Measurement of a person’s size is, typically, a fundamental practice in health assessment. Waist girth is used to determine an individual’s risk to conditions associated with obesity and metabolic syndrome (type 2 diabetes, cardiovascular disease). However, waist girth is a blunt tool that misses a significant sub-population of normal weight, metabolically obese individuals. As a result, treatment can be ineffective or even absent.
We have developed a method of measuring a person’s shape as well as their size. Our method extracts complex shape parameters from 3D body scanner of their torso. We have shown that our measures of shape contain additional information compared to size alone, and that they improve predictions of fat amount and distribution (which are closely related to metabolic syndrome).
In order to measure a person’s shape we currently require large, high-cost body scanning equipment. In order for shape measurement to have future practical applications we must develop more accessible methods of measurement. We want to develop quicker and lower cost methods of measuring shape so that they can be used routinely in clinical practice.
We would expect the student to formulate appropriate aims and objectives, however, we have included a possible aim and project structure below to provide context.
The aim of this PhD is to develop a low-cost, clinically accessible method of torso shape measurement.
The project aim will be achieved by addressing 4 main objectives
· Review torso shape measurement methods and refine existing methods
· Review feasible methods of body shape measurement using single RGB or RGB-D cameras
· Develop and refine a body shape measurement tool
· Validate the new (low-cost) method against existing (full body scanner) methods.
The project will be centred around 4 studies to address these objectives
1. A review of existing shape methods with scope to adapt current algorithms
2. A scoping review examining state of the art computer vision techniques in shape identification and measurement
3. A pilot study to assess feasibility and efficacy of new shape method measurements
4. A cohort study to validate the new method against current methods – are they measuring the same thing?
Candidate Requirements
Applicants should have a strong undergraduate degree (2.1 or above) and/or a relevant Masters qualification in computer science, computer engineering or a related field, or be able to demonstrate efficacy in this area. Experience of computer vision using RGB or RGB-D cameras, experience with large datasets and related data processing and a desire to work with human participants as part of data collection is advantageous .
Although the project will be based primarily at Sheffield Hallam the student will also complete a period of research at La Trobe in Australia.
How to apply
We strongly recommend you contact the lead academic, Dr Simon Choppin, [Email Address Removed], to discuss your application.
Start date for studentship: February 2023
Interviews are scheduled for: TBC
For information on how to apply and more about the Joint Programme with La Trobe, Australia, please click Global partnership: Joint PhD programme with La Trobe University, Australia | Sheffield Hallam University (shu.ac.uk)
Your application should be emailed to [Email Address Removed] by the closing date of 31st October 2022.