NIMG-22. RAPID ACCURATE INTRAOPERATIVE DIAGNOSIS OF PEDIATRIC BRAIN TUMORS USING STIMULATED RAMAN HISTOLOGY

Accurate histopathologic diagnosis is essential for providing optimal surgical management of pediatric brain tumors. Current methods for intraoperative histology are time- and labor-intensive intensive and often introduce artifact that complicates interpretation. The ideal system of intraoperative histology would be built on a rapid, standardized workflow with minimal tissue processing and diagnostic-quality images. Stimulated Raman histology (SRH) is a new label-free imaging technique that provides intraoperative histologic images of fresh, unprocessed surgical specimens. Here, we evaluate the capacity of SRH for use in the intraoperative diagnosis of pediatric brain tumors. SRH revealed key diagnostic features in fresh tissue specimens collected from 33 prospectively enrolled pediatric brain tumor patients, preserving tumor cytology and histoarchitecture in all specimens. We simulated an intraoperative neuropathology consultation for 25 patients with specimens imaged using both SRH and standard H&E histology. SRH-based diagnoses achieve near-perfect diagnostic concordance (Cohen’s kappa, κ > 0.90) and an accuracy of 92-96%. Leveraging the image contrast generated by the intrinsic biochemical data embedded in SRH images, we developed a quantitative histologic method that allows for rapid image feature extraction. Nuclear density, tumor-associated macrophage infiltration, and nuclear morphology parameters extracted from 3337 SRH fields of view were used to develop and train a decision tree machine-learning model. Using SRH image features, our model correctly identified normal versus lesional tissue and low-grade versus high-grade tumors in 25 fresh pediatric surgical specimens with 100% accuracy. Our results provide insight into how SRH can deliver rapid diagnostic histologic data during pediatric brain tumor surgery.