J Neurosurg 143:323–331, 2025
Combining 5-aminolevulinic acid (5-ALA) fluorescence and intraoperative ultrasound (ioUS) significantly improved the sensitivity and specificity for detecting high-grade glioma during surgery, compared to either technique alone. A machine learning algorithm (HGGPredictor) further enhanced intraoperative tumor margin prediction, suggesting a new standard for safer, more effective resections.
• Combining 5-ALA fluorescence and intraoperative ultrasound (ioUS) improves the accuracy of high-grade glioma (HGG) resection compared to using either method alone.
• 5-ALA shows higher sensitivity (84.9%), while ioUS provides higher specificity (84.5%); combined, they reach sensitivity of 91% and specificity of 86%.
• The combined approach is especially valuable for maximizing tumor removal while minimizing neurological damage, particularly near eloquent brain regions.
• A machine learning algorithm (HGGPredictor) was developed to predict tumor presence during surgery based on 5-ALA and ioUS results.
• The study included 72 patients and 301 biopsies, with histological analysis as the reference standard.
• The benefit of combination is greatest for strong fluorescence or hyperechogenicity; ioUS is particularly helpful when 5-ALA fluorescence is weak.
• The combined method is accessible and can be integrated into existing surgical protocols without major additional costs.
• Limitations include single-center design and lack of a control group, but results suggest a new standard for HGG resection.













You must be logged in to post a comment.