Automatic Mixed Reality Guidance for Obstetric Ulrasound in Remote and Resource Limited Settings – Globalink Research Internship at the University of Oxford, UK

Ultrasound (US) is an important diagnostic tool in healthcare, especially in low- and middle-income countries (LMICs) where it is often the only available imaging modality. However, in remote and resource-limited settings, multiple barriers limit access to ultrasound including a shortage of professional sonographers and long travel distances. Teleultrasound, where non-experts perform US procedures under the guidance of a remote expert, helps overcome these barriers, yet this still relies on hiring a sonographer which is expensive and limits its applicability in LMICs. My internship with Dr. Alison Noble at the University of Oxford will focus on developing a novel machine learning model that can automatically estimate US probe pose and guide non-experts to standard US fetal imaging planes. This model will be trained on a comprehensive dataset I will collect during the internship which will be the first in the field to contain ultrasound video as well as probe orientation, position, and interaction forces. Through this research project I will gain an increasingly interdisciplinary perspective and learn from leaders in the robotics, medical imaging, and healthcare fields. This project will establish a collaboration between UBC’s Robotics and Control Laboratory and the Noble lab at Oxford, enhancing research programs in both groups.

Faculty Supervisor:

Septimiu (Tim) Salcudean

Student:

Partner:

University of Oxford

Discipline:

Engineering

Sector:

Health and Related Sciences & Technology; Artificial Intelligence; Technology

University:

The University of British Columbia

Program: