EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access

Fast intraoperative detection of primary central nervous system lymphoma and differentiation from common central nervous system tumors using stimulated Raman histology and deep learning

David Reinecke iD, Nader Maarouf, Andrew Smith, Daniel Alber, John Markert, Nicolas K Goff, Todd C Hollon, Asadur Chowdury, Cheng Jiang, Xinhai Hou, Anna-Katharina Meissner, Gina Fürtjes und 11 weitere

Neuro-Oncology · 21.06.2025 · 27 (5) · 1297-1310

DOI: 10.1093/neuonc/noae270 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Neuro-Oncology, online erschienen am 14.12.2024. Verfasst von David Reinecke und 22 weiteren Autor:innen. Beteiligte Einrichtung: Department of General Neurosurgery, Center for Neurosurgery, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Ger…. Hauptthemen laut MEDLINE: Deep Learning, Central Nervous System Neoplasms, Lymphoma, Spectrum Analysis, Raman. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, German Research Foundation, NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate intraoperative diagnosis is crucial for differentiating between primary central nervous system (CNS) lymphoma (PCNSL) and other CNS entities, guiding surgical decision-making, but represents significant challenges due to overlapping histomorphological features, time constraints, and differing treatment strategies. We combined stimulated Raman histology (SRH) with deep learning to address this challenge.

Methods

We imaged unprocessed, label-free tissue samples intraoperatively using a portable Raman scattering microscope, generating virtual H&E-like images within <3 min. We developed a deep learning pipeline called RapidLymphoma based on a self-supervised learning strategy to (1) detect PCNSL, (2) differentiate from other CNS entities, and (3) test the diagnostic performance in a prospective international multicenter cohort and 2 additional independent test cohorts. We trained on 54 000 SRH patch images sourced from surgical resections and stereotactic-guided biopsies, including various CNS neoplastic/nonneoplastic lesions. Training and test data were collected from 4 tertiary international medical centers. The final histopathological diagnosis served as ground truth.

Results

In the prospective test cohort of PCNSL and non-PCNSL entities (n = 160), RapidLymphoma achieved an overall balanced accuracy of 97.81% ± 0.91, non-inferior to frozen section analysis in detecting PCNSL (100% vs. 77.77%). The additional test cohorts (n = 420, n = 59) reached balanced accuracy rates of 95.44% ± 0.74 and 95.57% ± 2.47 in differentiating IDH-wildtype diffuse gliomas and various brain metastasis from PCNSL. Visual heatmaps revealed RapidLymphoma’s capabilities to detect class-specific histomorphological key features.

Conclusions

RapidLymphoma proves reliable and valid for intraoperative PCNSL detection and differentiation from other CNS entities. It provides visual feedback within 3 min, enabling fast clinical decision-making and subsequent treatment strategy planning.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningCentral Nervous System NeoplasmsLymphomaSpectrum Analysis, RamanHumansDiagnosis, DifferentialProspective StudiesFemaleMaleMiddle AgedAgedAdultPrognosis

Stichwörter der Autor:innen

artificial intelligencebrain tumordeep learningintraoperative histopathologyoptical imagingprimary CNS lymphomastimulated Raman histology

Autor:innen und Institutionen

  • Department of General Neurosurgery, Center for Neurosurgery, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.
  • Department of Neurosurgery, New York University Grossman School of Medicine, New York, New York, USA.
  • Heersink School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama, USA.
  • Department of Neurosurgery, University of Texas at Austin Dell Medical School, Austin, Texas, USA.
  • Machine Learning in Neurosurgery Laboratory, Department of Neurosurgery, University of Michigan, Ann Arbor, Michigan, USA.
  • Center for Integrated Oncology, Faculty of Medicine and University Hospital Cologne, University of Cologne, Cologne, Germany.

ORCID-Profile: David Reinecke, Maximilian I Ruge, Georg Widhalm, Thomas Roetzer-Pejrimovsky, John G Golfinos, Matija Snuderl

Förderung

  • NCI NIH HHS
  • German Research Foundation
  • NIH HHS
  • National Institutes of Health

Zitieren

APA

Reinecke, D., Maarouf, N., Smith, A., Alber, D., Markert, J., Goff, N. K., Hollon, T. C., Chowdury, A., Jiang, C., Hou, X., Meissner, A. K., Fürtjes, G., Ruge, M. I., Ruess, D., Stehle, T., Al-Shughri, A., Körner, L. I., Widhalm, G., Roetzer-Pejrimovsky, T., …, & Orringer, D. A. (2025). Fast intraoperative detection of primary central nervous system lymphoma and differentiation from common central nervous system tumors using stimulated Raman histology and deep learning. Neuro-Oncology, 27(5), 1297-1310. https://doi.org/10.1093/neuonc/noae270

Vancouver

Reinecke D, Maarouf N, Smith A, Alber D, Markert J, Goff NK, et al. Fast intraoperative detection of primary central nervous system lymphoma and differentiation from common central nervous system tumors using stimulated Raman histology and deep learning. Neuro-Oncology. 2025;27(5):1297-1310. doi:10.1093/neuonc/noae270

BibTeX
@article{Reinecke2025,
  author = {Reinecke, David and Maarouf, Nader and Smith, Andrew and Alber, Daniel and Markert, John and Goff, Nicolas K and Hollon, Todd C and Chowdury, Asadur and Jiang, Cheng and Hou, Xinhai and Meissner, Anna-Katharina and Fürtjes, Gina and Ruge, Maximilian I and Ruess, Daniel and Stehle, Thomas and Al-Shughri, Abdulkader and Körner, Lisa I and Widhalm, Georg and Roetzer-Pejrimovsky, Thomas and Golfinos, John G and Snuderl, Matija and Neuschmelting, Volker and Orringer, Daniel A},
  title = {Fast intraoperative detection of primary central nervous system lymphoma and differentiation from common central nervous system tumors using stimulated Raman histology and deep learning},
  journal = {Neuro-Oncology},
  year = {2025},
  volume = {27},
  number = {5},
  pages = {1297--1310},
  doi = {10.1093/neuonc/noae270}
}
Kostenlos starten →