Edicio Clinica Felix Boada, Calle Bolivar, Nº 59, Baruta – Caracas – Venezuela
RID
REPORTE
Imagenológico Dentomaxilofacial
Sociedad Venezolana deSociedad Venezolana de
Radiología e ImagenologíaRadiología e Imagenología
DentomaxilofacialDentomaxilofacial
SVRID
08-11-2012
ISSN: 2791-1888. e-id: e20240302 Número 2 Volumen 3 Julio-Diciembre 2024
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
e-ISSN: 2791-1888|
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DOI: 10.60094/RID.20240302-38
1 de 8
REPORTE TÉCNICO
RESUMEN
Improved odontoma diagnosis via enamel mask in
cone beam computed tomography
Diagnóstico mejorado de odontoma complejo
por medio de máscara de esmalte en tomograa
computarizada de haz cónico
Recibido: 22/05/2024
Aceptado: 18/07/2024
Publicado: 30/07
/2024
Como citar: Pinto ASB, Ferraz MAAL, Falcão CAM, Pinto MSC, Cardoso LLC, de Sousa
ALA. Diagnóstico mejorado de odontoma complejo por medio de máscara de esmalte
en tomograa computarizada de haz cónico. Rep Imagenol Dentomaxilofacial
2024;3(2):e2024030202
Antonione Santos Bezerra Pinto
1
, Maria Ângela Arêa Leão Ferraz
2
, Carlos Alberto Monteiro Falcão
3
, Moara
e Silva Conceição Pinto
4
, Luana Leal Cosmo Cardoso
5*
, André Luca Araujo de Sousa
5
1
Assistant Professor of Pathology and Histology. Instituto de Educação Superior do Vale do Parnaíba
(FAHESP/IESVAP). Parnaíba-PI, Brasil. antonione.pinto@iesvap.edu.br ORCID: 0000-0002-6577-2816
2
Professor of Endodontics. Universidade Estadual de Piauí. Brasil. angelaferraz@phb.uespi.br ORCID:
0000-0001-5660-0222
3
Professor of Oral Radiology. Universidade Estadual de Piauí. Brasil. falcao@phb.uespi.br ORCID:
0000-0001-7787-0280
4
Post Graduation in Endodontics and Periodontics. Universidade Estadual de Piauí. Brasil. moara.c@hotmail.
com ORCID: 0000-0003-3518-3890
5
Undergraduation in Dentistry. Universidade Estadual de Piauí. Brasil. luanalccardoso@outlook.com ORCID:
0009-0000-5238-8789; a.lucaaraujo10@gmail.com ORCID: 0000-0003-4876-9188
Academic editor: Dra. Ana Isabel Ortega.
O
bjetivo: evaluar la efectividad del uso de la máscara de esmalte en el software InVesalius para
detectar regiones hiperdensas en imágenes de tomografía computarizada de haz cónico (TCHC) de
focos de esmalte en odontomas complejos. Materiales y métodos: se analizaron 25 escaneos TCHC
de pacientes que se sometieron a cirugía y diagnóstico histopatológico de odontomas complejos
utilizando el software InVesalius, donde se utilizó la segmentación para seleccionar los focos de
esmalte en la lesión. Se realizó un análisis estadístico descriptivo de los datos para evaluar la dispersión
de los mismos. Resultados: la técnica de segmentación de imágenes por umbral fue efectiva en la
identicación de áreas de esmalte en lesiones odontogénicas. El uso de la máscara de esmalte permitió
una mejor visualización de las áreas hiperdensas en las imágenes. Las pruebas estadísticas mostraron
una distribución normal para las medias de las áreas (p = 0,13, media = 3,74, DE = 0.38) y el tamaño
promedio de las lesiones (p = 0,02, media = 2,54, DE = 0,23). No hubo una relación signicativa entre
la media y el tamaño promedio (p = 0,67). Conclusión: el uso del software InVesalius asociado con la
máscara de esmalte demostró ser una herramienta efectiva en la detección de regiones hiperdensas
asociadas con odontomas complejos.
Palabras clave: Odontoma Complejo, tomografía computarizada de haz cónico, software (DeCS)
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
2 de 8e-ISSN: 2791-1888|
INTRODUCCTION
TECHNICAL REPORT
Objective: to evaluate the effectiveness of using the enamel mask in InVesalius software to
detect hyperdense regions in cone beam computed tomography (CBCT) images of enamel foci in
complex odontomas. Materials and methods: 25 CBCT scans of patients who underwent surgery and
histopathological diagnosis of complex odontomas were analyzed using the InVesalius software, where
segmentation was used to select the enamel foci in the lesion. A descriptive statistical analysis of the data
was performed to evaluate the data dispersion. Results: The threshold image segmentation technique
was effective in identifying areas of enamel in odontogenic lesions. The use of the enamel mask allowed
a better visualization of the hyperdense areas in the images. Statistical tests showed normal distribution
for the means of the areas (p = 0.13, mean = 3.74, SD = 0.38) and average size of the lesions (p = 0.02,
mean = 2.54, SD = 0.23). There was no signicant relationship between mean and average size (p =
0.67). Conclusion: the use of the InVesalius software associated with the enamel mask proved to be an
effective tool in the detection of hyperdense regions associated with complex odontomas.
Keywords: Complex Odontoma, cone-beam computed tomography, software (MeSH)
Cone beam computed tomography (CBCT)
is a diagnostic imaging tool that is increasingly
used in dentistry. It is able to provide accurate
information about the anatomy of the tooth and
adjacent tissues in three dimensions
1, 2
. With its
growing popularity from this technique, new
possibilities for the use of specic software for
analysis and processing of these images also
arise
3- 5
. InVesalius is a free and open source
image processing software that allows the
visualization and analysis of medical images in
three dimensions, including CBCT images. It
offers a variety of tools for image segmentation
and manipulation, making it possible to identify
different tissues and structures with high precision
6- 8
. In this context, the detection of hyperdense
regions on CBCT images becomes an important
challenge for dental diagnosis
9- 11
. These regions
may indicate the presence of enamel lesions,
which may be associated with a variety of dental
pathologies. However, identifying these lesions
is not always easy, especially in more complex
cases
2,13
. Thus, the aim of this study is to evaluate
the effectiveness of using the enamel mask in
InVesalius to detect hyperdense regions in CBCT
images of enamel foci in complex odontomas.
ABSTRACT
MATERIALS AND METHODS
The present study qualies as quantitative
in a cross-sectional way, based on the collection
of tomographic les for analysis using specic
software to satisfy the objective of the study.
The sample was census, all 25 patients with
tomographic images suggestive of complex
odontomas and with surgery performed for
surgical treatment and histopathological
diagnosis (Figure 1), observed from October 2022
to March 2023 were included. The observation
period was chosen based on the availability of
data and the timeframe required to collect and
analyze the CBCT images. This period allowed
for the inclusion of a sufcient number of cases
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
e-ISSN: 2791-1888|
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
3 de 8
for robust statistical analysis and coincided with
the timeframe during which patients underwent
surgeries and histopathological diagnoses of
complex odontomas. Those patients who showed
signs of neoplasms, cysts or other lesions that
could inuence the identication of structures of
interest, in addition to patients under 18 years of
age, were excluded.
Figura 1. Figure 1. Histological sections of the odontoma stained with Hematoxylin-Eosin. Acellular matrix resembling sh scales
indicating tooth enamel (red arrow) and row of columnar cells bordering this matrix in an ameloblast-like fashion (green arrow),
followed by attened cells that abruptly move apart maintaining intercellular bridges, loosely arranged, indicating the reticulum
stellate of an enamel organ as in odontogenesis (star). Area showing basophilic mineralized material that forms spherical structures
called cementicles, characterizing a formation of disorganized cementoid tissue (yellow arrows).
