• Imaging Science in Dentistry
• /
• v.51 no.3
• /
• pp.299-306
• /
• 2021

Purpose: This study aimed to propose a fully automatic landmark identification model based on a deep learning algorithm using real clinical data and to verify its accuracy considering inter-examiner variability. Materials and Methods: In total, 950 lateral cephalometric images from Yonsei Dental Hospital were used. Two calibrated examiners manually identified the 13 most important landmarks to set as references. The proposed deep learning model has a 2-step structure-a region of interest machine and a detection machine-each consisting of 8 convolution layers, 5 pooling layers, and 2 fully connected layers. The distance errors of detection between 2 examiners were used as a clinically acceptable range for performance evaluation. Results: The 13 landmarks were automatically detected using the proposed model. Inter-examiner agreement for all landmarks indicated excellent reliability based on the 95% confidence interval. The average clinically acceptable range for all 13 landmarks was 1.24 mm. The mean radial error between the reference values assigned by 1 expert and the proposed model was 1.84 mm, exhibiting a successful detection rate of 36.1%. The A-point, the incisal tip of the maxillary and mandibular incisors, and ANS showed lower mean radial error than the calibrated expert variability. Conclusion: This experiment demonstrated that the proposed deep learning model can perform fully automatic identification of cephalometric landmarks and achieve better results than examiners for some landmarks. It is meaningful to consider between-examiner variability for clinical applicability when evaluating the performance of deep learning methods in cephalometric landmark identification.

• Journal of Oral Medicine and Pain
• /
• v.47 no.3
• /
• pp.135-143
• /
• 2022

Purpose: This study aimed to evaluate polysomnographic and cephalometric characteristics of patients with OSA according to obesity level based on the World Health Organization (WHO) Asian-Pacific BMI criteria. Methods: One hundred and thirty-one consecutive patients with obstructive sleep apnea (OSA) were evaluated using standard level 1 polysomnography and cephalometric analyses. The subjects were categorized into normal, overweight and obese groups according to the WHO Asian-Pacific BMI criteria. Respiratory indices and cephalometric parameters were compared among groups. Results: The 131 patients consisted of 111 males and 20 females, with a mean age of 44.1±12.4 years. The mean value of BMI was 25.3±3.4 kg/m² for all subjects, 20.6±2.2 kg/m² for normal (n=27), 24.0±0.5 kg/m² for overweight (n=33) and 27.6±2.2 kg/m² for obese (n=71). The obese group had a significantly higher apnea-hypopnea index (AHI) and respiratory arousal index and lower oxygen saturation level than the normal group (p<0.05). Total AHI, mean oxygen saturation level and respiratory arousal index were significantly correlated with BMI (p<0.001). A longer soft palate and anterior position of the hyoid bone were significantly correlated with BMI level (p ≤0.05). Conclusions: Obese patients have a higher risk of compromised craniofacial skeletal features and soft tissue structures, and severe OSA than non-obese patients.

• Journal of Oral Medicine and Pain
• /
• v.48 no.1
• /
• pp.37-37
• /
• 2023

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.5 no.1
• /
• pp.22-25
• /
• 1975

The author has studied on the profiles on Korean young male and female by the cephalometric roentgenograms, the subjects were consisted of 104 persons with normal occlusion and the measurements were done on the soft and hard structures. The mean and standard deviations in the subjects were calculated and compared between male and female. The author has obtained the following results: 1. In the linear measurement, male was longer than female in most measuring points. 2. Compared with the angulation of soft tissue, there were no differences between both sexes. 3. In the skeletal patterns, there were some differences between male and female, but there was a similarity to the facial profile. 4. The individual variations on the cephalometric profiles on the subjects were recognized, especially, in the lower facial part.

• The korean journal of orthodontics
• /
• v.15 no.1
• /
• pp.23-42
• /
• 1985

The morphology and growth changes of the Craniofacial Complex are important in orthodontics and cephalometrics for analysis and evaluation of facial bone growth are widely used. The author analyzed the data using lateral cephalometric roentgenogram of 75 Korean male and 50 Korean females age of 6 to 10 with normal occlusion to provide informations-relative rates of facial bone growth of Korean which is to be contributed in Korean standard. The results were as follows: 1. Means, standard deviation and coefficient of variation of Korean children were obtained. 2. The item which showed significent difference between male and female was craniofacial height in absolute dimension. 9. No difference of sex was showed in increment of craniofacial height and depth. 4. Among the craniofacial depth increments, the lower facial depth dimension increased most, midfacial depth dimension increased less, and cranial depth dimension increased the least. 5. The horizontal body of mandible showed rapid growing tendency more than did the ascending ramus.

• The korean journal of orthodontics
• /
• v.45 no.6
• /
• pp.322-332
• /
• 2015

A 12-year-old girl was referred to our clinic for evaluation of an unaesthetic dental appearance. All permanent teeth were erupted, while the deciduous maxillary right canine was retained. Cone-beam computed tomography revealed a complete transposition of the maxillary left canine and first premolar involving both the crowns and the roots. Initial cephalometric analysis showed a skeletal Class III pattern, with a slight maxillary retrusion and a compensated proclination of the upper incisors. The patient's teeth were considered to be in the correct position; therefore, we decided to attempt treatment by correcting the transposition and using only orthodontic compensation of the skeletal Class III malocclusion. After 25 months of active orthodontic treatment, the patient had a Class I molar and canine relationship on both sides, with ideal overbite and overjet values. Her profile was improved, her lips were competent, and cephalometric evaluation showed acceptable maxillary and mandibular incisor inclinations. The final panoramic radiograph showed that good root parallelism was achieved. Two-year follow-up intraoral photography showed stable results.

