Accurate detection of brain tumor boundaries is crucial for successful tumor removal and better patient outcomes. A novel method using label-free Fluorescence Lifetime Imaging (iFLIm) is presented in this study. The approach involved developing an optimized classification model based on tumor enhancement status, utilizing multispectral FLIm. The method was evaluated on 52 patients with adult-type diffuse glioma, demonstrating promising results with 87% sensitivity, 92% specificity, and an AUC of 0.90. This FLIm-based model has the potential to offer a non-invasive and real-time technique to assist neurosurgeons in accurately identifying tumor infiltrates, potentially improving tumor resection and patient outcomes.
|