Measurement of subcutaneous adipose tissue thickness by infrared images during liposuction surgery: brief report

Abstract

Background: Currently, there are many techniques to measure subcutaneous body fat but these methods have many limitations. In this study, we tried to provide a clustering algorithm to measure the thickness of subcutaneous fat in thermography images. Methods: For the detection of subcutaneous adipose tissue in the midline area (from pubis to the xiphoid process), imaging takes place in the right or left lateral sides of the concerned person and to detect this tissue at the left and right flank (from ribs to the iliac crest), imaging takes place from the front. This study was done on 100 subjects (50 female, 50 male) of patients referred to the Shahid Mobini Hospital of Sabzevar since April 2013/4 to December, 2013 and the thickness of their subcutaneous fat in midline abdomen from pubis to the xiphoid process and flank from ribs to the iliac crest were measured based on thermal model and using K-Means and Fuzzy c-means (FCM) clustering methods and also recursive connected components algorithm. Results: Subcutaneous fat tissue can quickly appear in the thermogram as an area of low temperature and since in the thermal images, temperature is characterized by the color, as a result, subcutaneous fat tissue must have lower levels of color (temperature) relative to internal body tissues. All the measurements based-on thermal images to determine the maximum thickness of subcutaneous fat were compared with ultrasound. The results of our method were similar to the results of ultrasound method done by a radiologist, with the acceptable approximation. Conclusion: The method presented in this paper is considered as a noninvasive and cost-effective method to measure the thickness of subcutaneous body fat.

Keywords

• clustering
• components labeling
• subcutaneous fat
• thermography