Deep learning in the diagnosis of maxillary sinus diseases: a systematic review.

OBJECTIVES: To assess the performance of deep learning (DL) in the detection, classification, and segmentation of maxillary sinus diseases. METHODS: An electronic search was conducted by two reviewers on databases including PubMed, Scopus, Cochrane, and IEEE. All English papers published no later than February 7, 2024, were evaluated. Studies related to DL for diagnosing maxillary sinus diseases were also searched in journals manually. RESULTS: Fourteen of 1167 studies were eligible according to the inclusion criteria. All studies trained DL models based on radiographic images. Six studies applied to detection tasks, one focused on classification, two segmented lesions, and five studies made a combination of two types of DL models. The accuracy of the DL algorithms ranged from 75.7% to 99.7%, and the area under curves (AUC) varied between 0.7 and 0.997. CONCLUSION: DL can accurately deal with the tasks of diagnosing maxillary sinus diseases. Students, residents, and dentists could be assisted by DL algorithms to diagnose and make rational decisions on implant treatment related to maxillary sinuses.

Deep learning-based automatic segmentation of bone graft material after maxillary sinus augmentation.

OBJECTIVES: To investigate the accuracy and reliability of deep learning in automatic graft material segmentation after maxillary sinus augmentation (SA) from cone-beam computed tomography (CBCT) images. MATERIALS AND METHODS: One hundred paired CBCT scans (a preoperative scan and a postoperative scan) were collected and randomly allocated to training (n = 82) and testing (n = 18) subsets. The ground truths of graft materials were labeled by three observers together (two experienced surgeons and a computer engineer). A deep learning model including a 3D V-Net and a 3D Attention V-Net was developed. The overall performance of the model was assessed through the testing data set. The comparative accuracy and inference time consumption of the model-driven and manual segmentation (by two surgeons with 3 years of experience in dental implant surgery) were conducted on 10 CBCT scans from the test samples. RESULTS: The deep learning model had a Dice coefficient (Dice) of 90.36 ± 2.53%, a 95% Hausdorff distance (HD) of 1.59 ± 0.82 mm, and an average surface distance (ASD) of 0.38 ± 0.11 mm. The proposed model only needed 7.2 s, while the surgeon took 19.15 min on average to complete a segmentation task. The overall performances of the model were significantly superior to those of surgeons. CONCLUSIONS: The proposed deep learning model yielded a more accurate and efficient performance of automatic segmentation of graft material after SA than that of the two surgeons. The proposed model could facilitate a powerful system for volumetric change evaluation, dental implant planning, and digital dentistry.

Feasibility of a Robot-Assisted Surgical Navigation System for Mandibular Distraction Osteogenesis in Hemifacial Microsomia: A Model Experiment.

This study aimed to investigate the feasibility and accuracy of osteotomy and distractor placement using a robotic navigation system in a model surgical experiment of mandibular distraction osteogenesis for hemifacial microsomia. Imaging data from 5 patients with Pruzansky-Kaban type II (IIa: 4; IIb: 1) mandibular deformities were used to print 3D models for simulated mandibular distraction osteogenesis. In the experimental group, a robot-assisted surgical navigation system was used to perform the surgery under robotic guidance following registration, according to the preoperative design. Conventional surgery was performed in the control group, in which the operation was based on intraoperative estimations of the preoperative design by experienced surgeons. The accuracies of the osteotomy and distractor placement were assessed based on distance and angular error. Osteotomy accuracy was higher in the experimental group than in the control group, and the distance error ( t =9.311, P <0.001) and angular error ( t =5.385, P =0.001) were significantly reduced. The accuracy of distractor placement was also significantly higher in the experimental group, while the distance error ( t =3.048, P =0.016) and angular error ( t =3.524, P =0.024) were significantly reduced. The present results highlight the feasibility of robot-assisted distraction osteogenesis combined with electromagnetic navigation for improved surgical precision in clinical settings.

Three-dimensional Printing Technologies in Craniofacial Plastic Surgery: An Institutional Experience.

