Three-dimensional changes in the cranial base associated with soft-diet feeding.

BACKGROUND: Masticatory activity affects the morphology of the maxillo-mandibular complex, however, its influence on the cranial base remains to be elucidated. The recent integration of quantitative morphometric analysis with 3D imaging enabled a comprehensive and high-resolution morphological characterization of the craniofacial complex. We aimed to investigate the influence of masticatory activity on the morphology of the growing cranial base by three-dimensional (3D) geometric morphometric approach using micro-CT. METHODS: The micro-CT data was reanalyzed to illustrate the 3D shape of the cranial base, and wireframe models were generated by connecting landmarks on the images. In the original study, mice were fed a soft diet (SD) of powdered pellets or a conventional hard diet (HD) for 6 weeks from 3 to 9 weeks of age, immediately after weaning. A principal component (PC) analysis analyzed shape variations and assessed their significance, while canonical variate (CV) analysis facilitated the comparison and differentiation of groups based on shape, unveiling meaningful shape distinctions. RESULTS: Three PCs were extracted that significantly separated the SD and HD groups among those explaining variations in shape. These PCs were related to the length of the sphenoid bone, the width of the anterior part of the sphenoid bone, and the length of the cranial base. Furthermore, one CV effectively distinguished SD from HD, and CV analysis showed that the sphenoid was shortened in the length and narrowed at the border of the temporal bone in SD mice. CONCLUSIONS: Masticatory loading affects the skeletal development of the cranial base. The morphology of the sphenoid bone was affected in both the sagittal and transverse axes.

Oral frailty is a risk factor for body weight loss after gastrectomy: a single-center, retrospective study.

BACKGROUND: Body weight loss (BWL) is a serious complication of gastrectomy in patients with gastric cancer (GC). Nutritional intervention alone is inadequate for preventing BWL, and a new approach is needed. Oral frailty among older adults has recently attracted attention. This study aimed to investigate masticatory ability and BWL after gastrectomy. METHODS: This was a single-center, retrospective study. Functional tooth units (FTU) were used to measure masticatory ability. Patients with FTU < 4 were defined as low FTU group and FTU ≥ 4 as high FTU group. The BWL was compared between the two groups. RESULTS: Sixty patients who underwent distal gastrectomy for GC from March 2022 to January 2023 were enrolled in this study. The median FTU was 3 (range 0-12). The low-FTU group (FTU < 4) included 29 patients, while the high-FTU group (FTU ≥ 4) included 31 patients. The %BWL in the low FTU group was significantly higher than that in the high-FTU group at 1 and 3 months (p = 0.003 and p = 0.017, respectively). The risk factors associated with a %BWL > 5 at 1 and 3 months after gastrectomy were analyzed using logistic regression analysis. Only FTU < 4 was an independent risk factor after gastrectomy for GC in univariate and multivariate analyses (p = 0.028 and p = 0.006, respectively). CONCLUSIONS: Low FTU in patients with preoperative GC was a risk factor for %BWL 1 and 3 months postoperatively. Appropriate oral interventions may be useful in improving the postoperative nutritional status after gastrectomy.

Applications and performance of artificial intelligence models in removable prosthodontics: A literature review.

PURPOSE: In this narrative review, we present the current applications and performances of artificial intelligence (AI) models in different phases of the removable prosthodontic workflow and related research topics. STUDY SELECTION: A literature search was conducted using PubMed, Scopus, Web of Science, and Google Scholar databases between January 2010 and January 2023. Search terms related to AI were combined with terms related to removable prosthodontics. Articles reporting the structure and performance of the developed AI model were selected for this literature review. RESULTS: A total of 15 articles were relevant to the application of AI in removable prosthodontics, including maxillofacial prosthetics. These applications included the design of removable partial dentures, classification of partially edentulous arches, functional evaluation and outcome prediction in complete denture treatment, early prosthetic management of patients with cleft lip and palate, coloration of maxillofacial prostheses, and prediction of the material properties of denture teeth. Various AI models with reliable prediction accuracy have been developed using supervised learning. CONCLUSIONS: The current applications of AI in removable prosthodontics exhibit significant potential for improving the prosthodontic workflow, with high accuracy levels reported in most of the reviewed studies. However, the focus has been predominantly on the diagnostic phase, with few studies addressing treatment planning and implementation. Because the number of AI-related studies in removable prosthodontics is limited, more models targeting different prosthodontic disciplines are required.

