News | Artificial Intelligence | August 10, 2017

RSNA Announces Pediatric Bone Age Machine Learning Challenge

Participants will design algorithms to assess bone age from pediatric radiographs; results will be presented at RSNA 2017 annual meeting

RSNA Announces Pediatric Bone Age Machine Learning Challenge

August 10, 2017 — The Radiological Society of North America (RSNA) is organizing a challenge intended to show the application of machine learning and artificial intelligence on medical imaging and the ways in which these emerging tools and methodologies may improve diagnostic care.

The RSNA Pediatric Bone Age Machine Learning Challenge addresses a familiar image analysis activity for pediatric radiologists: assessment of bone age from hand radiographs of pediatric patients used to evaluate growth and diagnose developmental disorders. The Challenge uses a dataset of hand radiographs provided by a consortium of leading research institutions — Stanford University, the University of California, Los Angeles and the University of Colorado — that have associated bone age assessments provided by multiple expert observers.

Participants in the challenge will be judged by how well the bone age evaluations produced by their algorithms accord with the expert observers’ evaluations. Participants will have the opportunity to directly compare their algorithms in a structured way using this carefully curated dataset. The RSNA Machine Learning Challenge organizing committee will select a small group of the most successful entries for recognition at the RSNA annual meeting, Nov. 26-Dec. 1 in Chicago. Recognition of Challenge participants will be part of a broad range of educational events and exhibits focusing on machine learning at the RSNA annual meeting. 

 The milestone activities scheduled for the Challenge include:

  • Training data phase: Aug. 1-30
  • Leader board phase: Sept. 1-Oct. 7
  • Submission of results: Oct. 7-15
  • Review and confirmation of results: Oct. 15
  • Notification of awardees: Oct. 15
  • Public announcement of results: Monday, Nov. 27 at the RSNA 2017 annual meeting

At the challenge site, prospective participants may review the terms and conditions, evaluation criteria and other details of participation as well as, register to participate and download the training dataset to begin the Challenge.

The Challenge is hosted on the MedICI platform (built by CodaLab) provided by Jayashree Kalpathy-Cramer, Ph.D., and Massachusetts General Hospital, supported through NIH grants and a contract from Leidos.

For more information: www.rsnachallenges.cloudapp.net

Related Content

Breast Cancer Screening Performance Impacted by Mean Mammographic Compression Pressure
News | Mammography | December 08, 2017
Dutch researchers demonstrated a strong relationship between compression pressure in mammography and breast cancer...
Sponsored Content | Videos | Magnetic Resonance Imaging (MRI) | December 07, 2017
Max Wintermark, M.D., professor of radiology and chief of neuroradiology, Stanford Hospital and Clinics, discussed MR
Technology | Neuro Imaging | December 07, 2017
Deep learning artificial intelligence (AI) medical company DeepRadiology  announced the world's first computed...
Sponsored Content | Videos | Advanced Visualization | December 07, 2017
Dianna Bardo M.D., director of body MR and co-director of the 3-D Innovation Lab at Phoenix Children's Hospital.
Sponsored Content | Videos | Artificial Intelligence | December 07, 2017
Adam Flanders, M.D., co-director, neuroradiology and vice-chair of informatics at Jefferson University Hospitals, Phi
Toshiba Highlights Ultra-High Resolution CT at RSNA 2017
News | Computed Tomography (CT) | December 06, 2017
December 6, 2017 — Toshiba Medical, a Canon Group company, showcased the Aquilion Precision, what it calls the world’
Canon Showcases New CXDI Digital Radiography Detectors at RSNA 2017
News | Digital Radiography (DR) | December 06, 2017
Canon U.S.A. Inc. and Virtual Imaging Inc. recently announced the launch of the CXDI-710C, CXDI-810C and CXDI-410C...
Medexprim's Radiomics Enabler Goes Open Source
Technology | PACS Accessories | December 06, 2017
December 6, 2017 — Medexprim announced its Radiomics Enabler, meant to enhance medical images archives for research,
How to allow legacy PACS systems to use a global worklist even with older, disparate PACS systems. Carestream.
Sponsored Content | Whitepapers | PACS | December 06, 2017
Myth: Legacy PACS systems in an M&A environment can not support a global worklist.
Sponsored Content | Videos | Mammography | December 06, 2017
Martin Yaffe, Ph.D., FAAPM, senior scientist, physical sciences/imaging research, Sunnybrook Health Sciences Centre,
Overlay Init