Rapid developments in 3-dimensional(3D) printing technologies in craniofacial plastic surgery have provided a new treatment modality for patients. In this article, we intend to share our institution's experience using 3D printing in 3 modes-namely, 3-dimensional printing for manufacturing contour models, guides, and implants. Fifty-nine patients were enrolled in our study between September 2009 and September 2021. Among the 3D printing-assisted technologies, 41 cases were used for congenital malformations, 82 for trauma repair, and 112 for cosmetic surgery. Preoperative design and postoperative data were compared and analyzed based on imaging data. In craniofacial plastic surgery, all patients had excellent postoperative objective bone measurements close to the preoperative design and improved esthetic appearance. Our survey of postoperative satisfaction showed that patients were quite satisfied with the surgery, especially concerning congenital deformities. Rapid prototyping 3-dimensional printing technology provides a practical and anatomically accurate means to produce patient-specific and disease-specific translational tools. These models can be used for surgical planning, simulation, and clinical evaluation. Expanding this technology in craniofacial plastic surgery will provide adequate assistance to practitioners and patients.

Radiomics and Artificial Intelligence Study of Masseter Muscle Segmentation in Patients With Hemifacial Microsomia.

BACKGROUND: Hemifacial microsomia (HFM) is one of the most common congenital craniofacial condition often accompanied by masseter muscle involvement. U-Net neural convolution network for masseter segmentation is expected to achieve an efficient evaluation of masseter muscle. METHODS: A database was established with 108 patients with HFM from June 2012 to June 2019 in our center. Demographic data, OMENS classification, and 1-mm layer thick 3-dimensional computed tomography were included. Two radiologists manually segmented masseter muscles in a consensus reading as the ground truth. A test set of 20 cases was duplicated into 2 groups: an experimental group with the intelligent algorithm and a control group with manual segmentation. The U-net follows the design of 3D RoI-Aware U-Net with overlapping window strategy and references to our previous study of masseter segmentation in a healthy population system. Sorensen dice-similarity coefficient (DSC) muscle volume, average surface distance, recall, and time were used to validate compared with the ground truth. RESULTS: The mean DSC value of 0.794±0.028 for the experiment group was compared with the manual segmentation (0.885±0.118) with α=0.05 and a noninferiority margin of 15%. In addition, higher DSC was reported in patients with milder mandible deformity ( r =0.824, P <0.05). Moreover, intelligent automatic segmentation takes only 6.4 seconds showing great efficiency. CONCLUSIONS: We first proposed a U-net neural convolutional network and achieved automatic segmentation of masseter muscles in patients with HFM. It is a great attempt at intelligent diagnosis and evaluation of craniofacial diseases.

Correlations Between Mandible Condylar Structures and External Ear Deformities in Hemifacial Microsomia With Three-Dimensional Analysis.

ABSTRACT: This study aimed to explore correlations between mandible and ear deformities and quantitative volumetric relations between condylar structures and external ear in hemifacial microsomia. The authors reconstructed three-dimensional craniofacial models from 212 patients with unilateral hemifacial microsomia (the unaffected side as the controls). Patients were evaluated by Pruzansky-Kaban and Marx classification, and divided into 3 age groups (0-6, 7-12, and >12 years of age). The mandible condylar structures, including condyle and the condylar skeletal unit, were selected (except the classification of the mandibular or ear deformities (M3)). Along with the external ear (except the classification of the mandibular or ear deformities (E4)), their volumes were measured and analyzed. Spearman correlation coefficient analysis was applied. There was a positive correlation between the mandible and ear deformities (r = 0.301, P   <  0.001). Either between the condyle and external ear ( P  = 0.071-0.493) or between the condylar unit and external ear ( P  = 0.080 - 0.488), there were no volumetric relations on the affected side, whereas on the unaffected side were (r = 0.492-0.929 for condyle, r = 0.443-0.929 for the condylar unit, P  < 0.05). In most cases, the condylar structures of the classification of the mandibular or ear deformities (M2b) were significantly smaller than the classification of the mandibular or ear deformities (M2a). Results suggested deformities of mandibular condylar structures and ear did not correlate, although deformities of mandible and ear did. The condylar deformity might develop independently from microtia and be more severe within relatively more abnormal temporomandibular joints.

OMENS+ Classification Correlations Analysis of Craniofacial Microsomia in China: The Relationship Between Macrostomia and Mandibular Hypoplasia.

