Utilizing Sonographer Visual Attention for Probe Movement Guidance in Cardiac Point of Care Ultrasound

Research output: Contribution to Book/Report typesContribution to conference proceedingspeer-review

Abstract

We present a proof-of-concept of a probe guidance system for cardiac ultrasound (US) aiming to assist novice sonographers during the imaging of the Apical 4 Chamber and Parasternal Long Axis standard views. Our system treats probe guidance as a sonographer behavior cloning task and employs a recurrent neural network to generate a probe-orientation guidance signal solely from US images. We trained the network using mainly US images and probe orientation data collected during US phantom scans performed by volunteer sonographers. However, our training method also utilizes an additional sonographer visual attention modality to feed an auxiliary training task. The evaluation of our system shows that it generates a guidance signal with an accuracy of at least 80% for the following angular ranges: [3°, 15°] for the y and z axes, and [4°, 15°] for the x-axis.

Original languageEnglish
Title of host publication2023 IEEE International Symposium on Biomedical Imaging, ISBI 2023
PublisherIEEE Computer Society
ISBN (Electronic)9781665473583
DOIs
Publication statusPublished - 2023
Event20th IEEE International Symposium on Biomedical Imaging, ISBI 2023 - Cartagena, Colombia
Duration: 18 Apr 202321 Apr 2023

Publication series

NameProceedings - International Symposium on Biomedical Imaging
Volume2023-April
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Conference

Conference20th IEEE International Symposium on Biomedical Imaging, ISBI 2023
Country/TerritoryColombia
CityCartagena
Period18/04/2321/04/23

!!!Keywords

  • Point-of-care ultrasound (POCUS)
  • ResNet
  • attention module
  • probe guidance
  • visual attention

Fingerprint

Dive into the research topics of 'Utilizing Sonographer Visual Attention for Probe Movement Guidance in Cardiac Point of Care Ultrasound'. These topics are generated from the title and abstract of the publication. Together, they form a unique fingerprint.

Cite this