As part of the 26th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2023), the Neurofibromatosis Tumor Segmentation on Whole-Body MRI (WBMRI-NF) Challenge is now open for participants to begin work.
Neurofibromatosis (NF) is a group of genetic disorders that cause tumors to grow on nerve tissues anywhere in the body. Whole-Body MRI (WB-MRI) has become the standard imaging solution for early detection, monitoring, and surgical planning for NF-related clinical care and trials. MRI segmentation tasks often use U-Net models and variants, however this type of model shows significantly lower accuracy in segmenting NF tumors due to their high variation in tumor location, size, shape, infiltration, and heterogeneity.
Representatives from Intel, Massachusetts General Hospital (MGH), Harvard Medical School, NIH/National Cancer Institute, Sage Bionetworks, and University of Wisconsin-Madison have organized this challenge, which seeks new solutions for deep learning based segmentation.
Last year at MICCAI 2022 in Singapore, this consortium was a participant in the Multi-Modality Abdominal Multi-Organ Segmentation Challenge (AMOS 2022), finishing in 4th place. The goal of that challenge was to find a segmentation algorithm for a wide variety of organs, diseases, and data sources. The group is organizing this challenge to focus on WGMRI-NF, which is only a single disease but with a wide variety of characteristics.
This challenge will offer the first large-scale WBMRI dataset for NF tumors, consisting of WBMRI scans with radiologist-confirmed segmentation masks and volumetric tumor-burden reports. Participants will develop models for automated segmentation and detection of NF tumors on WBMRI scans.
The full challenge schedule is available on its website. The training data will be available starting April 1, and the final top-ranking teams will be announced September 1. The final winners announced at the MICCAI 2023, which is being held from October 8-12, 2023 in Vancouver Convention Centre Canada.
The long-term goal is to use this challenge to begin a research community focused on WBMRI image analysis techniques. Solving the challenges posed by such a complex disease requires collaboration between researchers and practitioners of medical image analysis, computer vision, and machine learning.
If you are interested in joining this challenge, learn more at the WBMRI-NF challenge website. And when putting together your end-to-end deep learning pipeline, be sure to check out the open source AI tools and frameworks optimized for Intel hardware.
Technical marketing manager for Intel AI/ML product and solutions. Previous to Intel, I spent 7.5 years at MathWorks in technical marketing for the HDL product line, and 20 years at Cadence Design Systems in various technical and marketing roles for synthesis, simulation, and other verification technologies.