Validation and Reliability of Sizestream 3D Scanner for Human Body Measurement

Abstract

3D scanners are widely used in industrial design, engineering, and manufacturing as they are able to capture the required data quickly in a precise and reliable manner, compared to manual methods, which can be too costly and time-consuming. 3D whole-body scanners have been a preferred means of data collection for many countries who have undertaken the National Sizing Survey of their respective country. To ensure the accuracy of the measurements, it is imperative to test and compare the 3D scan measurements to the manually taken body measurements. This research paper describes the procedure to check the validity and reliability of the anthropometric measurements extracted from 3D whole Body Scanner in comparison to the measurements taken through the traditional manual method. For this research, the process of establishing validity and reliability was formulated in ISO 20685:2005 and ISO 20685-1:2018. In total fifty subjects (25 male and 25 female subjects) of different body shapes/body types were measured manually as well as using 3D wholebody scanner by Sizestream – SS14. Intra-class correlation coefficients (ICC) was used to check for the consistency and repeatability between different scan measurements. The mean difference (Mdiff) between scanner extracted measurement and manually taken measurement was calculated and the standard deviation of this difference was arrived at to determine the upper and lower limits at 95% confidence limits. The values were further compared with the acceptable standard values as prescribed in ISO 8559:1989 and ISO 20685:2005. It was observed that the Sizestream—SS14 scanners used in the validation exercise were consistent in taking human body measurements, however, it indicated a consistent systematic error which leads to missing achieving of expected accuracy level between manual and scanner measurements (as per protocols of ISO 8559:1989 and ISO 20685:2005) for the majority of measurements. Subsequently, a correction factor in the form of bias-shift was determined, and the same was incorporated into the scanner extracted measurements. This corrected the systematic error and provided the desired accuracy in scanner measurement for all the anthropometric dimensions making the data acceptable for use for any further analysis.