In search of truth: evaluating concordance of AI-based anatomy segmentation models.

Lena Giebeler^1,2, Deepa Krishnaswamy², David Clunie³, Jakob Wasserthal⁴, Lalith Kumar Shiyam Sundar⁵, Andres Diaz-Pinto⁶, Klaus H Maier-Hein⁷, Murong Xu⁷, Bjoern Menze⁸, Steve Pieper⁹, Ron Kikinis², Andrey Fedorov²

¹RWTH Aachen University, Aachen, Germany.

²Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, United States.

³PixelMed Publishing, Bangor, Pennsylvania, United States.

⁴University Hospital of Basel, Basel, Switzerland.

⁵Ludwig Maximilian University Munich, Munich, Germany.

⁶NVIDIA, Santa Clara, California, United States.

⁷German Cancer Research Center, Heidelberg, Germany.

⁸University of Zurich, Zurich, Switzerland.

⁹Isomics Inc, Cambridge, Massachusetts, United States.

PubMed PMC DOI PDF

Abstract

PURPOSE: Artificial intelligence based methods for anatomy segmentation can help automate characterization of large imaging datasets. The growing number of similar functionality models raises the challenge of evaluating them on datasets that do not contain ground truth annotations. We introduce a practical framework to assist in this task. APPROACH: We harmonize the segmentation results into a st…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.