ABSTRACT: Macrostomia is arare congenital craniofacial deformity that influences the appearance and function of patients. In most cases, it coexists with craniomaxillofacial deformities such as craniofacial microsomia (CFM). This study aimed to analyze the relationship between macrostomia and mandibular hypoplasia so as to facilitate the early detection and diagnosis of children with CFM. It included 236 patients diagnosed with CFM. All underwent facial expression analysis, multi-angle photography, computed tomography, and three-dimensional reconstruction of soft and hard tissues. The clinical classification was performed according to OMENS+. Spearman (rank) correlation analysis was used to analyze the relationship between the severity of macrostomia (C1 and C2) and the degree of mandibular involvement (M1, M2a, M2b, and M3), and the correlation among the components of OMENS+. Of the 80 cases of macrostomia (34%) reported, 72 cases (90%) were C1 and 8 (10%) were C2. The analysis of OMENS+ revealed significant correlations among OMENS+ components. Also, a high correlation was observed between macrostomia (C) and hypoplasia of the mandible (M) ( P  = 0.002). Macrostomia was closely related to mandibular hypoplasia among children diagnosed with CFM. These results suggested that patients with macrostomia, who might also have craniofacial malformations caused by other first branchial arch anomalies, should be comprehensively physically examined for other syndromes.

Hyperbaric CO2 Cryotherapy for Managing Swelling After Mandibular Angle Ostectomy.

ABSTRACT: Postoperative facial swelling after mandibular angle ostectomy is a concern for patients as it affects their quality of life. This study aimed to evaluate the effect of hyperbaric CO2 cryotherapy in relieving postoperative swelling. Thirty-seven patients (mean age: 22.95 ±â€Š3.49 years) who underwent bilateral mandibular angle ostectomy from April to October 2019, were included in this study. A split-mouth design was adopted and through a random sequence, either side of the face was designated as the experimental side. Both received routine nursing care, while the experimental side underwent an additional hyperbaric CO2 cryotherapy treatment. The facial structure was recorded by a 3-dimensional laser scanner pre- and postoperatively. Geomagic Studio was used for alignment, visualization, and quantification of the swelling. The largest deviation value on each side was adopted to assess the overall swelling. Parameters were compared using the paired t-test, and P<0.05 was considered statistically significant. No necrosis of the skin and adjacent structures or other complications occurred in these patients. After the first day of cryotherapy, the deviation in the experimental and the control groups was (8.40 ±â€Š1.95) mm and (10.42 ±â€Š2.03) mm, respectively. The next day, after cryotherapy, the value further reduced to (5.42 ±â€Š1.36) mm and (8.24 ±â€Š2.22) mm for the experimental and control groups, respectively. And the effect remained till the seventh day. No difference was observed in terms of volume of drainage. Hyperbaric CO2 cryotherapy is safe and effective in relieving postoperative swelling and seems to be more effective than the traditional cold-pack treatment after mandibular angle ostectomy.

A Simple Scoring System for Quick, Accurate, and Reliable Early Diagnosis of Hand, Foot, and Mouth Disease.

BACKGROUND To construct an accurate, reliable, and simple scoring system of improving HFMD diagnosis. MATERIAL AND METHODS Based on the following 3 steps, a simple scoring diagnostic system was built: (1) we selected basic markers (age and sex), markers recommended in HFMD diagnosis guidelines, and significant biomarkers among severity groups found in a large dataset; (2) we used positive constituent ratio for determining scores of each marker; and (3) we applied receiving operating curve in an external dataset to determine the optimal cut-off score. RESULTS The selected markers were sex, age, fever, skin rashes, nervous system disorder, respiratory system disorder, digestive system disorder and cardiopulmonary complications, C-reactive-protein, White Blood Cell, Creatinine Kinase, Creatinine Kinase Isoenzyme, Gamma-Glutamyl Transpeptidase, Albumin, Globulin, Albumin/Globulin Ratio, Natrium, Chloride, Calcium, and Glucose. A simple scoring system with 3.9684 as the lower cut-off was constructed. The AUC was 0.918 (95% CI: 0.874-0.963, P<0.01). The sensitivity, specificity, and Youden Index, which were based on the validation dataset of 200 subjects (80 cases, 120 non-cases with skin rashes or fever), were 0.95, 0.90, and 0.85, respectively. CONCLUSIONS This simple scoring system is an effective method to diagnose HFMD.