Sex differences in airway volume and 3-dimensional shape in Japanese adults.

(1) To establish normative data for three-dimensional (3D) measurements of the upper airway in young Japanese adults, and (2) to investigate sex-related differences in linear and volumetric measurements, as well as shape. This study employed cone-beam computed tomography (CBCT) images of 56 Japanese young adults preselected from among 1000 patients, so that samples matched a historic 2D cephalometric cohort with normal occlusion using propensity score matching. Three-dimensional models of the oropharynx and hypopharynx were reconstructed from CBCT images and their volumes were calculated. We defined 20 landmarks on the surface of the 3D model and performed seven linear measurements between them. The mean and standard deviation of the linear measurements were calculated as the normative data for each sex as well as the volumes. Sex-related differences were analyzed using t-test (p < 0.05). Principal component discriminant analysis of the coordinate values of the landmarks was also performed to examine sex differences in shape. The normative ranges of the 3D measurements of the oropharynx and hypopharynx were determined according to sex. Sex-related differences in the measurement results were observed in hypopharyngeal length but not in volume. The hypopharynx length in males was significantly longer than that in females. The discriminant analysis showed that males tended to show longer and straight shapes, while females showed inversed triangular shapes from the frontal view. This result will allow clinicians to evaluate how patient airway characteristics differ from the normative 3D morphology of the upper airway.

Orofacial Pain, Musical Performance and Associated Coping Behaviors, Psychological Distress and Disability among Asian Young Adults.

Musicians often report orofacial pain (OFP) and performance-related psychological distress related to occupational neuromuscular overuse, but to date, no study has been performed in Asian musicians to assess these factors. This study evaluated OFP, psychological distress, coping behaviors, and disability among Asian musical performers. A total of 201 participants in Singaporean music ensembles were surveyed from which 159 met the inclusion criteria for vocalists or instrumentalist musicians (mean age 20.26 ± 2.20 years). Self-administered questionnaires assessed musical practices, jaw/neck pre-conditioning exercises, pain-related temporomandibular disorders (TMD), OFP descriptors, pain chronicity and disability, coping behaviors and psychological distress. Univariate and multi-variate analyses were carried out. OFP, while performing, was more than two-fold higher in instrumentalists when compared to vocalists (41.4-48% vs. 17.2%, p = 0.002). A similar trend occurred for OFP that progresses while playing (p = 0.035) and for persistent OFP that reduces playing (p = 0.001). There were no differences in psychological distress, pain coping and disability between groups. Vocalists were found to practice jaw/neck pre-conditioning exercises more frequently (75% vs. 4-12.9% in instrumentalists, p < 0.0001). While performing, Asian vocalists reported less OFP when compared to instrumentalists. Future prospective studies are needed to confirm if pre-conditioning exercises play a protective role against OFP in vocalists.

Morphologic variations in the craniofacial structures in Japanese adults and their relationship with sex differences.

