Document Type
Conference Proceeding
Publication Date
7-1-2020
Abstract
The increasing prevalence and adaptability of 3D optical scan (3DO) technology has invoked many recent studies which use 3DO scanning as a convenient and inexpensive means for predicting body composition and health risks. The Shape Up studies seek a device-agnostic solution for body composition estimation based on principal component analysis (PCA). This paper reports a progress made on Shape Up's previous work which served as a criterion analysis for PCA-based body composition and health risk prediction. This study presents proof-of-concept for a novel automated landmark detection step that allows for a fully automated PCA-based approach to body composition estimation that facilitates a practical device-agnostic PCA-based solution to body composition estimation from 3DO scans. Our results show that replacing expensive and time-consuming manual point placement with the proposed automated landmarks will not diminish the quality of body composition estimates allowing for a more practical pipeline that can be used in real-world settings.
Publication Source (Journal or Book title)
Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society EMBS
First Page
1853
Last Page
1858
Recommended Citation
Sobhiyeh, S., Borel, N., Dechenaud, M., Graham, C., Wong, M., Wolenski, P., Shepherd, J., & Heymsfield, S. (2020). Fully Automated Pipeline for Body Composition Estimation from 3D Optical Scans using Principal Component Analysis: A Shape Up Study. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society EMBS, 2020-July, 1853-1858. https://doi.org/10.1109/EMBC44109.2020.9175211