Accuracy and clinical validity of automated cephalometric analysis using convolutional neural networks.

BACKGROUND: This study aimed to assess the error range of cephalometric measurements based on the landmarks detected using cascaded CNNs and determine how horizontal and vertical positional errors of individual landmarks affect lateral cephalometric measurements. METHODS: In total, 120 lateral cephalograms were obtained consecutively from patients (mean age, 32.5 ± 11.6) who visited the Asan Medical Center, Seoul, Korea, for orthodontic treatment between 2019 and 2021. An automated lateral cephalometric analysis model previously developed from a nationwide multi-centre database was used to digitize the lateral cephalograms. The horizontal and vertical landmark position error attributable to the AI model was defined as the distance between the landmark identified by the human and that identified by the AI model on the x- and y-axes. The differences between the cephalometric measurements based on the landmarks identified by the AI model vs those identified by the human examiner were assessed. The association between the lateral cephalometric measurements and the positioning errors in the landmarks comprising the cephalometric measurement was assessed. RESULTS: The mean difference in the angular and linear measurements based on AI vs human landmark localization was .99 ± 1.05°, and .80 ± .82 mm, respectively. Significant differences between the measurements derived from AI-based and human localization were observed for all cephalometric variables except SNA, pog-Nperp, facial angle, SN-GoGn, FMA, Bjork sum, U1-SN, U1-FH, IMPA, L1-NB (angular) and interincisal angle. CONCLUSIONS: The errors in landmark positions, especially those that define reference planes, may significantly affect cephalometric measurements. The possibility of errors generated by automated lateral cephalometric analysis systems should be considered when using such systems for orthodontic diagnoses.

Growth Patterns of the Maxillomandibular Complex in Preadolescent Patients With Pierre-Robin Sequence Using Cluster Analysis and Longitudinal Follow-Up Cephalometric Data.

OBJECTIVE: To investigate the growth patterns of the maxillomandibular complex in preadolescent patients with Pierre-Robin sequence (PRS). METHODS: The samples consisted of 20 preadolescent PRS patients who had cleft palate and did not undergo growth-modification therapy or surgical intervention [6 boys and 14 girls; mean age of lateral cephalograms taken, 8.8 y (T1) and 13.7 y (T2)]. According to k-means cluster analysis, four clusters were defined over 3 major groups at T1: (1) Both very retrusive maxilla and mandible group: Cluster-4 [n=4, very large gonial angle, very low facial height ratio (FHR)] and Cluster-1 (n=5, small gonial angle, normal FHR); (2) Retrusive maxilla and very retrusive mandible group (Cluster-3, n=7, normal gonial angle, very low FHR); and (3) Both normal maxilla and mandible group (Cluster-2, n=4, very large gonial angle, low FHR). Seven angular and ratio variables [sella-nasion-A point (SNA), sella-nasion-B point (SNB), A point-nasion-B point (ANB), saddle angle, gonial angle, mandibular-body-length/anterior-cranial-base-length (MBL/ACBL), and FHR] at T1 and T2 and growth change from T1 to T2 were investigated. RESULTS: At T1, SNA, SNB, saddle angle, gonial angle (all P<0.05), and FHR (P<0.01) showed significant difference among 4 clusters. At T2, SNA, SNB and gonial angle (all P<0.05) still showed significant difference among 4 clusters. During T1 to T2, there was no significant change in variables at each cluster except an increase in MBL/ACBL in Cluster-1 and FHR in Cluster-3 (all P<0.05). CONCLUSIONS: Preadolescent PRS patients exhibited diverse skeletal phenotypes at T1, which did not change significantly from their original skeletal patterns by growth during T1 to T2.

Catch-Up Growth Pattern in Cleft Palate: A Longitudinal Study from Infancy to Adolescence.

