Towards Machine Learning-based Quantitative Hyperspectral Image Guidance for Brain Tumor Resection

TL;DR

This study explores the use of hyperspectral imaging and machine learning to classify brain tumor types, grades, and margins with high accuracy, leveraging fluorophore spectra as optical biomarkers for improved intraoperative guidance.

Contribution

It demonstrates the effectiveness of five fluorophore spectra combined with machine learning models for accurate brain tumor classification during surgery.

Findings

Achieved 84-87% accuracy in tumor type classification

Achieved 96% accuracy in tumor grade classification

Fluorophore abundances significantly differ across tissue types and tumor characteristics

Abstract

Complete resection of malignant gliomas is hampered by the difficulty in distinguishing tumor cells at the infiltration zone. Fluorescence guidance with 5-ALA assists in reaching this goal. Using hyperspectral imaging, previous work characterized five fluorophores' emission spectra in most human brain tumors. In this paper, the effectiveness of these five spectra was explored for different tumor and tissue classification tasks in 184 patients (891 hyperspectral measurements) harboring low- (n=30) and high-grade gliomas (n=115), non-glial primary brain tumors (n=19), radiation necrosis (n=2), miscellaneous (n=10) and metastases (n=8). Four machine learning models were trained to classify tumor type, grade, glioma margins and IDH mutation. Using random forests and multi-layer perceptrons, the classifiers achieved average test accuracies of 84-87%, 96%, 86%, and 93% respectively. All five fluorophore abundances varied between tumor margin types and tumor grades (p < 0.01). For tissue type, at least four of the five fluorophore abundances were found to be significantly different (p < 0.01) between all classes. These results demonstrate the fluorophores' differing abundances in different tissue classes, as well as the value of the five fluorophores as potential optical biomarkers, opening new opportunities for intraoperative classification systems in fluorescence-guided neurosurgery.