INTRODUCTION: The morphology of the human face varies broadly, with genetic and environmental factors determining these variations. Examining variations in the 3-dimensional (3D) craniomandibular morphology and identifying related factors (eg, sex differences) are important in orthodontic clinics. This study observed shape variations in the 3D facial morphology of Japanese adults showing skeletal Class 1 malocclusion and examined the association of sexual dimorphism with shape variations. METHODS: Sixty cone-beam computed tomography images of Japanese adults (30 males and 30 females) with skeletal Class I malocclusion were employed. In each cone-beam computed tomography image, wire mesh fitting was conducted as previously described. A principal component (PC) analysis after Procrustes registration and the PC clustering method was conducted to observe the shape variations. A PC regression analysis was conducted to determine the sexual morphologic characteristics. RESULTS: Nine PCs depicting 62% of the morphology were determined. Four typical phenotypes were found, mainly related to mandibular protrusion (PC1) and the vertical divergence of the face (PC2). PCs related to sex determination were PC3 (robustness of the mandibular angle in males), PC5 (greater size and shape of the coronoid and mastoid processes in males), and PC7 (greater maxillary width in males), accounting for 16% of total variations. CONCLUSIONS: The major shape variations in skeletal Class 1 subjects were related to nonsexual dimorphic characteristics (ie, mandibular protrusion and facial divergence). Sexual dimorphic characteristics were evaluated in detail and accounted for 16% of total morphologic variations.

Prognostic Factors for Orthognathic Surgery in Children With Cleft Lip and/or Palate: Dentition and Palatal Morphology.

To determine dental and palatal morphology in children with cleft lip and/or palate (CL/P) and identify morphological prognostic factors for orthognathic surgery (OGS).Retrospective cohort study.Orthodontic department of a university dental hospital.This study included 80 patients with bilateral and unilateral CL/P who had lateral cephalograms at the ages of 7 (T1), 15 (T2) years, and a dental plaster model at T1.Plaster models at T1 were scanned with a three-dimensional (3D) scanner. Morphological features were extracted from 3D models with geometric morphometrics software as principal components (PCs). The combinations of the PCs and other predictive factors (ie, the No. of clefts in the lip and alveolus, the palatal repair method, sex, cephalometric variables at T1, and the No. of missing teeth) were examined by logistic regression to determine the predictability for OGS. The need for OGS and skeletal and dental discrepancies at T2 were examined as outcomes.Shrinkage of the palate, including vertical shallowing and transverse narrowing of the posterior maxilla and cleft-side asymmetry of the anterior maxilla at T1, as well as the No. of clefts in the lip and alveolus, the palatal repair method, male sex, several cephalometric variables for the sagittal and vertical dimensions, and the No. of missing teeth, were found to be predictive factors for OGS.Morphological prognostic factors for OGS in children with CL/P were determined.

Relationship Between Stigma Experience and Self-Perception Related to Facial Appearance in Young Japanese Patients with Cleft lip and/or Palate.

OBJECTIVE: To examine the relationship between stigma experience related to facial appearance in Japanese youths with cleft lip and/or palate (CL/P) and their self-perception. DESIGN: A cross-sectional study. PARTICIPANTS: Sixty-nine Japanese youths with CL/P (11-18 years old). OUTCOME MEASURES: The participants' stigma experience in relation to facial appearance (measured with 7 single contextual scale items) and their self-perception (measured with 5 domain scores based on 30 perceptual items) were assessed using the Japanese version of the Youth Quality of Life Instrument-Facial Differences Module. Participants were categorized into high and low self-perception subgroups with a threshold of 1 standard deviation for each domain. The frequency of stigma experiences was compared between the following 2 subgroups: age, sex, cleft palate only versus other cleft, and high versus low self-perception. Correlations between the responses regarding stigma and all domain scores were examined. RESULTS: Sixteen percent of the participants reported experiencing stigma. Hearing others say something about their face occurred significantly more frequently in youths 15 to 18 years of age than in youths 11 to 14 years of age. Stigma frequency was not found to differ by sex or cleft type. Stigma experiences were significantly more frequent for youth with higher scores across negative self-perception domains as well as higher coping skills. Significant correlations were identified between responses regarding stigma items and all domain scores (r = 0.27-0.63, p < .05). CONCLUSIONS: It was found that stigma experiences related to facial appearance may influence negative self-perceptions of facial differences as well as higher coping skills among Japanese youths with CL/P.

Influence of the nutritional status on facial morphology in young Japanese women.

