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

David Reinecke,Nader Maarouf,Andrew Smith,Daniel Alber,John Markert,Nicolas K. Goff,Todd C. Hollon,Asadur Chowdury,Cheng Jiang,Xinhai Hou,Anna-Katharina Meissner,Gina Fuertjes,Maximilian I. Ruge,Daniel Ruess,Thomas Stehle,Abdulkader Al-Shughri,Lisa I. Koerner,Georg Widhalm,Thomas Roetzer-Pejrimovsky,John G. Golfinos,Matija Snuderl,Volker Neuschmelting,Daniel A. Orringer

DOI: https://doi.org/10.1101/2024.08.25.24312509

2024-08-26

Abstract:Accurate intraoperative diagnosis is crucial for differentiating between primary 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. We imaged unprocessed, label-free tissue samples intraoperatively using a portable Raman scattering microscope, generating virtual H&E-like images within less than three minutes. 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 two additional independent test cohorts. We trained on 54,000 SRH patch images sourced from surgical resections and stereotactic-guided biopsies, including various CNS tumor/non-tumor lesions. Training and test data were collected from four tertiary international medical centers. The final histopathological diagnosis served as ground-truth. In the prospective test cohort of PCNSL and non-PCNSL entities (n=160), RapidLymphoma achieved an overall balanced accuracy of 97.81%, non-inferior to frozen section analysis in detecting PCNSL (100% vs. 78.94%). The additional test cohorts (n=420, n=59) reached balanced accuracy rates of 95.44% and 95.57% in differentiating IDH-wildtype diffuse gliomas and various brain metastasis from PCNSL. Visual heatmaps revealed capabilities to detect class-specific histomorphological key features. RapidLymphoma proves reliable and valid for intraoperative PCNSL detection and differentiation from other CNS entities. It provides visual feedback within three minutes, enabling fast clinical decision-making and subsequent treatment strategy planning.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to rapidly and accurately detect primary central nervous system lymphoma (PCNSL) during the surgical process and distinguish it from other common central nervous system (CNS) tumors. Specifically, the main challenges faced by the researchers include:

1. **Diagnostic accuracy**: Traditional intraoperative histopathological analysis methods such as frozen sections or cytological smears have limitations in correctly detecting PCNSL and differentiating non - PCNSL entities. Their accuracy rates range from 30.8% to 89.6%, and it is even difficult to make a definite diagnosis using immunohistochemical techniques.

2. **Time constraints**: Since PCNSL is a rare disease but requires urgent personalized treatment, rapid intraoperative decision - making is crucial for further treatment planning of patients. Traditional diagnostic methods are often time - consuming and cannot meet this need.

3. **Similarities in imaging and clinical manifestations**: PCNSL has similarities in clinical and imaging aspects with other common CNS tumors (such as gliomas and metastatic cancers), which increases the difficulty of accurate intraoperative diagnosis.

To address these challenges, the researchers combined **stimulated Raman histology (SRH)** and **deep - learning techniques** to develop a self - supervised learning model named **RapidLymphoma**. This model can generate virtual H&E - stained images during the operation and complete the detection and classification of PCNSL within less than three minutes through a deep - learning pipeline. The research team verified the effectiveness and reliability of this model in an international multi - center prospective clinical trial. The results showed that the overall balanced accuracy rate of RapidLymphoma in detecting PCNSL reached 97.81%, which was significantly better than the traditional frozen - section analysis method.

Through this method, RapidLymphoma not only improves the accuracy of intraoperative diagnosis but also provides real - time visual feedback for surgeons and pathologists, which is helpful for making rapid clinical decisions and treatment plans.