The RSNA, the American Society of Neuroradiology, and the Medical Image Computing and Computer Assisted Interventions society plan to focus on brain tumors in the latest iteration of the RSNA's artificial intelligence (AI) challenge.
AI developers can participate in one or two challenges for this year's event. The first task is a brain tumor segmentation challenge that requires participants to build models that produce detailed segmentation of brain tumor subregions that correspond to those created by neuroradiologists. This challenge may enable clinical improvements in areas that include radiotherapy guidance and computer-assisted surgery.
The second task involves a brain tumor radiogenomic classification challenge. It requires participants to build models that use multiparametric MRI images to predict O6-methylguanine-DNA methyltransferase promoter methylation status. This task may improve efficiency and accuracy of diagnosis and treatment planning for glioblastoma patients.
Final submissions of AI algorithms are due October 12. For more information, visit here.