Validation and Reliability of Sizestream 3D Scanner …
21
the dimensions was within the ISO 8559:1989 limits. Hence, the scanner extracted
measurements were considered as “PASS”.
In view of the observed difference between measurement reported by the scanner
and one took manually, the same set of subjects were scanned by two other SS14
scanners, under the same scanning conditions, to examine scanner consistency and
eliminate the possibility of scanner misreporting. It was seen that similar results (i.e.
deviations) were reported from all the three scanners. Intraclass correlation ICC was
calculated between the measurements extracted from all the three scanners and the
result of ICC ranged between 0.997 to 1.00 for all the anthropometric measurements.
This concludes that scanners had high levels of consistency with a systematic error.
This pattern has been observed with other scanner technologies as well.
Han et al. [5] conducted a comparative analysis of 3D body scan measurements
and manual measurements of Size Korea adult females using 3D whole-body laser
Scanning technology of Cyberware (WB4) and observed that scanner measurements
were generally larger than manual measurements with that the mean differences of
circumferences being larger than those of lengths and heights [5].
Mckinnon and Istook [1] undertook a similar comparative study of 10 anthropometric measurements of dress forms extracted from two 3D whole-body white
light scanning technology i.e. the Image Twin (2T4) and the 3T6, with the measurements obtained manually and reported that the scanned measurements from both the
scanning systems exceeded the manual measurements except crotch circumference
which was found consistently smaller [1].
Gordon et al. [14] in Anthropometric Survey of U.S. Army Personnel- Methods
and Summary Statistics report (ANSUR II conducted using 3D whole-body laser
Scanning technology Cyberware WBX 3D) stated that on comparison of measurement extracted through traditional anthropometry methods with the one extracted
using 3D Scanning technology, scan-generated measurements were observed to be
significantly larger [14].
Koepke et al. [9] while comparing 3D laser-based photonic scans (using 3D wholebody laser Scanning technology of Human Solution—Vitus Smart XXL) and manual
anthropometric measurements for Swiss men reported the scanned derived values of
height were found systematically shorter by 2.1 cm, while the waist, hip and bust
circumferences were observed larger in the scans by 1.17–4.37 cm [9].
Šimenko and ˇ
Cuk [15] in testing Reliability and Validity of 3D whole-body white
light technology of TC2 NX-16 confirmed that 3D body scan values were larger but
systematic with a high level of consistency [15].
There have been many such similar studies by Kouchi [6], Kuehnapfel et al. [12],
Wang et al. [16], Wells et al. [17], Zancanaro et al. [18], Bougourd et al. [19], Paquette
et al. [20], Robinettea and Daanend [21], and Weinberg et al. [22] where comparison
of 3D scan extracted measurements to manual measurements using various 3D wholebody scanning technologies have been done and deviations recorded from those
of manual measurements. These researches have also confirmed that the scanned
derived measurements show a high level of consistency in comparison to manual
measurements [6, 12, 16–22].
21
the dimensions was within the ISO 8559:1989 limits. Hence, the scanner extracted
measurements were considered as “PASS”.
In view of the observed difference between measurement reported by the scanner
and one took manually, the same set of subjects were scanned by two other SS14
scanners, under the same scanning conditions, to examine scanner consistency and
eliminate the possibility of scanner misreporting. It was seen that similar results (i.e.
deviations) were reported from all the three scanners. Intraclass correlation ICC was
calculated between the measurements extracted from all the three scanners and the
result of ICC ranged between 0.997 to 1.00 for all the anthropometric measurements.
This concludes that scanners had high levels of consistency with a systematic error.
This pattern has been observed with other scanner technologies as well.
Han et al. [5] conducted a comparative analysis of 3D body scan measurements
and manual measurements of Size Korea adult females using 3D whole-body laser
Scanning technology of Cyberware (WB4) and observed that scanner measurements
were generally larger than manual measurements with that the mean differences of
circumferences being larger than those of lengths and heights [5].
Mckinnon and Istook [1] undertook a similar comparative study of 10 anthropometric measurements of dress forms extracted from two 3D whole-body white
light scanning technology i.e. the Image Twin (2T4) and the 3T6, with the measurements obtained manually and reported that the scanned measurements from both the
scanning systems exceeded the manual measurements except crotch circumference
which was found consistently smaller [1].
Gordon et al. [14] in Anthropometric Survey of U.S. Army Personnel- Methods
and Summary Statistics report (ANSUR II conducted using 3D whole-body laser
Scanning technology Cyberware WBX 3D) stated that on comparison of measurement extracted through traditional anthropometry methods with the one extracted
using 3D Scanning technology, scan-generated measurements were observed to be
significantly larger [14].
Koepke et al. [9] while comparing 3D laser-based photonic scans (using 3D wholebody laser Scanning technology of Human Solution—Vitus Smart XXL) and manual
anthropometric measurements for Swiss men reported the scanned derived values of
height were found systematically shorter by 2.1 cm, while the waist, hip and bust
circumferences were observed larger in the scans by 1.17–4.37 cm [9].
Šimenko and ˇ
Cuk [15] in testing Reliability and Validity of 3D whole-body white
light technology of TC2 NX-16 confirmed that 3D body scan values were larger but
systematic with a high level of consistency [15].
There have been many such similar studies by Kouchi [6], Kuehnapfel et al. [12],
Wang et al. [16], Wells et al. [17], Zancanaro et al. [18], Bougourd et al. [19], Paquette
et al. [20], Robinettea and Daanend [21], and Weinberg et al. [22] where comparison
of 3D scan extracted measurements to manual measurements using various 3D wholebody scanning technologies have been done and deviations recorded from those
of manual measurements. These researches have also confirmed that the scanned
derived measurements show a high level of consistency in comparison to manual
measurements [6, 12, 16–22].