The results are in! The American College of Radiology (ACR) and Society for Imaging Informatics in Medicine (SIIM) announced the winners of the groups’ machine learning challenge during SIIM’s Conference on Machine Learning in Medical Imaging in Austin, Texas.
More than 1,400 teams participated in the Machine Learning Challenge on Pneumothorax Detection and Localization, developing detection algorithms that could potentially speed up patient care. A total of 352 teams submitted results during the competition’s evaluation phase.
The challenge ran on a Kaggle platform using data from the National Institutes of Health. The Society of Thoracic Radiology and MD.ai also contributed.
The challenge’s top 10 winning teams were:
1. [dsmlkz] sneddy
4. [ods.ai] amirassov
7. See & Eduardo
8. Ian Pan & Felipe Kitamura
9. [ods.ai] Scizzo
10. [ods.ai] Yury & Konstantin
“SIIM is very pleased to have cooperated with the ACR, Google, Kaggle and the Society of Thoracic Radiology in hosting this challenge,” Steve Langer, PhD, CIIP, informatics physicist and radiology imaging architect at Mayo Clinic and co-chair of the SIIM Machine Learning Committee, said in a prepared statement. “In addition to the medical and data science aspects, SIIM introduced the use of FHIR and DICOMweb in a medical imaging data challenge for the first time in Kaggle’s history, as those API’s are key in moving AI tools into clinical production.”
“Kaggle challenges like this one now incorporate some useful parameters that are more likely to result in the winners producing AI tools with potential for clinical production,” Bibb Allen Jr., MD, ACR Data Science Institute chief medical officer, said in the same statement. “Congratulations to the winners. They have developed new healthcare solutions that may one day improve patient care.”
Prior coverage of this challenge can be read here.