Towards machine learning-based quantitative hyperspectral image guidance for brain tumor resection

• Published in: Communications medicine 2024-07, Vol.4 (1), p.131-11, Article 131
• Main Authors: Black, David, Byrne, Declan, Walke, Anna, Liu, Sidong, Di Ieva, Antonio, Kaneko, Sadahiro, Stummer, Walter, Salcudean, Tim, Suero Molina, Eric
• Format: Article
• Language: English
• Subjects: 14/34; 631/378/116; 631/67/1922; 692/53/2421; Brain cancer; Brain research; Classification; Clinical outcomes; Glioma; Light; Machine learning; Medicine; Medicine & Public Health; Metastasis; Mutation; Neurosurgery; Spectrum analysis; Surgery; Tumors
• ISSN: 2730-664X; 2730-664X
• DOI: 10.1038/s43856-024-00562-3

Background 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. Methods 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. Results Using random forests and multilayer perceptrons, the classifiers achieve average test accuracies of 84–87%, 96.1%, 86%, and 91% respectively. All five fluorophore abundances vary between tumor margin types and tumor grades ( p