Standard Plane Navigation using Fetal Ultrasound Description

  • Leveraged multi-modal vision-language models to establish a connection between textual descriptions and ultrasound (US) images for standard plane navigation.
  • Curated a dataset featuring text descriptions for each US image, showing the landmarks and artifacts.
  • Aiming to aid novice radiologists in quickly navigating to the correct standard planes while reducing variance among expert radiologists.
Rabin Adhikari
Rabin Adhikari
Research Assistant

Currently researching on semi-supervised and multi-modal learning, Medical Imaging, and Natural Language Processing