OBJECTIVE: To characterize the patterns of somatic catch-up growth from infancy to adolescence in patients with cleft palate (CP). STUDY DESIGN: We assessed 474 nonsyndromic patients with isolated cleft palate (n = 69) and unilateral and bilateral cleft lip and palate (n = 271; n = 134) who underwent palatoplasty between 1988 and 2017 and had longitudinal physical growth data at birth (T0), cheiloplasty (T1), palatoplasty (T2), childhood (T3), and adolescence (T4). The z scores of weight (ZWT), height (ZHT), and body mass index (ZBMI) were compared among the CP types (isolated cleft palate, unilateral cleft lip and palate, and bilateral cleft lip and palate) and time points (T1, T2, T3, and T4). Subgroup analyses were performed to investigate the growth of patients with malnourishment (z score < -1) at T1 or T2. A generalized linear model was used to investigate the effects of gestational age and cardiac anomalies on the longitudinal changes in ZHT and ZBMI. RESULTS: Regardless of the time point, the overall ZHT, ZWT, and ZBMI approximated 0 in all CP types, indicating few differences from the mean values of noncleft children. Significant catch-up growth occurred in ZHT and ZWT from T1 to T4 for all CP types (all P < .05). Despite the recovery of ZHT and ZBMI in most patients with malnourishment, these values remain relatively low until adolescence. Patients who were born at preterm stage or had surgically repaired cardiac anomalies grew well. CONCLUSIONS: Even in infants with CP and malnutrition, preterm birth, or cardiac anomalies, rapid catch-up growth can occur prior to palatoplasty with the help of comprehensive cleft care.

Stability of clockwise rotation of the maxillary occlusal plane in skeletal Class III patients treated with two-jaw surgery.

OBJECTIVE: The objective of the study was to investigate the post-surgical stability of clockwise rotation of the maxillary occlusal plane (MXOP) after Le Fort I osteotomy for posterior impaction and advancement, in skeletal Class III patients who had undergone two-jaw orthognathic surgery (2J-OGS). MATERIALS AND METHODS: The sample set consisted of 46 patients (18 males and 28 females). Using lateral cephalograms taken at the initial evaluation (T0), before 2J-OGS (T1), after 2J-OGS (T2) and at debonding (T3), the amount of MXOP change (criterion: 2° in ΔMXOP [between T2 and T3]) was used to assign patients to Group 1 (G1, high relapse, -3.09°) or Group 2 (G2, low relapse, -0.99°). Findings were statistically compared between the groups. RESULTS: Compared with G2, G1 exhibited more severe skeletal Class III relationships and a flatter MXOP at T0, a greater increase in ΔANB, and more clockwise rotation of ΔMXOP and ΔFMA between T1 and T2. G1 also showed a greater decrease in ΔANB, higher counterclockwise rotation of ΔMXOP and upward movement of U1 between T2 and T3. Regression analysis yielded the following equation: MXOP (ΔT2-T3) = -0.37 X MXOP (ΔT1-T2) -0.43. CONCLUSIONS: The higher the clockwise rotation of the MXOP during 2J-OGS in skeletal Class III patients, the greater the subsequent relapse (counterclockwise rotation) of the MXOP.

Treatment Strategy for Condyle/Ramus Deficiency Using 3D Virtual Surgery and CAD/CAM Technologies; Total Joint Replacement or Inverted-L Osteotomy.

OBJECTIVE: To introduce a guideline for selecting proper surgical modalities for correction of skeletal Class II malocclusion with moderate-to-severe vertical height deficiency (VHD) of the condyle/ramus complex (CRC) using bimaxillary surgery in conjunction with total joint replacement (TJR) or inverted-L osteotomy (ILO) assisted by 3D virtual surgical planning and CAD/CAM technologies. CASES: Four cases of severe skeletal Class II patients with moderate-to-severe VHD of CRC were presented. They underwent bimaxillary surgery using Le Fort I osteotomy in the maxilla and TJR or ILO in the mandible, with the help of virtual surgical planning (VSP) and computer-aided design/computer-aided manufacturing (CAD/CAM)-printed surgical guides and osteosynthesis plates. Guidelines are as follows: (1) if a patient has a moderate degree of VHD and the function of the temporomandibular joint (TMJ) is normal, ILO would be preferred for vertical elongation of the ramus; (2) if a patient has congenitally small condyle or severe condylar resorption, TJR would be preferred to resolve severe VHD of CRC; and (3) when a patient has a metal allergy, foreign body reaction or anatomic limitations (ie, thin cranial base cortex for fossa fixation), ILO would be a better option than TJR. RESULTS: TJR or ILO using VSP and CAD/CAM-printed surgical guides, wafers, and customized plates can provide a proper selection of the surgical plan, accurate transfer of surgical plans to actual surgical procedures, and esthetic improvement of the facial profile. CONCLUSION: This guideline based on the degree of VHD and functional aspects of CRC might help clinicians to select effective surgical modality for correction of skeletal Class II malocclusion with moderate-to-severe VHD of CRC.