• The korean journal of orthodontics
• /
• v.19 no.1 s.27
• /
• pp.7-20
• /
• 1989

A serial cephalometric study was undertaken to define the growth of the soft tissue facial profile in Korean children. The sample was composed of 25 males and 15 females for whom yearly cephalometric records were taken from the ages of 6 to 13 years. From the tracings, points on skeletal and soft tissue profiles were located and recorded on magnetic tape utilizing a Calcomp Talos RP660 X-Y digitizer. Linear and angular measurements of soft tissues were made directly from tape in a Cyber 174-16 computer after cephalometric enlargement had been corrected. A statistical evaluation was made of the data and the average profile diagrams in male and female were described by a Calcomp 960 pen plotter. On the basis of the findings of this study, the following trends were established. 1. The most prominent growth in soft tissue facial profile thickness was the nose and the least was the forehead. 2. The general growth direction of the soft facial tissue to the cranium described the downward and forward. 3. The degree of soft tissue facial convexity was decidely more than that exhibited earlier in life even though the soft tissue chin had protruded to the cranium. 4. The measurements indicated a general tendency for males to have larger nose and more convex and long soft tissue facial profile than did females. 5. Males showed significantly more growth than females in base of the upper lip and height of the upper anterior facial profile. 6. There was a difference between males and females in the rates of soft tissue facial profile growth. 7. Korean children showed less convex in the soft tissue profile convexity than did American children.

• Imaging Science in Dentistry
• /
• v.41 no.2
• /
• pp.79-84
• /
• 2011

Purpose : The aim of the present study was to investigate the disagreement of cephalometric analysis depending on the reference determination of midsagittal plane on three-dimensional computed tomography. Materials and Methods : A total of 102 young women with class III dentofacial deformity were evaluated using three-dimensional computed tomography. The cranial and facial midsagittal planes were defined and the amounts of jaw deviation were calculated. The amounts of jaw deviation were compared with paired t-test (2-tailed) and Bland-Altman plot was drawn. Results : The landmark tracing were reproducible ($r{\ge}.978$). The jaws relative to the cranial midsagittal plane were 10-17 times more significantly deviated than to the facial midsagittal plane (P<.001). Bland-Altman plot demonstrated that the differences between the amounts of jaw deviation from two midsagittal planes were not normally distributed versus the average of the amounts of jaw deviation from two midsagittal planes. Conclusion : The cephalometric analyses of facial asymmetry were significantly inconsistent depending on the reference determination of midsagittal plane. The reference for midsagittal plane should be carefully determined in three-dimensional cephalometric analysis of facial asymmetry of patients with class III dentofacial deformity.

• The korean journal of orthodontics
• /
• v.50 no.3
• /
• pp.181-187
• /
• 2020

Objective: This study aimed to evaluate the differences in cephalometric characteristics and skeletal maturation in girls with central precocious puberty (CPP) via lateral and hand-wrist radiographs. We also aimed to identify the indicators that are most effective for determining skeletal maturity in these patients. Methods: The study included 70 Korean girls (mean age, 8.5 ± 0.5 years) diagnosed with CPP at the Department of Pediatrics, and 48 normal healthy age-matched girls who visited the Department of Orthodontics and had no history of hormone treatment or growth problems. Skeletal maturation was evaluated using lateral cephalometric and hand-wrist radiographs using cervical vertebrae maturation indicators (CVMI) and skeletal maturity indicators (SMI). Results: The mean mandibular plane angle was smaller in the CPP group than in the control group (35.8° ± 4.9° vs. 39.0° ± 6.5°), resulting in greater posterior facial height (p = 0.003). SMI was significantly greater in the CPP group (3.5 ± 1.4 vs. 2.0 ± 1.0) than in the control group (p = 0.001) and was significantly associated with CPP (r = 0.492; p = 0.001), whereas CVMI was not. Conclusions: In comparison with the control group, the CPP group exhibited a smaller mandibular plane angle, greater posterior facial height, and greater skeletal maturation. SMI may be more suitable than CVMI for determining skeletal maturation in CPP. Hand-wrist radiography is recommended in addition to lateral cephalogram for predicting growth in girls with CPP.

• The korean journal of orthodontics
• /
• v.51 no.2
• /
• pp.77-85
• /
• 2021

Objective: To evaluate the accuracy of a multi-stage convolutional neural network (CNN) model-based automated identification system for posteroanterior (PA) cephalometric landmarks. Methods: The multi-stage CNN model was implemented with a personal computer. A total of 430 PA-cephalograms synthesized from cone-beam computed tomography scans (CBCT-PA) were selected as samples. Twenty-three landmarks used for Tweemac analysis were manually identified on all CBCT-PA images by a single examiner. Intra-examiner reproducibility was confirmed by repeating the identification on 85 randomly selected images, which were subsequently set as test data, with a two-week interval before training. For initial learning stage of the multi-stage CNN model, the data from 345 of 430 CBCT-PA images were used, after which the multi-stage CNN model was tested with previous 85 images. The first manual identification on these 85 images was set as a truth ground. The mean radial error (MRE) and successful detection rate (SDR) were calculated to evaluate the errors in manual identification and artificial intelligence (AI) prediction. Results: The AI showed an average MRE of 2.23 ± 2.02 mm with an SDR of 60.88% for errors of 2 mm or lower. However, in a comparison of the repetitive task, the AI predicted landmarks at the same position, while the MRE for the repeated manual identification was 1.31 ± 0.94 mm. Conclusions: Automated identification for CBCT-synthesized PA cephalometric landmarks did not sufficiently achieve the clinically favorable error range of less than 2 mm. However, AI landmark identification on PA cephalograms showed better consistency than manual identification.