16
M. Tiwari and N. Anand
Standard Deviation (SD) of the difference between scanner measurements and the
manual measurements to check the comparability of the two measurement techniques
was calculated [5].
Step 2. Error limits were calculated using the mean of the differences (between
scanner measurement and manual measurement) for all the subjects and reported
with its associated standard deviation, sample size and 95% confidence Interval. If
the 95% confidence interval for the mean of scan-minus-manual measure differences
is within the plus or minus interval defined by the ISO 20685:2005 values, then the 3D
scanning system can be considered as acceptable as per the International standards
[6].
95% Standard Error = 1.96 ∗ std. Deviation of difference/SQRT(N)
Upper Limit = Mean difference(+)95% Standard Error
Lower Limit = Mean difference(−)95% Standard Error
Please refer Table 1 for the scanner validation analysis on the actual extracted
measurements from the scanner. Table 1 illustrates the summary statistics for scan
measurements-manual measurements for different body dimensions measured during
the scanner validation exercise. The upper limit and lower limits of the differences
as derived by following the process mentioned in step 2 were compared to the ISO
20685:2005 standard values for the permissible error. The result of the comparison of
error values obtained from scanner measurements to the ISO 20685:2005 permissible
error is shown as “Outcome” with P for Pass or F for Fail. It can be observed that the
upper and lower limit of all the differences were observed beyond the permissible
limit except WBL (Waist back length) and WH (Waist height), where the lower limit
and the upper limit was observed within the permissible range for WBL (Waist back
length) and WH (Waist height), respectively.
To accept the scanner extracted measurements, both the upper value and lower
value should be within the ISO 20685:2005 prescribed limit. For example, for Stature
the upper limit and lower limit are 1.42 and 1.17 cm, respectively, while the ISO
20685:2005 permissible error value is ± 0.4 cm. Here both the values of the upper
limit and lower limit are beyond the ISO 20685:2005 permissible error value, hence
all the scanner extracted measurements were considered as “FAIL”.
Further, the same procedure was followed while checking the accuracy as per
ISO 8559:1989 standard by comparing the difference between scanner measurements and manual measurement to the ISO 8559:1989 values (an accuracy of ± 1%
or ± 5.0 m, whichever is the smaller) [7]. As indicated in Table 1, While comparing
the mean difference of scanner measurement – manual measurements to the ISO
8559:1989 prescribed error limits, it was observed that all the differences were
observed beyond the permissible limit. However for the WH (Waist height), the
upper limit was observed in the ISO Range (Error of −0.43 against standard acceptable error value of −0.497) and a lower limit was observed out of the ISO Range
(Error of −1.02 against standard acceptable error value of −0.497), hence all the
scanner extracted measurements were considered as “FAIL”.
M. Tiwari and N. Anand
Standard Deviation (SD) of the difference between scanner measurements and the
manual measurements to check the comparability of the two measurement techniques
was calculated [5].
Step 2. Error limits were calculated using the mean of the differences (between
scanner measurement and manual measurement) for all the subjects and reported
with its associated standard deviation, sample size and 95% confidence Interval. If
the 95% confidence interval for the mean of scan-minus-manual measure differences
is within the plus or minus interval defined by the ISO 20685:2005 values, then the 3D
scanning system can be considered as acceptable as per the International standards
[6].
95% Standard Error = 1.96 ∗ std. Deviation of difference/SQRT(N)
Upper Limit = Mean difference(+)95% Standard Error
Lower Limit = Mean difference(−)95% Standard Error
Please refer Table 1 for the scanner validation analysis on the actual extracted
measurements from the scanner. Table 1 illustrates the summary statistics for scan
measurements-manual measurements for different body dimensions measured during
the scanner validation exercise. The upper limit and lower limits of the differences
as derived by following the process mentioned in step 2 were compared to the ISO
20685:2005 standard values for the permissible error. The result of the comparison of
error values obtained from scanner measurements to the ISO 20685:2005 permissible
error is shown as “Outcome” with P for Pass or F for Fail. It can be observed that the
upper and lower limit of all the differences were observed beyond the permissible
limit except WBL (Waist back length) and WH (Waist height), where the lower limit
and the upper limit was observed within the permissible range for WBL (Waist back
length) and WH (Waist height), respectively.
To accept the scanner extracted measurements, both the upper value and lower
value should be within the ISO 20685:2005 prescribed limit. For example, for Stature
the upper limit and lower limit are 1.42 and 1.17 cm, respectively, while the ISO
20685:2005 permissible error value is ± 0.4 cm. Here both the values of the upper
limit and lower limit are beyond the ISO 20685:2005 permissible error value, hence
all the scanner extracted measurements were considered as “FAIL”.
Further, the same procedure was followed while checking the accuracy as per
ISO 8559:1989 standard by comparing the difference between scanner measurements and manual measurement to the ISO 8559:1989 values (an accuracy of ± 1%
or ± 5.0 m, whichever is the smaller) [7]. As indicated in Table 1, While comparing
the mean difference of scanner measurement – manual measurements to the ISO
8559:1989 prescribed error limits, it was observed that all the differences were
observed beyond the permissible limit. However for the WH (Waist height), the
upper limit was observed in the ISO Range (Error of −0.43 against standard acceptable error value of −0.497) and a lower limit was observed out of the ISO Range
(Error of −1.02 against standard acceptable error value of −0.497), hence all the
scanner extracted measurements were considered as “FAIL”.