Classifications of Prominent Mandibular Angle and Surgical Considerations for Mandibuloplasty.

Contouring surgery is an operation that makes a smooth, oval-shaped face by removing protruding parts of the face. In particular, mandibuloplasty is a procedure that softens the shape of the mandible and is a popular procedure among Asians with a dominant mesocephalic profile. In this study, we tried to discuss the considerations before surgery by dividing the mandibular shape into morphologic aspects in the lateral and frontal views and anatomic considerations. In the lateral view, it is essential to consider the posterior ramal height and mandibular body length of each patient. These factors are closely related to the postoperative outcomes. Consideration of the mandibular plane angle is also essential. In the frontal view, the position of the gonial angle relative to the ramal body should be considered because the position of the gonial angle can affect the difficulty of operation. Anatomic considerations include antero-posterior ramal width, bucco-lingual width of mandibular body, amount of mouth opening, and location of inferior alveolar nerve. These are important factors that determine the difficulty of the operation. In this study, classification according to the shape and shape of the mandible was presented, and considerations for each classification were discussed. Cosmetic surgery is a socio-psychologically meaningful procedure that not only improves the esthetics but also improves the mental well-being of the patients. Therefore, it will be possible to provide better medical services to patients if clinicians are fully aware of the factors that can increase the difficulty of the operation and the factors that affect the surgical outcome.

Classification of Skeletal Phenotypes of Preadolescent Patients With Isolated Cleft Palate Using Principal Component Analysis and Cluster Analysis.

This study aimed to classify the skeletal phenotypes of preadolescent patients with isolated cleft palate using principal component analysis and cluster analysis. Sixty-four preadolescent female patients with isolated cleft palate (incomplete hard palate and complete soft palate cleft group, n=51; complete cleft of the hard and soft palate group, n=13; the mean age when lateral cephalograms were taken, 7.08±0.76 y) were included. Ten angular and 2 ratio cephalometric variables were measured on a lateral cephalogram. Cluster analysis was performed using 3 representative variables obtained from principal component analysis (SN-GoMe, SNA, and SNB). The differences in the variables among the clusters were characterized using the Kruskal-Wallis test. As a result of the analysis, 6 clusters were obtained from 3 groups: the retrusive maxilla and mandible group: cluster 3 (14.1%, moderately hyperdivergent pattern), cluster 5 (17.2%, severely hyperdivergent pattern); the normal maxilla and mandible group: cluster 1 (23.4%, normodivergent pattern), cluster 4 (12.5%, moderately hyperdivergent pattern), cluster 6 (20.3%, severely hyperdivergent pattern); the normal maxilla and protrusive mandible group: cluster 2 (12.5%, normodivergent pattern). The distribution of isolated cleft palate types did not differ among the 6 clusters ( P >0.05). Two thirds of the patients (68.7%, clusters 1, 2, 4, and 6) had a normal anteroposterior position of the maxilla, while one third of the patients (31.3%, clusters 3 and 5) showed a retrusive mandible. These results indicate that isolated cleft palate patients have diverse maxillo-mandibular growth patterns compared with patients with cleft lip and palate.

Distribution and Phenotype of Goldenhar Syndrome and Its Association With Other Anomalies.

The purpose of this study was to investigate the distribution and phenotype of Goldenhar syndrome (GS) and its association with other anomalies. The samples consisted of 18 GS patients (6 males and 12 females; mean age at investigation, 7.4 ± 4.8 y) who were treated or followed up at the Department of Orthodontics, Seoul National University Dental Hospital between 1999 and 2021. The prevalence of side involvement and degree of mandibular deformity (MD), midface anomalies, and association with other anomalies were evaluated using statistical analysis. The prevalence of unilateral and bilateral MD did not differ (55.6% versus 44.4%). In unilateral MD cases, there was a tendency for higher prevalence of more severe Pruzansky-Kaban types than mild ones (type I, 10%; type IIa, 10%; type IIb, 50%; type III, 30%). Despite hypoplasia of the condyle/ramus complex, compensatory mandibular body growth occurred in 33.3% of GS patients (more severe side in bilateral MD cases, 37.5%, and ipsilateral side in unilateral MD cases, 30%). Class II molar relation was more prevalent than class I and class III molar relations (72.2% versus 11.1% versus 16.7%, P <0.01). Al total of 38.9% of patients had congenitally missing tooth. #7 facial cleft was found in 44.4% of patients. In midface anomalies, ear problem was the most common anomaly, followed by hypoplasia/absence of zygomatic arch and eye problem (88.9% versus 64.3% versus 61.1%, P <0.01). Association with the midface, spine, cardiovascular, and limb anomalies did not differ between unilateral and bilateral MD cases. These results might provide a basic guideline for diagnosis and treatment planning for GS patients.

