Segmentation of visceral and subcutaneous fat tissues in computed tomography images using finite mixture models

• Grant number: 16/19021-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: December 01, 2016
• End date: November 30, 2017
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Ricardo José Ferrari
• Grantee: Vinícius Ito Nagura
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Associated research grant: 13/00798-2 - The extracellular matrix in aging, exercise and in the tumor microenvironment, AP.TEM

Abstract

Osteoarthritis is a joint disorder that leads to aging and degeneration of the cartilage and underlying bone. The increase in fat mass overloads joints like knees and hips, therefore it is strongly associated with osteoarthritis. However, the prevalence of osteoarthritis is increasing in other joints, such as hands, which do not suffer from overload. Recent studies have speculated that, in addition to overloading, leptin at high levels, resulted from the consumption of saturated fatty acids in excess, assumes inflammatory characteristics and acts on the articular cartilage, triggering an inflammatory and degenerative process - the osteoarthritis. Volume measurement of adipose tissue for the study of diseases related to obesity is currently performed manually (or semi manually) outlining the fat regions in 3D computed tomography images (CT). This procedure, in addition to be time consuming for the radiologist, is prone to a wide variability intra- and inter-observers. In this context, this research project aims to develop an automatic computational technique for segmentation of visceral and subcutaneous adipose tissues in 3D CT images using both Gaussian and Student's-t finite mixture models. The developed technique will be quantitatively assessed using images from an osteoarthritis study acquired at UFSCar's University Hospital. (AU)