Excess body fat is a key underlying factor in the development of numerous chronic diseases, including type II diabetes, heart disease, stroke, and cancer. The AMA recently declared that obesity, itself, is a disease. Most epidemiologic studies utilize Body Mass Index (BMI) to classify people as underweight, normal, overweight, or obese because it is a convenient and simple method that has been shown to correlate with disease risk. Since the majority of the health risks associated with obesity are more directly linked to an overabundance of body fat than weight, measuring body fat is essential for more precise guidelines. However, accurate methods of assessing body fat are expensive, inconvenient, and require immobile equipment. Consequently, the AMA has called for more cost effective and convenient methods to assess body composition to assist doctors in their assessment and treatment. Virtual modeling of humans in particular has provided ways to scan and analyze the body and its motion. Supervised Machine Learning (SML), a sub-field of artificial intelligence, has made great progress in taking measured data to infer new relationships. It is our belief that virtual modeling and SML can provide the techniques necessary to conveniently and accurately calculate the percentage of body fat (%BF) and to provide new tools in treating obesity based on body shapes. The project will develop a system that uses commercially available depth cameras such as the Microsoft Kinect(r) to capture the surface of the human body. This will be accomplished by developing a new algorithm to perform deformable registration of several RGB-Depth views of the body. A new algorithm that uses SML will be developed to calculate percentage body fat using the surface data. The system will be trained and validated by collecting data from a number of subjects. The surface captured will be used to explore the role of visual body representation in motivation and adherence. The developed systems can be implemented in clinical or personal settings and be utilized as a public health research tool and deployed widely given the low-cost of the hardware required. In addition to the immediate impact that the system will have on managing obesity, the project will have a broad impact on a number of areas. A large database of such shapes captured over time may lead to ways to predict how an individual's body shape will change given a particular intervention. Certain medical conditions that result in body shape change, such as those involving lymphatic circulations, may be diagnosed and tracked more easily. Growth patterns of children may be tracked by change of body shapes. Further research can be conducted to determine the effect of body shape on %BF using data mining techniques.

Public Health Relevance

The project will develop a new method to capture the 3D surface and shape of a human body and a new method to use these data to calculate percent body fat. By making these tools widely available and economical, the proposed approach has the potential for major contributions in the assessment and treatment of obesity.

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Kidney, Nutrition, Obesity and Diabetes (KNOD)
Program Officer
Fine, Larry
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
George Washington University
Biostatistics & Other Math Sci
Biomed Engr/Col Engr/Engr Sta
United States
Zip Code