All procedures performed in studies involving
human participants were in accordance
with the ethical standards of the institutional
research committee and with the 1964 Helsinki
declaration and its comparable ethical standards
(Ethics Committee Submission Code: 6.065.998).
Informed consent was obtained from all individual
participants included in the study for publication
of their data.
CBCT scans were used using the Carestream
CS 8100 scanner (Carestream Dental LLC, Atlanta,
Brazil), with the following settings: 90 kVp, 5
mA and exposure time of 8 seconds. Images
were selected in DICOM (Digital Imaging and
Communications in Medicine) format.
The DICOM images were opened in the
InVesalius software version 3.1.0, a free software
for visualization, processing and analysis of 3D
medical images, widely used in Dentistry, where
segmentation was used to select the enamel foci
in the lesion. For this, the enamel mask available
in the software was used, and the threshold values
were adjusted to segment only pixels with intensity
within the desired range. The segmented areas
were measured in square millimeters (mm²) and
the spatial coordinates were recorded in relation to
the anatomical reference of the lesion. Data were
stored in electronic spreadsheets for statistical
analysis.
Standardization of Hounseld Units (HU)
The standardization of Hounseld Units
(HU) in CBCT scans is a fundamental element
to ensure the reliability and consistency of
measurements obtained in the images. Although
HU values are typically used in conventional
computed tomography and not adapted for
CBCT due to differences in image acquisition and
reconstruction processes, a rigorous calibration
approach was adopted in this study to ensure HU
standardization. A specially designed calibration
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
4 de 8e-ISSN: 2791-1888|
phantom containing reference materials with
known densities was used, covering a wide range
of HU values including soft tissues, bones, and
dental enamel.
Calibration Phantom
Specially designed calibration phantom was used for
this study. The phantom contained reference materials
with known densities that covered a wide range of HU
values, including densities representing soft tissues,
bones, and dental enamel. Each material within the
phantom was characterized by its specic HU value.
Calibration Procedure
The meticulous calibration procedure
involved several specic steps. Before each
image acquisition, the calibration phantom was
positioned in the same location as the patients.
Images of the phantom were acquired under the
same exposure settings used for the patients.
Subsequently, the CS 8100 tomograph underwent
a specic calibration procedure, where images
of the phantom were acquired under the same
exposure settings used for patients.
Comparison and Adjustment
After acquiring the phantom images, the HU
values measured in the regions of interest (ROI)
corresponding to the reference materials within
the phantom were compared with the known HU
values. Any discrepancies were adjusted through
a calibration correction procedure. This involved
modifying the CBCT scan acquisition parameters,
if necessary, to ensure that the HU values
measured in the images accurately matched the
reference values.
Continuous Verication
Before each new image acquisition session, the
calibration procedure was repeated to ensure that
the HU values remained consistent throughout
the study. The standardization process ensures the
reliability and consistency of HU measurements by
using a calibration phantom with known densities
and repeating the calibration procedure before
each new image acquisition session. Challenges
include variations in acquisition parameters and
the inuence of surrounding tissues, which were
minimized through careful adjustments and
continuous verication
This rigorous approach to HU standardization
ensured that density values in the obtained
images were reliable and independent of any
variations in surrounding tissues or exposure
parameters. In this way, we could guarantee the
accuracy and validity of the measurements made
during the study.
Statistical analysis
All data were organized and distributed in the
Excel program (2019), and comparative analyses
were performed using the SPSS Statistics
software version 25.0 (IBM, Armonk, USA) and
R (Statistical and graphical programming
language, Lucent Technologies, Murray Hill,
USA). Thus, to fulll the research objectives,
Kolmogorov-Smirnov tests were conducted to
assess the normality of sample distribution, and
for correlation between groups, the Pearson
Product-Moment Test was employed, with a
signicance level set at 5% (p < 0.05).