Evidence regarding the possible influence of nutritional status on the facial morphology has thus far been insufficient. We examined whether or not the physical body compositions and dietary behaviors were correlated with any morphological characteristics of the face. One hundred and fifteen young Japanese women participated. Variables representing the dietary behaviors were extracted from self-reported survey data, and corresponding three-dimensional (3D) facial images and body compositions were examined. Multivariate analyses identified significant relationships between the nutritional status and facial topography (p < 0.05). The clustering method revealed the existence of three dietary condition patterns ("balanced diet", "high-calorie-diet" with obesity tendency, and "imbalanced low-calorie-diet" with sarcopenic obesity tendency). Among these three patterns, a round face (increased facial width; analysis of variance [ANOVA], p < 0.05) was observed in the high-calorie-diet pattern, while the imbalanced low-calorie-diet pattern showed a more masculine face (increased face height, decreased eye height, increased non-allometric sexual shape differences; ANOVA, p < 0.05), thus suggesting the possibility of sex-hormonal influences. In summary, the body composition and dietary behaviors were found to influence the facial morphology, and potential biological influences were discussed.

Hemodynamic Changes in the Masseter and Superior Orbicularis Oris Muscles before and after Exercise Load: A Comparison between Young Adult Women and Middle-Aged to Old Adult Women.

Background: The vascularity index (VI) is useful for measuring the hemodynamics on ultrasound imaging. However, there are no reports concerning the application of the VI to facial muscles. Objective: The aim of this study was (1) to establish a method of measuring the hemodynamics in facial muscles in a constant way and (2) to evaluate the hemodynamic changes in the masseter and superior orbicularis oris muscles (SOOMs) before and after exercise load in two subject groups of females of different ages. Methods: (1) The VI in the SOOM was calculated, and the test-retest reliability was assessed in seven healthy adults. (2) The VIs in the left-side masseter and SOOM were calculated in 3 sessions: before exercise loading (T0), immediately after loading (T1), and 5 minutes after T1 (T2) for the young adult group (YAG, n = 20; age range, 20-35 years) and the middle-aged to old group (MOG, n = 20; age range, 50-70 years). Tasks were gum chewing for the masseter muscle and lip sealing for the SOOM. The differences in the mean peak flows between two sessions were examined. Results: (1) Significant differences were not noted for the repeatedly measured average volumes of blood flow with good test-retest agreement (intraclass correlation coefficient = 0.81). (2) In both muscles of the YAG, there were a significant increase in T1 compared with T0 and a significant decrease in T2 compared with T1 (all p < 0.05). In both muscles of the MOG, no significant differences were noted in either comparison. Conclusions: A method of measuring the hemodynamics in facial muscles was developed and showed good reliability. Changes in the blood flow after exercise load in these muscles may vary with age in women.

A three-dimensional cephalometric analysis of Japanese adults and its usefulness in orthognathic surgery: A retrospective study.

This study aimed to establish a three-dimensional (3D) cephalometric analysis of craniofacial morphology and discuss its theoretical usefulness in orthognathic patients. Cone-beam computed tomography (CBCT) images of Japanese subjects with skeletal Class I malocclusion before treatment were selected from among 1000 patients so that samples matched a historic 2D cephalometric cohort with normal occlusion using propensity score matching. In each CBCT image, 67 3D measurements were calculated based on manually identified landmarks. The mean and standard deviation of the measurements were calculated and used as the normative range for each sex. To confirm the usefulness of the 3D measurements, pre- and post-treatment CT data of nine jaw deformity patients who underwent orthognathic surgery with two-dimensional planning (2DP) in the past were used. Pre- and post-treatment CT values were evaluated with a paired t-test as well as a Z-score, which was calculated using the aforementioned normative range, and then categorized into five groups ("deteriorated", "no improvement", "over-treatment", "no change", "improvement") with -1 < Z-score < 1 considered normal. Fifty-six patients were matched to normal skeletal 1 subjects. The normative range of 67 items indicating 3D craniofacial morphology of the Japanese was calculated. Postoperatively, the horizontal position of the pogonion to the mid-sagittal plane significantly decreased (p = 0.043) and "improved"; however, the ramus axis on the right side significantly increased (p = 0.005) and "deteriorated". Maxillary yaw and the horizontal position of the gonion also tended to "deteriorated". The normative range for the 3D cephalometric analysis in Japanese has been established. Given findings of deteriorated maxillomandibular yawing after surgery when using conventional 2DP, 3D cephalometric measurements should be used when planning jaw positions after surgery for orthognathic patients.