Comparison of Ultrasound-Guided Percutaneous Polidocanol Injection Versus Percutaneous Ethanol Injection for Treatment of Benign Cystic Thyroid Nodules.

OBJECTIVES: We compared the efficacy, safety, and cost-effectiveness of ultrasound-guided percutaneous polidocanol injection and percutaneous ethanol injection for the treatment of benign cystic and predominantly cystic thyroid nodules. METHODS: A total of 135 cystic thyroid nodules treated by percutaneous ethanol injection and 136 cystic thyroid nodules treated by percutaneous polidocanol injection were enrolled retrospectively in this study from May 2010 to March 2016. The nodules were followed after 1, 3, 6, and 12 months. Nodule volumes, symptoms scores, and cosmetic scores were assessed before treatment and at follow-up. The therapeutic success rate, safety, and cost-effectiveness between the groups were also compared. RESULTS: No significant differences in the reduction of the nodule volume, volume reduction rate, and therapeutic success were observed between the groups with cystic and predominantly cystic thyroid nodules during follow-up (P > .05). Neither the cosmetic scores (P = .59; P = .42) nor the symptom scores (P = .32; P = .73) in the cystic and predominantly cystic nodules were significantly different between the groups at the last follow-up. The complication rates for ethanol were higher than those for polidocanol (P < .05). However, the cost of polidocanol injection was higher than that of ethanol injection for cystic thyroid nodules (mean ± SD, US$97.18 ± US$22.17 versus US$43.36 ± US$5.51; P < .01). CONCLUSIONS: Ultrasound-guided percutaneous polidocanol injection can be an alternative for sclerotherapy of cystic or predominantly cystic thyroid nodules. However, its cost was higher than that of percutaneous ethanol injection.

Application of Real-Time Surgical Navigation for Zygomatic Implant Insertion in Patients With Severely Atrophic Maxilla.

PURPOSE: Computer-aided treatment technology has extended its applications to oral implantology. This report describes the authors' initial clinical experience on the application of a commercially available navigation system (VectorVision) in zygomatic implant (ZI) insertion in the severely atrophic maxilla. MATERIALS AND METHODS: This was a retrospective longitudinal study. Eligible patients with maxillary edentulism who were treated with ZI placement were enrolled. Treatment planning was performed on the computer based on previously obtained 3-dimensional imaging data. The surgical procedure was carried out under the guidance of a surgical navigation system. The outcome variable was safety and additional variables were ZI survival rate and radiologic bone-to-implant contact (rBIC) area in the zygoma. Statistical analysis was performed with SPSS 16.0 for Windows (SPSS, Inc, Chicago, IL). RESULTS: Fifteen patients (8 men, 7 women; age range, 30 to 69 yr; average age, 43 ± 3.5 yr) were eligible for the study and were enrolled from May 2015 through September 2016. Of the included patients, each of 4 patients received 1 ZI on each side of the zygomatic bone and 2 to 4 standard implants in the edentulous anterior maxilla; the other 11 received a ZI "quad approach" without standard implant insertion. All ZIs were anchored in the site of the maxillary alveolar process and zygomatic bone, and no critical anatomic structure injuries occurred during insertion and postoperative radiographic examination. All ZIs achieved osseointegration, for an overall survival rate of 100% after early healing. The overall rBIC area of ZIs in the study was 4.1 to 24.7 mm (average, 14.5 ± 4.6 mm). CONCLUSION: For the limited clinical cases treated in this study, the procedure for ZI placement was feasible and reliable with the guidance of the surgical navigation system. In addition, the potential risk of complications was minimized and ZIs were placed to make the best possible use of the available bone volume.

Comparison of the accuracy of dental implant placement using dynamic and augmented reality-based dynamic navigation: An in vitro study.