The Efficacy of 3D Virtual Surgery, CAD/CAM, and 3D Printing Technology for Maxillomandibular Advancement in Obstructive Sleep Apnea Patients.

OBJECTIVE: Maxillomandibular advancement (MMA) is the most effective surgical method for treating obstructive sleep apnea, and it moves the maxillomandibular complex forward to increase the entire upper airway volume. By using 3-dimensional (D) virtual surgery, computer-aided design/computer-aided manufacturing, and 3D printing technologies, it is possible to overcome all the limitations of conventional methods. MATERIALS AND METHODS: In this study, (modified) MMA was performed by applying 3D technologies to obstructive sleep apnea patients. Virtual surgery was done as surgical plan, cutting guides, and customized plates were made by computer-aided design/computer-aided manufacturing and 3D printing technologies for surgical procedures. RESULTS: After surgery, all patients improved their appearance, quality of sleep, and sleep apnea level were dramatically improved. Through these results, it was found that there are many advantages in using 3D technologies for preparing and implementing MMA. CONCLUSIONS: It was confirmed that the accuracy and efficiency of surgery were increased by applying 3D technologies. This suggests that 3D technologies are very useful tools in surgical area.

Causes of Reoperation Consultation and Clinical Review of Actual Reoperation After Previous Orthognathic Surgery.

OBJECTIVE: To investigate the causes of reoperation consultation, and the actual percentage and procedures of reoperation after previous orthognathic surgery. METHODS: The samples consisted of 30 patients who visited our clinic for reoperation consultation from October 2015 to September 2021 (6 males and 24 females; mean age at reoperation consultation, 28.4 y). Patient's causes of reoperation consultation were divided into "esthetic dissatisfaction," "airway changes," "temporomandibular disorders," "uncomfortable occlusion," and "other complications". In terms of esthetic dissatisfaction, the more detailed esthetic problem was evaluated by the clinical chart, facial photographs, and radiographs. In patients who actually underwent reoperation, the actual percentage and procedures of reoperation were investigated. RESULTS: The most prevalent causes for reoperation consultation were "esthetic dissatisfaction" (n = 21, 70.0%), followed by "airway changes" (n = 11, 36.7%), "uncomfortable occlusion" (n = 8, 26.7%), "other complications" (n = 5, 16.7%), and "temporomandibular disorder" (n = 4, 13.3%). Less than half of patients actually underwent reoperation (n = 13, 43.3%). Actual reoperation procedures included minor revision surgery, reconstruction surgery, or complete reoperation according to the patient's need. In case of complete reoperation, more accurate and predictable results were obtained by using virtual surgical planning, customized surgical guides, titanium surgical plates made with computer-aided design and computer-aided manufacturing technique, and a 3-dimensional printing method. CONCLUSION: It is important to communicate with patients about expectations for facial esthetic improvement by orthognathic surgery for obtaining the patient's postoperative satisfaction.

Classification of Skeletal Phenotypes of Adult Patients With Cleft Skeletal Class III Malocclusion Using Principal Component Analysis and Cluster Analysis.