RESULTS
Through the analysis of the DICOM images
obtained in the InVesalius software, it was possible
to identify hyperdense areas of enamel in the
lesions (Figure 2). The Kolmogorov-Smirnov tests
to assess the normality of sample distribution
demonstrated that the means of the areas
(p-value = 0.13) have a normal distribution (mean of
3.74 and SD + 0.38), and the average size (p-value =
0.02) has a normal distribution (mean of 2.54 and
SD + 0.23). (Figure 3) (Figure 4).
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
e-ISSN: 2791-1888|
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
5 de 8
Figure 2. A, B, C: Sagittal and coronal tomographic slices and maximum intensity projection. In the sagittal and coronal
tomographic sections, the areas colored green correspond to the applied enamel mask. D: Detected hyperdense foci. The
arrow indicates areas of dental enamel in a complex odontoma.
Furthermore, the results of the Pearson
Product-Moment Correlation to analyze
whether there is a direct or indirect relationship
between the mean and the average size
demonstrated a statistically non-signicant
correlation (p-value = 0.67), indicating that
the groups are independent and do not vary
in a correlated manner. Therefore, it was
possible to observe that the threshold image
segmentation technique was effective in
identifying areas of enamel in odontogenic
lesions.
Figure 3. Kolmogorov-Smirnov test for enamel areas. The means of the areas (p-value = 0.13) have a normal distribution (Mean of
3.74 and SD+0.38).
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
6 de 8e-ISSN: 2791-1888|
DISCUSSION
Figure 4. Kolmogorov-Smirnov test for mean tooth size. Average size (p-value = 0.02) has a normal distribution (Mean of 2.54 and
SD + 0.23).
The use of advanced technologies in dentistry,
such as CBCT, has been of great importance
in the diagnosis and treatment of various
injuries. However, the interpretation of the
images obtained can be challenging, especially
when dealing with lesions that have similar
characteristics1,8. In this context, the InVesalius
software, associated with the enamel mask, has
proven to be an effective tool for detecting areas of
enamel in odontogenic lesions, such as complex
odontoma.The use of an enamel mask allowed
a better visualization of the hyperdense areas in
the CBCT images, facilitating the diagnosis and
helping to identify lesions that present similar
tomographic characteristics.
Several studies have highlighted the
importance of CBCT in dental diagnostics
and the challenges associated with image
interpretation in complex cases. The use of
segmentation software like InVesalius, coupled
with enamel masks, provides a signicant
advantage in visualizing hyperdense regions. For
instance, Lo Giudice et al.
7
demonstrated the
accuracy of imaging software for 3D analysis of
mandibular condyles, supporting our ndings
on the utility of InVesalius. Additionally, Kim
9
explored the potential of CBCT in assessing
bone mineral density, which aligns with our
efforts to standardize HU values for reliable
measurements. These studies underscore the
potential of InVesalius in routine evaluations for
tomographic diagnosis, offering a reliable and
cost-effective tool compared to other advanced
imaging techniques.
The use of the InVesalius software for the
detection of enamel areas in odontogenic lesions
is possible thanks to the segmentation resource
available in the program. This resource allows
the selection of certain types of tissues present
in the image, allowing the visualization of areas
of enamel among other structures. For this, it is
necessary to dene a threshold that allows the
selection of pixels with color intensity similar to
that of enamel.
With the use of the enamel mask, it is also possible
to isolate the selected areas in the original image,
which facilitates their visualization and interpretation.
This tool is especially useful in the differential diagnosis
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
e-ISSN: 2791-1888|
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
7 de 8
LESIONS RADIOLOGICAL CHARACTERISTICS
Cementoblastoma
Accumulation of cementoblasts in the periapical region of teeth;
It presents as a well-delineated round radiopaque mass of disorganized cementum, surrounded by a
radiolucent halo of germ cells.
Periapical cemento-osseous
dysplasia
In a more advanced stage, it presents as a lesion with a dense nucleus and a radiopaque mineralized
mass surrounded by a radiolucent zone.