The validation of orthodontic artificial intelligence systems that perform orthodontic diagnoses and treatment planning.

AIM: This study was aimed to evaluate two artificial intelligence (AI) systems that created a prioritized problem list and treatment plan, and examine whether the performance of the aforementioned systems was equivalent to orthodontists. MATERIALS AND METHODS: A total of 967 consecutive cases [800: training; 67: validation; 100: evaluation (40: randomly selected for the clinical evaluation)] were used. We used a stored document that describes (1) the patient's clinical information, (2) the prioritized list, and (3) a treatment strategy without digital tooth movement. Sentences of (1) were vectorized according to the bag of words method (V); sentences of (2) and (3) were relabelled with 423 and 330 labels, respectively. AI systems that output labels for the prioritized list (subtask 1) and treatment planning (subtask 2) based on the vectors V were developed using a support vector machine and self-attention network, respectively, while the system was trained to improve precision and recall. Clinical evaluations were conducted by four orthodontists (no faculty or residents; peer group) in two sessions: in the first session, peer group and the developed AI systems created problem lists and treatment plans; in the second session, two of the peer group (not AI) evaluated these lists and plans, including the lists and plans of the AIs, by scoring them using 4-point scales [unacceptable (1) to ideal (4)]. Scores were compared among the system and peer group (Wilcoxon signed-rank test, P < 0.05). RESULTS: The precision after system training was 65% and 48% for subtasks 1 and 2 respectively, with recall of 55% and 48%, respectively. The clinical evaluation of the AI system for subtask 1 showed a mid-rank. For subtask 2, the AI system had a significantly lower score than the three panels but the same rank with one panel. CONCLUSIONS: Two AI systems that output a prioritized problem list and create a treatment plan were developed. The clinical system ability of the former system showed a mid-rank in the peer group, and the latter system was almost equivalent to the worst orthodontist.

Comparison of clinical outcomes between Invisalign and conventional fixed appliance therapies in adult patients with severe deep overbite treated with nonextraction.

INTRODUCTION: The use of aligner therapy for orthodontic treatment has increased substantially in the past decade. However, no study has compared treatment outcomes between the conventional fixed appliance and Invisalign therapies in patients with a severe deep overbite. METHODS: This study included 50 consecutive adult patients who underwent treatment with either Invisalign (n = 25; mean age, 23.3 ± 8.5 years) or a conventional fixed appliance (n = 25; mean age, 23.1 ± 6.5 years) to correct overbite >5 mm and >60% deep overbite. Cephalometric analysis and peer assessment rating was used to compare the clinical outcome between groups. RESULTS: Cephalometric analysis showed significant differences in N-Me (P = 0.0005) and Mp-L6 (P = 0.0001) between Invisalign and fixed appliance treatment groups. No significant differences were observed in the peer assessment rating analysis or total treatment duration between the 2 groups. CONCLUSIONS: Both Invisalign and conventional fixed appliances were effective in treating patients with a severe deep overbite. Invisalign therapy may be preferable over conventional fixed appliance therapy in patients with high angle and deep overbite. However, because this study had a retrospective design, the results should be viewed with caution.

Comparison of 3-D mandibular surfaces generated by MRI and CT.