The purpose of this study was to classify the skeletal phenotypes of adult patients with skeletal class III (C-III) malocclusion and unilateral or bilateral cleft lip and palate using principal component analysis and cluster analysis. The samples consisted of 81 adult C-III patients with cleft lip and palate (CLP) who underwent orthognathic surgery (OGS) or distraction osteogenesis (59 males and 22 females; 50 unilateral cleft lip and palate and 31 bilateral cleft lip and palate; mean age when lateral cephalograms were taken, 22.2±4.6 y). Thirteen angular and one ratio cephalometric variables were measured. Using 4 representative variables obtained from principal component analysis (SNA, SNB, Gonial angle, and Bjork sum), K-means cluster analysis was performed to classify the phenotypes. Then, statistical analysis was conducted to characterize the differences in the variables among the clusters. Five clusters were obtained from 3 groups: severely retrusive maxilla and moderately retrusive mandible group: cluster-1 (23.5%, severely hyperdivergent pattern), cluster-4 (27.2%, moderately hyperdivergent pattern), and cluster-5 (11.1%, normodivergent pattern); moderately retrusive maxilla and normal mandible group: cluster-2 (30.9%, normodivergent pattern); normal maxilla and moderately protrusive mandible group: cluster-3 (7.4%, normodivergent pattern). Although skeletal phenotypes were diverse, distribution of sex and cleft type did not differ among 5 clusters ( P >0.05). Sixty-two percent of cleft patients showed a severely retrusive maxilla and moderately retrusive mandible (cluster-1, cluster-4, and cluster-5), which indicated that these are the main cause of skeletal C-III malocclusion in CLP patients who were treated with OGS. Therefore, it is necessary to consider presurgical orthodontic treatment and surgical planning based on the skeletal phenotypes of CLP patients.

The Accuracy and Stability of Intentional Change of Frontal-Ramal Inclination in Orthognathic Surgery for Facial Asymmetry Patients.

Facial asymmetry can be defined as differences in the left and right sides of the face, and most of the patients with facial asymmetry have different left and right frontal-ramal inclinations (FRIs). Restoring the symmetry of both FRIs is important in the surgery of facial asymmetry patients, but it is very difficult to achieve perfect symmetry through conventional orthognathic surgery. However, by using 3-dimensional (3D) virtual planning and CAD/CAM technolo gies, intentional change of FRIs can be possible so the symmetry can be improved. The purpose of this study is to evaluate the surgical accuracy and long-term stability of intentional change of FRIs by 3D virtual surgery, CAD/CAM-assisted orthognathic surgery for patients with facial asymmetry. The study included 20 patients who had undergone orthognathic surgery for skeletal class III malocclusion from January 2019 to December 2021. To evaluate the accuracy of surgery, 3D facial cone beam computed tomography (CBCT) taken immediately after surgery (T1) and virtual surgery data (Tv) were measured and the difference values were calculated. The evaluation of the long-term stability of intentional change of FRI was performed by measuring T1 and T2 (3D facial cone beam computed tomography images taken 6 mo after surgery) and the difference values were calculated. The difference values of FRIs in the left and right proximal segments of each patient were calculated. And then, for comparison depending on the direction of rotation, increased FRI groups (n=20, medial rotation) and decreased FRI groups (n=20, lateral rotation) were analyzed separately. As a result, all difference values at both (ΔT1-Tv) and (ΔT2-T1) were <1 degree. As a result of analyzing the entire FRI by dividing it into decreasing and increasing groups, the mean value of (ΔT1-Tv) was 0.225 degrees for the decreasing group and 0.275 degrees for the increasing group. It means that the proximal segment moved less than the movement implemented by the virtual surgery through actual surgery but it shows a very small error, which means that the overall operation almost accurately implements the virtual surgical planning. Compared with (ΔT1-Tv), the mean value of (ΔT2-T1) showed a much smaller error value, and no specific tendency was observed. This indicates that the stability after surgery is very good. Based on this study, using 3D virtual surgery planning and CAD/CAM technologies for treating patients with facial asymmetry was very useful, and surgery could be performed accurately and predictably. In particular, left-right symmetry was almost perfectly achieved through virtual simulation and could be implemented through actual surgery. Therefore, it can be said that the use of these 3D technologies is recommended for the surgical approach of facial asymmetry.

Facial Asymmetry Phenotypes in Adult Patients With Unilateral Cleft Lip and Palate and Skeletal Class III Malocclusion Using Principal Component Analysis and Cluster Analysis.

