;8(1). doi:10.1093/noajnl/vdag267
BibTeX
@article{Frtjes2026,
author = {Fürtjes, Gina and Al Shughri, Abdulkader and Stehle, Thomas and Reinecke, David and Meißner, Anna-Katharina and Jung, Nina and Mercan, Zeynep and Neidert, Nicolas and Scheffler, Pierre and Leyer, Lili and Wang, Claire and Gueorguiev, Zachari and von Spreckelsen, Niklas and Ruess, Daniel and Freudiger, Christian and Khalid, Florian and Ion-Margineanu, Adrian and Goldbrunner, Roland and Beck, Jürgen and Neuschmelting, Volker and Straehle, Jakob},
title = {Performance of intraoperative artificial intelligence-assisted label-free stimulated Raman histology in recurrent brain tumors},
journal = {Neuro-Oncology Advances},
year = {2026},
volume = {8},
number = {1},
doi = {10.1093/noajnl/vdag267}
}
EUVIMED