Benjamin Li
Millburn High School
Millburn, New Jersey
Multimodal, Depth-Aware Brain Tumor Segmentation in Sub-Saharan Populations
Benjamin created a tool that uses AI to improve how brain tumors are found in low-quality medical images.
View Poster
Benjamin Li, 18, of Short Hills, created an AI tool to identify brain tumors from low-quality MRI scans for his Regeneron Science Talent Search computational biology and bioinformatics project. AI models that analyze medical images can help doctors make a diagnosis quickly and accurately. However, many conventional models are only trained on high-quality data and struggle to handle low-quality images. This makes it harder to diagnose brain tumors in places with fewer resources, like Sub-Saharan Africa, and in populations that are not often used in training sets. Delayed or incorrect diagnoses lead to worse outcomes.
For his project, Benjamin developed a model, MD-SA2, adapting and transforming aspects from state-of-the-art image segmentation algorithms. This model was more efficient than existing models and more accurate with low-quality images. Benjamin’s AI tool can make it easier to diagnose brain tumors using limited imaging equipment.
Benjamin, the child of Shijian Li and Rong Chen, attends Millburn High School. Benjamin has a passion for ethical AI and volunteers at Jersey Cares. He received the President’s Volunteer Service Award.
Beyond the Project
Benjamin founded the computer science club at his high school and mentors others interested in AI. He helped edit his school’s science journal.
FUN FACTS: Benjamin is a varsity track and field athlete and an avid outdoorsman. He has hiked Yosemite National Park and recently finished the highly competitive 6.6-mile Loon Mountain Race in N.H.
