student profile: Mr Sam Houston


Thesis work

Thesis title: Low Cost Non Invasive Body Composition Measurement

Supervisors: Philip Heng Wai LEONG , Alistair MCEWAN

Thesis abstract:

Human Body Composition Measurementhas evolved since the 1940s. Studies in male cadavers created early information in the science. It is important that body composition remains in a healthy range. Obesity leads to metabolic syndrome and related diseases including diabetes, and coronary heart disease. It is also important to manage nutrition in those who are ill. Body composition is a strong nutritional status indicator. A current gold standard for measuring composition is the molecular 4 compartment (4C) model, where the body is assumed to consist of: fat, water, mineral, and protein. This is the model to which other methods are baselined. However the techniques and tools used to determine body composition in this model are prohibitively expensive, require highly skilled resources, and take several hours for each subject. They are typically only affordable in research institutions. A number of non-invasive, relatively fast methods to determine body composition have emerged and evolved such as Bioelectrical Impedance Analysis (BIA), and Near InfraRed Spectroscopy (NIRS). The algorithms created for these offerings have been created using linear regression against the 4C model and are therefore highly population dependent and typically This research aims to examine current methods and create a fast, low-cost non-invasive approach that should work for any human by approaching the problem from a data science perspective.

Note: This profile is for a student at the University of Sydney. Views presented here are not necessarily those of the University.