Background/purpose: Augmented reality has been gradually applied in dental implant surgery. However, whether the dynamic navigation system integrated with augmented reality technology will further improve the accuracy is still unknown. The purpose of this study is to investigate the accuracy of dental implant placement using dynamic navigation and augmented reality-based dynamic navigation systems. Materials and methods: Thirty-two cone-beam CT (CBCT) scans from clinical patients were collected and used to generate 64 phantoms that were allocated to the augmented reality-based dynamic navigation (ARDN) group or the conventional dynamic navigation (DN) group. The primary outcomes were global coronal, apical and angular deviations, and they were measured after image fusion. A linear mixed model with a random intercept was used. A P value < 0.05 was considered to indicate statistical significance. Results: A total of 242 dental implants were placed in two groups. The global coronal, apical and angular deviations of the ARDN and DN groups were 1.31 ± 0.67 mm vs. 1.18 ± 0.59 mm, 1.36 ± 0.67 mm vs. 1.39 ± 0.55 mm, and 3.72 ± 2.13° vs. 3.1 ± 1.56°, respectively. No significant differences were found with regard to coronal and apical deviations (P = 0.16 and 0.6, respectively), but the DN group had a significantly lower angular deviation than the ARDN group (P = 0.02). Conclusion: The augmented reality-based dynamic navigation system yielded a similar accuracy to the conventional dynamic navigation system for dental implant placement in coronal and apical points, but the augmented reality-based dynamic navigation system yielded a higher angular deviation.

A hybrid robotic system for zygomatic implant placement based on mixed reality navigation.

BACKGROUNDS: Zygomatic implant (ZI) placement surgery is a viable surgical option for patients with severe maxillary atrophy and insufficient residual maxillary bone. Still, it is difficult and risky due to the long path of ZI placement and the narrow field of vision. Dynamic navigation is a superior solution, but it presents challenges such as requiring operators to have advanced skills and experience. Moreover, the precision and stability of manual implantation remain inadequate. These issues are anticipated to be addressed by implementing robot-assisted surgery and achieved by introducing a mixed reality (MR) navigation-guided hybrid robotic system for ZI placement surgery. METHODS: This study utilized a hybrid robotic system to perform the ZI placement surgery. Our first step was to reconstruct a virtual 3D model from preoperative cone-beam CT (CBCT) images. We proposed a series of algorithms based on coordinate transformation, which includes image-phantom registration, HoloLens-tracker registration, drill-phantom calibration, and robot-implant calibration, to unify all objects within the same coordinate system. These algorithms enable real-time tracking of the surgical drill's position and orientation relative to the patient phantom. Subsequently, the surgical drill is directed to the entry position, and the planned implantation paths are superimposed on the patient phantom using HoloLens 2 for visualization. Finally, the hybrid robot system performs the processed of drilling, expansion, and placement of ZIs under the guidance of the MR navigation system. RESULTS: Phantom experiments of ZI placement were conducted using 10 patient phantoms, with a total of 40 ZIs inserted. Out of these, 20 were manually implanted, and the remaining 20 were robotically implanted. Comparisons between the actual implanted ZI paths and the preoperatively planned ZI paths showed that our MR navigation-guided hybrid robotic system achieved a coronal deviation of 0.887 ± 0.213 mm, an apical deviation of 1.201 ± 0.318 mm, and an angular deviation of 3.468 ± 0.339° This demonstrates significantly better accuracy and stability than manual implantation. CONCLUSION: Our proposed hybrid robotic system enables automated ZI placement surgery guided by MR navigation, achieving greater accuracy and stability compared to manual operations in phantom experiments. Furthermore, this system is expected to apply to animal and cadaveric experiments, to get a good ready for clinical studies.

Changes of masseter muscle after mandible distraction osteogenesis in patients with Hemifacial microsomia: a retrospective study.

Introduction: Mandible distraction osteogenesis (MDO) is widely used to reconstruct the mandible in patients with mild Hemifacial microsomia (HFM). However, the masseter's response to mandible distraction remains unclear. Methods: In this study, we analyze the effect of the surgical intervention on masseter muscle by a retrospective analysis. The procedure consisted of a five-day latent period, a three-week distraction period, and a six-month consolidation period. CT data were manually segmented and measured with Mimics software before surgery, within 3 months, and more than 1 year postoperatively. Masseter volume, masseter length, masseter width, and mandible ramus height were measured and analyzed using paired t-test, Pearson, and Spearman correlation analysis. Results: We included 21 patients with HFM who underwent mandible distraction osteogenesis from 2015 to 2020. The masseter volume on the affected side increased immediately after surgery from (6,505.33 ± 3,671.95) mm3 to (10,194.60 ± 5638.79) mm3, but decreased to (8,148.38 ± 3,472.57) mm3 at the second follow-up correlated to mandible ramus height (r = 0.395, P = 0.038). A similar trend was observed in changes in masseter length. Symmetry and width of masseter muscle had no longitudinal statistical significance. Discussion: Masseter muscle involvement benefits from MDO in the short term. To achieve long-term efficacy, more attention should be paid to muscle reconstruction.

