Name
Abstract | ||
|---|---|---|
The ability to accurately measure subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) from images is important for improved assessment and management of patients with various conditions such as obesity, diabetes mellitus, obstructive sleep apnea, cardiovascular disease, kidney disease, and degenerative disease. Although imaging and analysis methods to measure the volume of these tissue components have been developed [1, 2], in clinical practice, an estimate of the amount of fat is obtained from just one transverse abdominal CT slice typically acquired at the level of the L4-L5 vertebrae for various reasons including decreased radiation exposure and cost [3-5]. It is generally assumed that such an estimate reliably depicts the burden of fat in the body. This paper sets out to answer two questions related to this issue which have not been addressed in the literature. How does one ensure that the slices used for correlation calculation from different subjects are at the same anatomic location? At what anatomic location do the volumes of SAT and VAT correlate maximally with the corresponding single-slice area measures? To answer these questions, we propose two approaches for slice localization: linear mapping and non-linear mapping which is a novel learning based strategy for mapping slice locations to a standardized anatomic space so that same anatomic slice locations are identified in different subjects. We then study the volume-to-area correlations and determine where they become maximal We demonstrate on 50 abdominal CT data sets that this mapping achieves significantly improved consistency of anatomic localization compared to current practice. Our results also indicate that maximum correlations are achieved at different anatomic locations for SAT and VAT which are both different from the L4-L5 junction commonly utilized. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1117/12.2044254 | Proceedings of SPIE |
Keywords | Field | DocType |
body fat quantification,CT imaging,landmarks,image standardization | Computer vision,Obstructive sleep apnea,Data set,Abdomen,Anatomic Location,Clinical Practice,Kidney disease,Correlation,Adipose tissue,Artificial intelligence,Radiology,Physics | Conference |
Volume | ISSN | Citations |
9034 | 0277-786X | 0 |
PageRank | References | Authors |
0.34 | 1 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yubing Tong | 1 | 93 | 22.73 |
| Jayaram K. Udupa | 2 | 2481 | 322.29 |
| D. A. Torigian | 3 | 81 | 21.68 |