Abstract
Recent artificial intelligence (AI) algorithms aided intraoperative decision-making via stimulated Raman histology (SRH) during craniotomy. This study assesses deep-learning algorithms for rapid intraoperative diagnosis from SRH images in small stereotactic-guided brain biopsies. It defines a minimum tissue sample size threshold to ensure diagnostic accuracy. A prospective single-center study examined 121 SRH images from 84 patients with unclear intracranial lesions undergoing stereotactic brain biopsy. Unprocessed, label-free samples were imaged with a portable fiber-laser Raman scattering microscope. Three deep-learning models were tested to (I) identify tumorous/non-tumorous tissue as qualitative biopsy control, (II) subclassify into high-grade glioma (CNS WHO grade 4), diffuse low-grade glioma (CNS WHO grade 2-3), metastases, lymphoma, or gliosis, and (III) molecularly subtype IDH- and 1p/19q-status of adult-type diffuse gliomas. Model predictions were evaluated against frozen section analysis and final neuropathological diagnoses. The first model identified tumorous/non-tumorous tissue with 91.7% accuracy. Sample size on slides impacted accuracy in brain tumor subclassification (81.6%, κ=0.72 frozen section; 73.9%, κ=0.61 second model), with SRH being smaller than H&E (4.1±2.5mm² vs 16.7±8.2mm², p<0.001). SRH images with over 140 high-quality patches and a mean squeezed sample of 5.26mm² yielded 89.5% accuracy in subclassification and 93.9% in molecular subtyping of adult-type diffuse gliomas. AI-based SRH image analysis is non-inferior to frozen section analysis in detecting and subclassifying brain tumors during small stereotactic-guided biopsies once a critical squeezed sample size is reached. Beyond frozen section analysis, it enables valid molecular glioma subtyping, allowing faster treatment decisions in the future. Refinement is needed for long-term application.
Overview
- The study assesses deep-learning algorithms for rapid intraoperative diagnosis from stimulated Raman histology (SRH) images in small stereotactic-guided brain biopsies. It defines a minimum tissue sample size threshold to ensure diagnostic accuracy. A prospective single-center study examined 121 SRH images from 84 patients with unclear intracranial lesions undergoing stereotactic brain biopsy. Unprocessed, label-free samples were imaged with a portable fiber-laser Raman scattering microscope. Three deep-learning models were tested to (I) identify tumorous/non-tumorous tissue as qualitative biopsy control, (II) subclassify into high-grade glioma (CNS WHO grade 4), diffuse low-grade glioma (CNS WHO grade 2-3), metastases, lymphoma, or gliosis, and (III) molecularly subtype IDH- and 1p/19q-status of adult-type diffuse gliomas. Model predictions were evaluated against frozen section analysis and final neuropathological diagnoses. The first model identified tumorous/non-tumorous tissue with 91.7% accuracy. Sample size on slides impacted accuracy in brain tumor subclassification (81.6%, κ=0.72 frozen section; 73.9%, κ=0.61 second model), with SRH being smaller than H&E (4.1±2.5mm² vs 16.7±8.2mm², p<0.001). SRH images with over 140 high-quality patches and a mean squeezed sample of 5.26mm² yielded 89.5% accuracy in subclassification and 93.9% in molecular subtyping of adult-type diffuse gliomas. AI-based SRH image analysis is non-inferior to frozen section analysis in detecting and subclassifying brain tumors during small stereotactic-guided biopsies once a critical squeezed sample size is reached. Beyond frozen section analysis, it enables valid molecular glioma subtyping, allowing faster treatment decisions in the future. Refinement is needed for long-term application.
Comparative Analysis & Findings
- The study compared the outcomes observed under different experimental conditions or interventions detailed in the study. The three deep-learning models were tested for accuracy in identifying tumorous/non-tumorous tissue, subclassifying brain tumors, and molecularly subtyping gliomas. The first model identified tumorous/non-tumorous tissue with 91.7% accuracy. Sample size on slides impacted accuracy in brain tumor subclassification, with SRH being smaller than H&E. SRH images with over 140 high-quality patches and a mean squeezed sample of 5.26mm² yielded 89.5% accuracy in subclassification and 93.9% in molecular subtyping of adult-type diffuse gliomas. AI-based SRH image analysis is non-inferior to frozen section analysis in detecting and subclassifying brain tumors during small stereotactic-guided biopsies once a critical squeezed sample size is reached. Beyond frozen section analysis, it enables valid molecular glioma subtyping, allowing faster treatment decisions in the future. Refinement is needed for long-term application.
Implications and Future Directions
- The study's findings have significant implications for the field of research or clinical practice. AI-based SRH image analysis is non-inferior to frozen section analysis in detecting and subclassifying brain tumors during small stereotactic-guided biopsies once a critical squeezed sample size is reached. Beyond frozen section analysis, it enables valid molecular glioma subtyping, allowing faster treatment decisions in the future. However, refinement is needed for long-term application. Future research directions could explore the use of AI-based SRH image analysis in larger biopsies, in other types of brain tumors, and in combination with other imaging modalities. Additionally, further studies are needed to evaluate the long-term safety and efficacy of AI-based SRH image analysis in clinical practice.