Stereotactic radiosurgery (SRS), a validated treatment for brain metastases and brain tumors, requires accurate tumor contouring for treatment planning and delivery. This manual segmentation process is time-consuming and prone to substantial inter-practitioner variability even amongst experts, leading to large variations in care quality. The research team of the NTUH Cyberknife Center and the healthcare Artificial Intelligence (AI) company, Vysioneer Inc., have developed an AI-powered tumor automated brain tumor segmentation (ABS) system, VBrain. It has received the U.S. Food and Drug Administration (FDA) clearance for the three most common types of brain tumors: brain metastasis, meningioma, and acoustic neuroma.  As the first-ever AI-powered tumor auto-contouring solution cleared by the FDA, this ABS system enables a quicker response time for performing SRS with more precision in targeting these common brain tumors. Manual contouring time that may take hours by clinicians is completed within minutes with the assistance of the AI. As the first of its kind, the fully automated solution also ensures precision mapping of brain tumors with closer cuts and the ability to identify additional lesions that may be missed by the human eye. In addition to its value in clinical practice, we also demonstrated that the system impacts medical education and training for radiation oncologists or neurosurgeons who are not experts in the field of intracranial SRS. Leveraging the most advanced clinical expertise and cutting-edge computer vision techniques, the industry-university cooperation serves as a role model in championing bioinformatics technology and transforming the healthcare industry in Taiwan.

Figure. The machine learning model for automated brain tumor segmentation (ABS).