OBJECTIVE: The purpose of this study was to evaluate the errors of three-dimensional mandibular surfaces generated using magnetic resonance imaging (MRI) when computed tomography (CT) was set as the gold standard. SETTINGS AND SAMPLE POPULATION: Seven patients with orthognathic deformities who had undergone CT and MRI scans were included in the study. MATERIALS AND METHODS: Mandibular surfaces were generated on each CT and MR image by the surface-rendering method. Intra-individual reliability between CT and MRI was statistically tested by the confidence limits of agreement (LOA) for systematic error, 95% confidence interval minimal detectable change (MDC95 ) for random error and intra-class correlation coefficient (ICC). RESULTS: The average total error was 1.6 mm. The greatest MDC95 was observed in the coronoid region in all directions. The other regions showed MDC95 values of < 1.8 mm (transvers direction), 3.5 mm (vertical direction) and 1.7 mm (antero-posterior direction). ICCs showed 'almost perfect' agreement with respect to all regions. CONCLUSION: Random errors were quantified for 3-D rendering of the mandible from MRI data. Although the coronoid region showed the greatest errors, the other regions of the mandibular surfaces generated using MRI were able to be evaluated.

Evaluation of Facial Appearance-Related Quality of Life in Young Japanese Patients With Cleft Lip and/or Palate.

OBJECTIVE: (1) To confirm the reliability of a Japanese version of the Youth Quality of Life Instrument-Facial Differences Module (YQOL-FD); (2) to assess the quality of life (QoL) related to facial difference in Japanese youths with cleft lip and/or palate (CL/P) using this instrument; and (3) to examine the QoL differences according to age, sex, and cleft type. DESIGN: A cross-sectional study. SETTING: Japanese youths with CL/P were recruited through our hospital and asked to complete the YQOL-FD. PARTICIPANTS: Sixty-nine Japanese youths (age, 11-18 years) with CL/P. OUTCOME MEASURES: The domain scores of stigma, negative consequences, negative self-image, positive consequences, and coping in the YQOL-FD, and the reliability of such scores were evaluated. RESULTS: The instrument showed an acceptable internal consistency (Cronbach α = 0.74-0.92) and test-retest reliability (intraclass correlation coefficient = 0.94-0.98), except for the coping domain. The individual's domain scores were spread out from the lowest score to the high scores among all domains, thus indicating the negative and positive impacts of living with facial differences regarding their QoL may vary among individuals with CL/P. All domain scores in the 15- to 18-year-old group were significantly higher than those in 11- to 14-year-old group; there were no significant differences according to sex or cleft type. CONCLUSIONS: The instrument showed acceptable reliability, except for the coping domain. There were individual variations in QoL concerning the facial difference among Japanese youths with CL/P as measured by the YQOL-FD, suggesting the importance of individual evaluations. Perceptions were influenced by age, but not sex or cleft type.

Development of novel artificial intelligence systems to predict facial morphology after orthognathic surgery and orthodontic treatment in Japanese patients.

From a socio-psychological standpoint, improving the morphology of the facial soft-tissues is regarded as an important therapeutic goal in modern orthodontic treatment. Currently, many of the algorithms used in commercially available software programs that are said to provide the function of performing profile prediction are based on the false assumption that the amount of movement of hard-tissue and soft-tissue has a proportional relationship. The specification of the proportionality constant value depends on the operator, and there is little evidence to support the validity of the prediction result. Thus, the present study attempted to develop artificial intelligence (AI) systems that predict the three-dimensional (3-D) facial morphology after orthognathic surgery and orthodontic treatment based on the results of previous treatment. This was a retrospective study in a secondary adult care setting. A total of 137 patients who underwent orthognathic surgery (n = 72) and orthodontic treatment with four premolar extraction (n = 65) were enrolled. Lateral cephalograms and 3-D facial images were obtained before and after treatment. We have developed two AI systems to predict facial morphology after orthognathic surgery (System S) and orthodontic treatment (System E) using landmark-based geometric morphometric methods together with deep learning methods; where cephalometric changes during treatment and the coordinate values of the faces before treatment were employed as predictive variables. Eleven-fold cross-validation showed that the average system errors were 0.94 mm and 0.69 mm for systems S and E, respectively. The total success rates, when success was defined by a system error of < 1 mm, were 54% and 98% for systems S and E, respectively. The total success rates when success was defined by a system error of < 2 mm were both 100%. AI systems to predict facial morphology after treatment were therefore confirmed to be clinically acceptable.