The purpose of this study was to classify and characterize facial asymmetry (FA) phenotypes in adult patients with unilateral cleft lip and palate (UCLP) and skeletal class III malocclusion. The samples comprised 52 adult UCLP patients (36 men and 16 women; mean age, 22.43 y) who had undergone orthognathic surgery for correction of class III malocclusion. After measurement of 22 cephalometric parameters in posteroanterior cephalograms taken 1 month before orthognathic surgery, principal component analysis was performed to obtain 5 representative parameters [deviation (mm) of ANS (ANS-dev), maxillary central incisor contact point (Mx1-dev), and menton (Me-dev); cant (degree) of the maxillary anterior occlusal plane (MxAntOP-cant) and mandibular border (MnBorder-cant)]. K-means cluster analysis was conducted using these representative parameters. The differences in cephalometric parameters among the clusters were statistically analyzed. The FA phenotypes were classified into 4 types: No-cant-and-No-deviation type (cluster-4, n=16, 30.8%); MxMn-cant-MxMn-dev to the cleft-side type (cluster-3, n=4, 7.7%); Mx-cant-Mn-shift to the cleft-side type (cluster-2, n=15, 28.8%); and Mn-cant-Mn-dev to the noncleft-side type (cluster-1, n=17, 32.7%). Asymmetry in the maxilla and/or mandible were observed in 70% of patients. One third of patients (cluster-2 and cluster-3; sum, 36.5%) exhibited significant cant of MxAntOP induced by cleft and cant or shift of the mandible to the cleft side. Another one third of patients (cluster-1, 32.7%) demonstrated significant deviation and cant of the mandible to the noncleft-side despite cleft in the maxilla. This FA phenotype classification might be a basic guideline for diagnosis and treatment planning for UCLP patients.

Maxillary Total Elongation Surgery using 3D Virtual Surgery, CAD/CAM and 3D Printing Technology: Surgical Convenience and Accuracy.

After Le Fort I osteotomy was first performed by von Langernbeck in 1859, there has been many improvements since. Research on and development of Le Fort I osteotomy procedure has shown downward movement of the maxilla to exhibit lowest stability and accuracy. However, maxillary downgraft movement is necessary in orthognathic patients with insufficient vertical length of the maxilla, but fixation of the maxilla after elongation is often very inaccurate. In this study, the authors utilized 3D virtual surgery, CAD/CAM-assisted 3D printing technology to overcome such limitations of maxillary total elongation. In addition, accuracy at 7 different landmarks from superimposition of virtual simulation data and postoperative Cone-beam computed tomography (CBCT) data were measured. Although posterior maxilla exhibited bigger range of errors, an error of <1 mm was measured at all 7 landmarks. Operation time was greatly shortened with cutting guides and customized plates. Although this study is a single-case study, this study shows increased accuracy and efficacy from application of 3D virtual surgery, CAD/CAM, and 3D printing technology.

Surgical Accuracy of Positioning the Maxilla in Patients With Skeletal Class II Malocclusion Using Computer-Aided Design and Computer-Aided Manufacturing-Assisted Orthognathic Surgery.

OBJECTIVE: To evaluate the surgical accuracy of positioning the maxilla in patients with skeletal class II malocclusion using computer-aided design and computer-aided manufacturing (CAD/ CAM)-assisted orthognathic surgery. MATERIALS AND METHODS: The samples consisted of 10 patients with skeletal class II malocclusion, whose cone-beam computed tomographys taken before and immediately after surgery were available and who underwent bimaxillary orthognathic surgery by a single surgeon using Le Fort I osteotomy and bilateral sagittal split ramus osteotomy at the Department of Oral and Maxillofacial Surgery, Seoul National University Dental Hospital, Seoul, South Korea between January 2018 and December 2019. After virtual surgical planning was performed using the FACEGIDE system (Korea), surgical cutting guides, intermediate splints, and custom-made titanium mini-plates were fabricated using CAD/CAM technique. Using 8 landmarks (anterior nasal spine, point A, #16, #13, contact point between #11 and #21, #23, #26, posterior nasal spine), the mean differences between the virtually planned (Virtual) and actual postsurgical position of the maxilla (Actual) in the three-dimensional coordinates (ΔActual-Virtual) and their mean absolute deviations were investigated. RESULTS: The mean differences of 8 landmarks were 0.42 mm left side movement in the transverse coordinate, 0.15 mm forward movement in the sagittal coordinate and 0.10 mm downward movement in the vertical coordinate. Their mean absolute deviations were 0.98, 0.67, and 0.62 mm in the sagittal, vertical, and transverse coordinates, respectively. CONCLUSIONS: Since the mean difference was less than 0.5 mm and the range of error was less than 1.0 mm, CAD/CAM-assisted orthognathic surgery might have a high degree of surgical accuracy and clinical relevance in the positioning of the maxilla.

Accuracy of Intentional Change of Frontal Ramal Inclination From Virtual to Actual Orthognathic Surgery Using Computer-Aided Design and Computer-Aided Manufacturing-Made Customized Metal Plates.