Cemento-Ossifying Fibroma
Mass of brous and mineralized tissue of slow growth;
It is presented as a radiopaque-radiolucent mixture. More advanced lesions become radiopaque due
to progressive mineralization;
Condensing Osteitis Focal area of well-circumscribed or ill-dened bony sclerosis around tooth roots
Idiopathic Osteosclerosis Radiopaque area with sharp angular margins. Resembles a dense bone island
Dentinogenic Ghost Cell
Tumor
Density mixed with the degree of calcication. They are typically well dened and may result in resorp-
tions.
Adenomatoid Odontogenic
Tumor
Characterization of mixed density with calcications with the appearance of snow locos, usually with
the presence of an impacted tooth
Calcifying Epithelial Odonto-
genic Tumor
It has varying densities from radiolucent to dense radiopaque. Its characteristic is that intralesional
calcication is around a crown and is often associated with impaction of one or more teeth.
Table 1. Neoplasms of the orofacial complex that make dierential diagnosis with odontoma.
of complex odontoma, which can be confused with
other lesions that present hyperdense calcications
(Table 1)
14,16
. With the detection of areas of enamel
in the lesions, it is possible to conrm the presence
of odontogenic tissue and rule out other possible
diagnostic hypotheses.
The results obtained in this study showed
an average of 3.8 enamel areas per complex
odontoma lesion, with an average size of 2.5 mm².
These values are consistent with data found in
the literature, which suggests that the use of the
enamel mask on the InVesalius can be a reliable
tool for detecting areas of enamel in odontogenic
lesions
16
.
Source: Ghita et al.
14
; Vanhoenacker et al.
15
It is important to emphasize that the detection
of areas of enamel in odontogenic lesions through
the use of enamel mask should not be considered
as a denitive diagnosis. It is necessary that the
interpretation of the images be carried out by a
qualied and experienced professional, who can
evaluate all the clinical and tomographic aspects
of the patient.
CONCLUSION
In conclusion, the use of the InVesalius software
associated with the enamel mask proved to be
an effective tool for detecting areas of enamel
in odontogenic lesions, enabling the differential
diagnosis of complex odontoma and other lesions
that present hyperdense calcications. However,
the interpretation of the images must be carried
out by a qualied professional, who considers all
clinical and tomographic aspects of the patient.
Conicts of Interest: The authors declare that
they have no conicts of interest.
Corresp
onding author:
Luana Leal. Afonso Pena Street, 1528, Pindorama.
Faculty of Dentistry, Universidade Estadual
de Piauí. Parnaíba, Brazil.
luanalccardoso@ outlook.com
Rep Imagenol Dentomaxilofacial 2024 julio-diciembre; 3(2): 2024030202
Disponible en: hps://publicaciones.svrid.org.ve/index.php/rid
DIAGNÓSTICO MEJORADO DE ODONTOMA COMPLEJO
POR MEDIO DE MÁSCARA DE ESMALTE
Bezerra, et al.
8 de 8e-ISSN: 2791-1888|
Prog Orthod. 2022 Jun 13;23(1):22. DOI: https://doi.
org/10.1186/s40510-022-00413-8
9. Kim DG. Can dental cone beam computed
tomography assess bone mineral density? J
Bone Metab. 2014 May;21(2):117-26. DOI: https://doi.
org/10.11005/jbm.2014.21.2.117
10. Adibi S, Zhang W, Servos T, O’Neill PN. Cone
beam computed tomography in dentistry: what
dental educators and learners should know. J Dent
Educ. 2012 Nov;76(11):1437-42.
11. Aldhuwayhi S, Bhardwaj A, Deeban YAM,
Bhardwaj SS, Alammari RB, Alzunaydi A. A narrative
review on current diagnostic imaging tools for
dentomaxillofacial abnormalities in children.
Children (Basel). 2022 Apr 27;9(5):621. DOI: https://
doi.org/10.3390/children9050621
12. Silva BSF, Bueno MR, Yamamoto-Silva
FP, Gomez RS, Peters OA, Estrela C. Differential
diagnosis and clinical management of periapical
radiopaque/hyperdense jaw lesions. Braz Oral Res.
