Page 81 - IJB-8-1

Liu, et al.
and subtractive techniques like milling or machining , implant at the time of placement is dependent on the
[6]
AM produces parts from 3D digital data using the layer- implant engaging the bone surrounding the tooth socket.
by-layer process of joining raw materials. For implants, If patients have insufficient bone around the tooth socket,
powder bed fusion (PBF) is the most established AM or if vital structures such as the maxillary sinus or inferior
processes. In PBF, layers of powder are thermally fused dental nerve are located close to the intended implant site,
using an energy source, which can be lasered in laser PBF insertion of an immediate implant will not be possible.
(L-PBF) or electron beam in electron beam PBF (EB- These standard-sized implants do not fit the shape of
PBF). L-PBF is also commonly referred to as selective the socket exactly, hence a lack of intimate contact with
laser melting (SLM) while EB-PBF is called electron surrounding alveolar bone is likely to compromise initial
beam melting. implant stability. Under such circumstances, patients
L-PBF has the capability to produce highly complex will need to undergo bone grafting for delayed implant
metallic parts with densities and mechanical performance placement. Having an intimate adaptation of the implant
matching those from conventional methods [7-9] . However, against its bony socket walls is important for promoting
geometrical accuracy remains a challenge due to primary stability and osseointegration in the longer term.
the multiple steps involved in the process chain of In this study, our aim is to investigate the geometric
fabrication [10,11] . Few groups have shown the feasibility accuracy of the 3D printed dental implant using a
of using L-PBF to manufacture root-analog dental monkey maxilla incisor model. Our goal is to identify
implants for immediate implant placement . Moin the potential sources of error in geometrical accuracy
[12]
et al. showed the feasibility of obtaining models using for the development of patient-specific 3D printed dental
cone-beam computed tomography (CT) scan and the implants that mimic patient’s tooth anatomy.
subsequent fabrication of the root analog implants (RAI) 2. Methods
using L-PBF [13,14] . Chen et al. evaluated the mechanical
and biomechanical performance of customized dental 2.1. Animals and ethics approval
implants fabricated by L-PBF. They obtained the digital
files using reverse engineering by obtaining CT images The animal experimental protocol was approved by
of a maxillary incisor which are then used to create the Institutional Animal Care and Use Committee of
the customized RAI . Li et al. studied how the build SingHealth, Singapore (IACUC #2018/SHS/1419).
[15]
orientation and scanning parameters affect the surface The animal laboratory is certified by the International
roughness of RAI fabricated using L-PBF. Based on Association for the Assessment of Laboratory Animal Care.
their preliminary studies, they further analyzed the effect 2.2. Segmentation
of using constant parameters and gradient parameters,
meaning different parameters at different segments of PET-CT DICOM images of the animals’ maxillofacial
the implants. It is found that using gradient parameters, region were taken and then imported into Mimics inPrint
consistent and low surface roughness can be obtained for 2.0 software (Materialize) for segmentation of the target
the RAI . Li et al. produced RAI with oval cross-section central upper incisor from the maxilla (n = 14). First, the
[16]
abutment design using L-PBF; however, the designs are “brush threshold” tool was adjusted to 300 – 5000 HU
not obtained using imaging techniques. Nonetheless, and used to manually select the area within the boundary
they showed the potential of the material performance of the target tooth in each individual sagittal cross-section
and feasibility of using L-PBF as the fabrication of the tooth at a contrast of −500 (minimum) to 3000
method . Similarly, Guo et al. studied the performance (maximum). Next, the volume selected was refined by
[17]
of RAI fabricated using L-PBF by finite element model filling up any holes or removing any voxels that fall out of
simulations and verified them using actual fabricated the boundary of the tooth using the “brush erase” tool on
samples, without consideration for the accuracy of the each of the frontal and transverse cross-sections. The final
fabricated samples . A similar study was also conducted volume of the target tooth was then exported to ProPlan
[18]
by Song et al. . CMF 3.0 software for further segmentation. The region
[19]
The ideal customized dental implant should fit of interest spans from 1 mm beneath the cementoenamel
precisely within the socket for immediate implant junction to the tooth root (Figure 1). This region, termed
placement upon tooth extraction. The immediate implant as the dental implant, was exported as STL data and sent
placement approach is advantageous over delayed for 3D printing (n = 14).
implants as it significantly shortens overall treatment 2.3. Optimisation
duration, reduces treatment costs, and preserves the
bone and gingival tissues . Current conventional dental To improve consistency in the segmentation process, the
[20]
implants are machined into pre-determined sizes with contrast and threshold parameters were standardized.
a standard screw-form design, and the stability of the A professional digital dentistry personnel was tasked to

International Journal of Bioprinting (2022)–Volume 8, Issue 1 67