OBJECTIVE: To evaluate the accuracy of intentional change of the frontal ramal inclination (FRI) from virtual to actual orthognathic surgery. MATERIALS AND METHODS: A total of 16 patients who underwent orthognathic surgery for correction of facial asymmetry and took cone-beam computed tomography 2 to 3 weeks before surgery (T0) and 3 days after surgery (T1A) were selected. After reorientation of cone-beam computed tomography, the digital imaging and communications in medicine data was converted to StereoLithography format for the merging with dental cast scan using R2GATE software (MegaGen Implant, Daegu, Republic of Korea). During virtual surgery using Geomagic Freeform Plus software (3D Systems, Rock Hill, SC), bilateral FRI discrepancy was corrected (T1V) and 3D-printed surgical wafers and customized metal plates were manufactured using 3D printer (Meg-printer II, MegaGen Implant, Daegu, Republic of Korea) and computer numerical control milling machine (ARDEN, TPS Korea Ltd., Gwangju, Republic of Korea) accordingly. During surgery, FRI correction was done using surgical guide and customized plates. The difference between preoperative and virtually corrected FRI (Δ T1V - T0) and the difference between preoperative and actually surgically corrected FRI (Δ T1A - T0) were measured, and the achieved FRI correction in percentage ([Δ T1A - T0]/[Δ T1V - T0] × 100) was calculated. RESULTS: The mean absolute value of virtual FRI correction (Δ T1V - T0) and actual FRI correction (Δ T1A - T0) were 3.0° and 2.9°, respectively (n = 32). In the increased (medial rotation) FRI group, mean virtual and actual correction amount were 3.2° and 2.8° (n = 17), whereas in the decreased (lateral rotation) group, mean virtual and actual correction amount were -2.7° and -3.1°, respectively (n = 15). No statistically significant difference between virtual and actual measurements were present in either group. The mean achievement rate was 102.8%. Tendency of undercorrection (91.2%) in the increased FRI group and overcorrection (116.0%) in the decreased FRI group were noted. DISCUSSION: With virtual planning and computer-aided design and computer-aided manufacturing-made customized plates, it is possible to correct FRI during orthognathic surgery of patients with facial asymmetry.

Three-Dimensional Changes in the Mandibular Proximal Segment After Using a Surgery-First Approach in Patients With Class III Malocclusion and Facial Asymmetry.

ABSTRACT: This study was performed to evaluate condylar position and angulation after asymmetric mandibular setback between a conventional (CA) and surgery-first approach (SFA) using three-dimensional analysis. The condylar positions of 30 patients with skeletal Class III malocclusion and facial asymmetry who underwent 1-jaw (sagittal split ramus osteotomy) or 2-jaw orthognathic surgery (Le Fort I osteotomy and sagittal split ramus osteotomy) with CA (n = 18) or SFA (n = 12) from 2 university hospitals were studied. The three-dimensional assessment of condylar changes was performed using computed tomography images at the initial time point (T0) and at least 6 months after surgery (T1). Segmentation of condyles and cranial base assessment from cone-beam computed tomography images were performed using ITK-SNAP software (version 3.4.0). Condylar position and angulation changes were calculated using 3D Slicer software (version 4.10.2), and statistical analysis was performed. No significant translational or rotational condylar changes were observed between the deviated and non-deviated sides in each group or between the CA and SFA groups except yaw ( p = 0.014). Linear mixed-model analysis and multi-variate analysis showed no significant difference between the CA and SFA groups. Surgery-first approach might not be associated with more harmful effects on the condylar position and angulation changes as compared with CA.

Characterization of Treatment Modalities for Patients With Syndromic Craniosynostosis in Relation to Degree of Midface Hypoplasia and Patient's Age Using Longitudinal Follow-Up Data.