2017 Jul 3;31:e52. DOI: https://doi.org/10.1590/1807-
3107BOR-2017.vol31.0052
13. Araki M, Matsumoto N, Matsumoto K,
Ohnishi M, Honda K, Komiyama K. Asymptomatic
radiopaque lesions of the jaws: a radiographic
study using cone-beam computed tomography.
J Oral Sci. 2011 Dec;53(4):439-44. DOI: https://doi.
org/10.2334/josnusd.53.439
14. Ghita I, Brooks JK, Bordener SL, Emmerling
MR, Price JB, Younis RH. Central compact osteoma
of the mandible: case report featuring unusual
radiographic and computed tomographic
presentations and brief literature review. J Stomatol
Oral Maxillofac Surg. 2021 Nov;122(5):516-20. DOI:
https://doi.org/10.1016/j.jormas.2020.09.014
15. Vanhoenacker FM, Bosmans F, Vanhoenacker
C, Bernaerts A. Imaging of mixed and radiopaque
jaw lesions. Semin Musculoskelet Radiol. 2020
Oct;24(5):558-569. DOI: https://doi.org/10.1055/s-
0039-3402766
16. MacDonald-Jankowski DS. Odontomas in
a Chinese population. Dentomaxillofac Radiol.
1996 Sep;25(4):186-92. DOI: https://doi.org/10.1259/
dmfr.25.4.9084271. PMID: 9084271
1. Nasseh I, Al-Rawi W. Cone beam
computed tomography. Dent Clin North Am.
2018 Jul;62(3):361-91. DOI: https://doi.org/10.1016/j.
cden.2018.03.002
2. Leonardi Dutra K, Haas L, Porporatti AL,
Flores-Mir C, Nascimento Santos J, Mezzomo LA,
Corrêa M, De Luca Canto G. Diagnostic accuracy
of cone-beam computed tomography and
conventional radiography on apical periodontitis:
a systematic review and meta-analysis. J Endod.
2016 Mar;42(3):356-64. DOI: https://doi.org/10.1016/j.
joen.2015.12.015
3. Ebert LC, Franckenberg S, Sieberth T,
Schweitzer W, Thali M, Ford J, Decker S. A review of
visualization techniques of post-mortem computed
tomography data for forensic death investigations.
Int J Legal Med. 2021 Sep;135(5):1855-67. DOI: https://
doi.org/10.1007/s00414-021-02581-4
4. Wildberger JE, Prokop M. Hounseld’s
Legacy. Invest Radiol. 2020 Sep;55(9):556-8. DOI:
https://doi.org/10.1097/RLI.0000000000000680
5. Tsoumpas C, Sauer Jørgensen J, Kolbitsch
C, Thielemans K. Synergistic tomographic image
reconstruction: part 2. Philos Trans A Math Phys
Eng Sci. 2021 Aug 23;379(2204):20210111. DOI: https://
doi.org/10.1098/rsta.2021.0111
6. Centro de Tecnologia da Informação
Renato Archer [Internet]. Software Inversalius:
Guia do Usuário [Citado 18 de mayo de 2024].
Recuperado a partir de: https://www.gov.br/cti/
pt-br/acesso-a-informacao/acoes-e-programas/
invesalius
7. Lo Giudice A, Quinzi V, Ronsivalle V, Farronato
M, Nicotra C, Indelicato F, Isola G. Evaluation of
imaging software accuracy for 3-Dimensional
analysis of the mandibular condyle. A comparative
study using a surface-to-surface matching
technique. Int J Environ Res Public Health. 2020
Jul 3;17(13):4789. DOI: https://doi.org/10.3390/
ijerph17134789
8. Lo Giudice A, Ronsivalle V, Gastaldi G, Leonardi
R. Assessment of the accuracy of imaging software
for 3D rendering of the upper airway, usable in
orthodontic and craniofacial clinical settings.
REFERENCES