Page 131 - IJB-7-4

Li, et al
and were able to produce good quality scans. Therefore, 4114 2.2 2400MHz 10-Core CPU, 256 GB DDR4 2666
each scan was manually checked against the following DIMM Memory, Nvidia Quadro P1000 4GB GFX). For
set of exclusion criteria to ensure that all scans were of each input scan, a solid respirator mask CAD model was
similar quality before submitted to the pipeline: generated by the pipeline described in Section 2.1. The
i. Being duplicates, that is, having two or more total run time for processing a scan and generating a CAD
submissions containing the same face model was recorded.
ii. Having any obstruction on the face that may interfere
with the mask, including beard, moustache, glasses, 2.5. Fit evaluation
and piercing. Euclidean distance was used to computationally evaluate
iii. Having poor quality, including poor reconstruction fit between the mask surface and the aligned face mesh
of facial geometry, corrupted file, scan resolution at (Figure 4). A nearest neighbor search was performed
[70]
>0.5 mm through a space-partitioning method called K-dimensional
iv. Having been modified by the participant to remove tree to pair each vertex on the mask surface with its
[71]
any holes/defects nearest neighboring vertex on the aligned mesh. The
v. Having non-neutral facial expression nearest neighbor search was defined as: given a set of
vi. Not human faces, that is, scans of other objects points u ∈ U (the aligned mesh), and a set of query points
vii. Being manifold, that is, enclosed to form a solid volume, v ∈ V (the mask surface), for all v, find the closest points
instead of being an open surface to U. The Euclidean distance was computed for each
closest pair. Subsequently, the maximum and root mean
A total of 322 submissions were received at the square error (RMSE) of the Euclidean distances of were
end of recruitment, of which 117 were excluded based computed. The Maximum Euclidean distance indicates the
on the above criteria. Figure 3 shows a summary of the maximum gap between a mask and its corresponding face
excluded scans. mesh, whereas the RMSE Euclidean distance indicates the
average gap between the mask and the face.
2.4. Computational time evaluation Maximum and RMSE Euclidean distance results
All 205 included scans were processed through the were grouped into age, gender, ethnicity, and BMI
pipeline on a remote Linux workstation (Intel® XEON® subcategories according to demographic data reported by
participants. For age, results have been grouped into three
subcategories: Young (18 – 39 years old), Middle aged
(40 – 54 years old), and Seniors (55 years old and above).
There is not an international standard on age classification
according to craniofacial shape change; however, various
studies have shown evidence of craniofacial change as a
result of aging [72-75] . Therefore, it is important to investigate
whether the current pipeline can deliver similar fit results
across different age groups. Age has generally been
grouped into three categories (young, middle aged, and
senior), and the cutoff points are approximately 15 – 25,
35 – 45, and 55 – 65 years old. In this study, the author,
therefore, used 18 – 39 years old for young adults, 40 –
54 years old for middle aged adults, and 55 years and above
for seniors. For gender, results were grouped into male
Figure 3. Facial scans exclusion. and female categories. BMI grouping was based on the

Figure 4. Euclidean distance between mask surface and a aligned face scan.

International Journal of Bioprinting (2021)–Volume 7, Issue 4 127