ABSTRACT: The purpose of this study was to investigate the type and frequency of use of treatment modalities (Tx-Mods) in patients with syndromic craniosynostosis (SC) using longitudinal follow-up data. A total of 28 patients with SC (24 Crouzon, 2 Apert, and 2 Antley-Bixler syndromes), who were treated at the Department of Orthodontics, Seoul National University Dental Hospital, Seoul, South Korea between 1998 and 2020, was included. According to the degree of midface hypoplasia (MH) at the initial visit (T1), the patients were divided into the mild-MH (78°≤SNA < 80°, n = 8), moderate-MH (76≤SNA < 78°, n = 7), and severe-MH (SNA < 76°, n = 13) groups. T1-age and Tx-Mods, including cal-varial surgery (CALS), orthopedic treatment (OPT), fixed orthodontic treatment, and midface advancement surgery in childhood (MAS-child) and adulthood (MAS-adult), were investigated. Complexity of MAS-adult was graded as follows: 0, no surgery; 1, orthognathic surgery; 2, distraction osteogenesis (DOG); 3, combination of distraction osteogenesis and orthognathic surgery. Then, statistical analysis was performed. Percentage distribution of Tx-Mods was 71.4% in CALS, 21.4% in MAS-child, 42.9% in OPT, 100% in fixed orthodontic treatment, and 89.3% in MAS-adult. 92.9% of patients underwent MAS more than once. The number of MAS increased according to the severity of MH ( P < 0.05). The complexity of MAS-adult increased as T1-age and severity of MH increased (all P < 0.05); whereas it decreased when CALS and OPT were performed (all P < 0.05). However, MAS in childhood did not guarantee the avoidance of additional MAS in adulthood ( P > 0.05). These findings may be used as basic guidelines for successful treatment planning and prognosis prediction in patients with SC.

Accuracy of auto-identification of the posteroanterior cephalometric landmarks using cascade convolution neural network algorithm and cephalometric images of different quality from nationwide multiple centers.

INTRODUCTION: The purpose of this study was to evaluate the accuracy of auto-identification of the posteroanterior (PA) cephalometric landmarks using the cascade convolution neural network (CNN) algorithm and PA cephalogram images of a different quality from nationwide multiple centers nationwide. METHODS: Of the 2798 PA cephalograms from 9 university hospitals, 2418 images (2075 training set and 343 validation set) were used to train the CNN algorithm for auto-identification of 16 PA cephalometric landmarks. Subsequently, 99 pretreatment images from the remaining 380 test set images were used to evaluate the accuracy of auto-identification of the CNN algorithm by comparing with the identification by a human examiner (gold standard) using V-Ceph 8.0 (Ostem, Seoul, South Korea). Pretreatment images were used to eliminate the effects of orthodontic bracket, tube and wire, surgical plate, and surgical screws. Paired t test was performed to compare the x- and y-coordinates of each landmark. The point-to-point error and the successful detection rate (range, within 2.0 mm) were calculated. RESULTS: The number of landmarks without a significant difference between the location identified by the human examiner and by auto-identification by the CNN algorithm were 8 on the x-coordinate and 5 on the y-coordinate, respectively. The mean point-to-point error was 1.52 mm. The low point-to-point error (<1.0 mm) was observed at the left and right antegonion (0.96 mm and 0.99 mm, respectively) and the high point-to-point error (>2.0 mm) was observed at the maxillary right first molar root apex (2.18 mm). The mean successful detection rate of auto-identification was 83.3%. CONCLUSIONS: Cascade CNN algorithm for auto-identification of PA cephalometric landmarks showed a possibility of an effective alternative to manual identification.

Vertical bony step between proximal and distal segments after mandibular setback is related with relapse: A cone-beam computed tomographic study.

INTRODUCTION: Vertical bony step (VBS) occurs between proximal and distal segments of the mandible during mandibular setback surgery with bilateral sagittal split ramus osteotomy. The purpose of this study was to investigate whether VBS is correlated with the relapse of mandibular setback using 3-dimensional models constructed from cone-beam computed tomography. METHODS: The subjects consisted of 30 patients who underwent bilateral sagittal split ramus osteotomy for a mandibular setback. Double jaw surgery was performed in 18 patients, and isolated mandibular setback surgery was performed in 12 patients. Cone-beam computed tomography scans were taken at pretreatment (T0), postsurgery (T1), and posttreatment (T2). Treatment changes and the correlations between measurements were evaluated. RESULTS: The mean mandibular setback was -11.9 mm, and the mean VBS was -5.6 mm. Correlations with the relapse of mandibular setback were found in the amount of mandibular setback (T1 - T0), development of VBS (T1 - T0), posterior movement of the proximal segment (T1 - T0), counterclockwise rotation of symphysis (T2 - T1), and the resolution of VBS (T2 - T1). CONCLUSIONS: The development and resolution of VBS were correlated with the relapse of mandibular setback. Minimizing VBS is recommended to reduce the relapse of mandibular setback.