Central nervous system (CNS) tumors are a leading cause of cancer-related deaths, particularly among children, and the primary treatment is neurosurgical resection. Precisely identifying the tumor type is crucial for developing an effective surgical plan. However, the current methods of histological assessment and molecular analysis face challenges in providing a clear diagnosis during surgery. In a study published in Nature, Vermeulen and colleagues developed Sturgeon, a deep learning model designed to classify CNS tumor type rapidly and accurately to improve surgical decision-making.