Population affinity and variation of sexual dimorphism in three-dimensional facial forms: comparisons between Turkish and Japanese populations.

Examining the extent to which sex differences in three-dimensional (3D) facial soft tissue configurations are similar across diverse populations could suggest the source of the indirect evolutionary benefits of facial sexual dimorphism traits. To explore this idea, we selected two geographically distinct populations. Three-dimensional model faces were derived from 272 Turkish and Japanese men and women; their facial morphologies were evaluated using landmark and surface-based analyses. We found four common facial features related to sexual dimorphism. Both Turkish and Japanese females had a shorter lower face height, a flatter forehead, greater sagittal cheek protrusion in the infraorbital region but less prominence of the cheek in the parotid-masseteric region, and an antero-posteriorly smaller nose when compared with their male counterparts. The results indicated the possible phylogenetic contribution of the masticatory organ function and morphogenesis on sexual dimorphism of the human face in addition to previously reported biological and psychological characteristics, including sexual maturity, reproductive potential, mating success, general health, immune response, age, and personality.

Clinical applicability of automated cephalometric landmark identification: Part II - Number of images needed to re-learn various quality of images.

AIM: To estimate the number of cephalograms needed to re-learn for different quality images, when artificial intelligence (AI) systems are introduced in a clinic. SETTINGS AND SAMPLE POPULATION: A total of 2385 digital lateral cephalograms (University data [1785]; Clinic F [300]; Clinic N [300]) were used. Using data from the university and clinics F and N, and combined data from clinics F and N, 50 cephalograms were randomly selected to test the system's performance (Test-data O, F, N, FN). MATERIALS AND METHODS: To examine the recognition ability of landmark positions of the AI system developed in Part I (Original System) for other clinical data, test data F, N and FN were applied to the original system, and success rates were calculated. Then, to determine the approximate number of cephalograms needed to re-learn for different quality images, 85 and 170 cephalograms were randomly selected from each group and used for the re-learning (F85, F170, N85, N170, FN85 and FN170) of the original system. To estimate the number of cephalograms needed for re-learning, we examined the changes in the success rate of the re-trained systems and compared them with the original system. Re-trained systems F85 and F170 were evaluated with test data F, N85 and N170 from test data N, and FN85 and FN170 from test data FN. RESULTS: For systems using F, N and FN, it was determined that 85, 170 and 85 cephalograms, respectively, were required for re-learning. CONCLUSIONS: The number of cephalograms needed to re-learn for images of different quality was estimated.

Clinical applicability of automated cephalometric landmark identification: Part I-Patient-related identification errors.

OBJECTIVES: To determine whether AI systems that recognize cephalometric landmarks can apply to various patient groups and to examine the patient-related factors associated with identification errors. SETTING AND SAMPLE POPULATION: The present retrospective cohort study analysed digital lateral cephalograms obtained from 1785 Japanese orthodontic patients. Patients were categorized into eight subgroups according to dental age, cleft lip and/or palate, orthodontic appliance use and overjet. MATERIALS AND METHODS: An AI system that automatically recognizes anatomic landmarks on lateral cephalograms was used. Thirty cephalograms in each subgroup were randomly selected and used to test the system's performance. The remaining cephalograms were used for system learning. The success rates in landmark recognition were evaluated using confidence ellipses with α = 0.99 for each landmark. The selection of test samples, learning of the system and evaluation of the system were repeated five times for each subgroup. The mean success rate and identification error were calculated. Factors associated with identification errors were examined using a multiple linear regression model. RESULTS: The success rate and error varied among subgroups, ranging from 85% to 91% and 1.32 mm to 1.50 mm, respectively. Cleft lip and/or palate was found to be a factor associated with greater identification errors, whereas dental age, orthodontic appliances and overjet were not significant factors (all, P < .05). CONCLUSION: Artificial intelligence systems that recognize cephalometric landmarks could be applied to various patient groups. Patient-oriented errors were found in patients with cleft lip and/or palate.