BONBID-HIE2023: 1st Workshop on HIE Lesion Segmentation
Introduction
Hypoxic ischemic encephalopathy (HIE) is a brain injury that occurs in 1 ∼ 5/1000 term-born neonates. HIE lesion detection is a crucial step in the clinical care of HIE. It could lead to a more accurate estimation of prognosis, a better understanding of neurological symptoms, and a timely prediction of response to therapy in this population. The rise of Artificial Intelligence (AI) brings hope to objectively and accurately finding HIE lesions. In order to bring together researchers interested in HIE lesion detection and facilitate research in this area, we release the first HIE dataset: BOston Neonatal Brain Injury Dataset for Hypoxic Ischemic Encephalopathy (BONBID-HIE), and host the first benchmark challenge on HIE Lesion Segmentation. The workshop consists of presentations by clinical and machine learning experts in the field and short talks regarding methods addressing the benchmark challenges in this disease.
BONBID-HIE Lesion Segmentation Challenge
We establish one challenge this year:
- BONBID-HIE Lesion Segmentation: to predict a binary mask of HIE lesion.
HIE Lesion Detection
Participants of the challenge are provided with prepared training, test datasets, and automated evaluation scripts. The top teams of the challenge will be invited to give a short talk describing their method during this workshop.
The challenge is hosted on Grand Challenge. If you would like to join the challenge, see more details here:
Important Dates
Training Data Release and Registration Open | 7/1/2023 |
Development Stage | 7/1/2023-8/31/2023 |
Eval Stage | 8/15/2023-9/20/2023 |
Test Stage | 9/1/2023-9/20/2023 |
Method Description and Codes Submission | 9/15/2023-9/20/2023 |
Notifications of Challenge Winners | 10/1/2023 |
BONBID-HIE 2023 Workshop Date | 10/16/2023 10:00 AM - 1:00 PM (EST) |
BONBID-HIE 2023 Workshop
10/16/2023 10:00 AM - 1:00 PM (EST) Zoom Link |
Website and Arrangements 1st BONBID-HIE Lesion Segmentation Workshop |
Workshop Video
The 3-hour-long workshop will be recorded and uploaded to youtube and this workshop website.
Invited Speakers
Seetha Shankaran, M.D. neonatologist at the Children’s Hospital of Michigan was the principal investigator of the 1st randomized controlled trial of whole body cooling for neonatal hypoxic ischemic encephalopathy (HIE). Considered a pioneer in finding effective methods for protecting HIE-affected newborns, she led a landmark 2005 study in the New England Journal of Medicine to help make the 72-hour “cooling blanket” procedure the standard of care in treating this condition. Several of her subsequent papers in the New England Journal of Medicine and JAMA have further tested the optimization of cooling therapy in newborns and long term outcomes of cooled infants. She was also the first author in the publications which have validated a scoring system for MRI as a prognostic marker for neonatal encephalopathy. |
Organizers
Postdoctoral Research Fellow, Boston Children's Hospital, Harvard Medical School
Senior Organizers
Associate Professor of Radiology, Harvard Medical School
Director of Fetal-Neonatal Neuroimaging and Developmental Science Center, Professor of Radiology and Pediatrics, Harvard Medical School
Contact
rina.bao@childrens.harvard.edu
Acknowledgments
Thanks to visualdialog.org for the webpage format.