Advancements in AI for Real-Time Brain Tumor Diagnosis and Classification
Researchers have developed a neural network classifier called Sturgeon that can rapidly classify central nervous system (CNS) tumours during surgery using nanopore DNA sequencing. The classifier, trained on simulated nanopore sequencing data, demonstrated high accuracy in classifying tumour samples based on methylation profiles, even with limited sequencing data. The study showed that Sturgeon could provide a molecular diagnosis within 25-50 minutes of sequencing, allowing for real-time decision-making during surgery. The classifier was also validated on real nanopore-sequenced samples, further confirming its robustness and accuracy. This technology has the potential to improve surgical strategies and reduce the need for additional surgeries in CNS tumour patients.