Accurate localization of inner ear regions of interests using deep reinforcement learning

Ana-Teodora Radutoiu*, Francois Patou, Jan Margeta, Rasmus Reinhold Paulsen, Paula Lopez Diez

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

115 Downloads (Pure)

Abstract

We propose a novel method for automatic ROI extraction. The method is implemented and tested for isolating the inner ear in full head CT scans. Extracting the ROI with high precision is in this case critical for surgical insertion of cochlear implants. Different parameters, such as CT equipment, image quality, anatomical variation, and the subject’s head orientation during scanning make robust ROI extraction challenging. We propose to use state-of-the-art communicative multi-agent reinforcement learning to overcome these difficulties. We specify landmarks specifically designed to robustly extract orientation parameters such that all ROIs have the same orientation and include the relevant anatomy across the dataset. 140 full head CT scans were used to develop and test the ROI extraction pipeline. We report an average overall estimated error for landmark localization of 1.07 mm. Extracted ROI presented an intersection over union of 0.84 and a Dice similarity coefficient of 0.91.
Original languageEnglish
Title of host publicationProceedings of the 13th International Workshop on Machine Learning in Medical Imaging
PublisherSpringer
Publication date2022
Pages416-424
ISBN (Print)9783031210136
DOIs
Publication statusPublished - 2022
Event13th International Workshop on Machine Learning in Medical Imaging - Resort World Convention Centre Singapore, Singapore, Singapore
Duration: 18 Sept 202218 Sept 2022
Conference number: 13

Conference

Conference13th International Workshop on Machine Learning in Medical Imaging
Number13
LocationResort World Convention Centre Singapore
Country/TerritorySingapore
CitySingapore
Period18/09/202218/09/2022

Keywords

  • Region of interest
  • Deep reinforcement learning
  • Computed tomography
  • Inner ear
  • Landmarks
  • Orientation

Fingerprint

Dive into the research topics of 'Accurate localization of inner ear regions of interests using deep reinforcement learning'. Together they form a unique fingerprint.

Cite this