Microplastics derived from plastic mulch films and their carrier function effect on the environmental risk of pesticides.

Plastic film mulching can maintain soil water and heat conditions, promote plant growth and thus generate considerable economic benefits in agriculture. However, as they age, these plastics degrade and form microplastics (MPs). Additionally, pesticides are widely utilized to control organisms that harm plants, and they can ultimately enter and remain in the environment after use. Pesticides can also be sorbed by MPs, and the sorption kinetics and isotherms explain the three stages of pesticide sorption: rapid sorption, slow sorption and sorption equilibrium. In this process, hydrophobic and partition interactions, electrostatic interactions and valence bond interactions are the main sorption mechanisms. Additionally, small MPs, biodegradable MPs and aged conventional MPs often exhibit stronger pesticide sorption capacity. As environmental conditions change, especially in simulated biological media, pesticides can desorb from MPs. The utilization of pesticides by environmental microorganisms is the main factor controlling the degradation rate of pesticides in the presence of MPs. Pesticide sorption by MPs and size effects of MPs on pesticides are related to the internal exposure level of biological pesticides and changes in pesticide toxicity in the presence of MPs. Most studies have suggested that MPs exacerbate the toxicological effects of pesticides on sentinel species. Hence, the environmental risks of pesticides are altered by MPs and the carrier function of MPs. Based on this, research on the affinity between MPs and various pesticides should be systematically conducted. During agricultural production, pesticides should be cautiously selected and used plastic film to ensure human health and ecological security.

The combined effects of azoxystrobin and different aged polyethylene microplastics on earthworms (Eisenia fetida): A systematic evaluation based on oxidative damage and intestinal function.

Pesticides and microplastics are common pollutants in soil environments, adversely affecting soil organisms. However, the combined toxicological effects of aged microplastics and pesticides on soil organisms are still unclear. In this study, we systematically studied the toxicological effects of azoxystrobin and four different aged polyethylene (PE) microplastics on earthworms (Eisenia fetida). The purpose was to evaluate the effects of aging microplastics on the toxicity of microplastics-pesticides combinations on earthworms. The results showed that different-aged PE microplastics promoted azoxystrobin accumulation in earthworms. Meanwhile, combined exposure to azoxystrobin and aged PE microplastics decreased the body weight of earthworms. Besides, both single and combined exposure to azoxystrobin and aged PE microplastics could lead to oxidative damage in earthworms. Further studies revealed that azoxystrobin and aged PE microplastics damage the intestinal structure and function of earthworms. Additionally, the combination of different aged PE microplastics and azoxystrobin was more toxic on earthworms than single exposures. The PE microplastics subjected to mechanical wear, ultraviolet radiation, and acid aging exhibited the strongest toxicity enhancement effects on earthworms. This high toxicity may be related to the modification of PE microplastics caused by aging. In summary, these results demonstrated the enhancing effects of aged PE microplastics on the toxicity of pesticides to earthworms. More importantly, aged PE microplastics exhibited stronger toxicity-enhancing effects in the early exposure stages. This study provides important data supporting the impact of different aged PE microplastics on the environmental risks of pesticides.

OrthoCalc: The six degrees of freedom measurement workflow of rotational and displacement changes for maxilla positioning evaluation.

BACKGROUND: This study is undertaken to establish the accuracy and reliability of OrthoCalc, a 3D application designed for the evaluation of maxillary positioning. METHODS: We registered target virtual planned models, maxillary models from pre-operative and post-operative CT scans, and post-operative intra-oral scans to a common reference system, allowing for digital evaluation. To assess rotational changes, we introduced a novel measurement method based on virtual cuboid models. Displacement errors were calculated based on proposed registration matrices. We also compared OrthoCalc to established commercial medical software as a benchmark. RESULTS: Statistical significance calculated showed no significant differences between OrthoCalc and commercial software. the biggest error of 0.04 degree in rotation change was found in the yaw. A maximum displacement change of 0.75 mm was found in the X direction. CONCLUSIONS: Our study validates OrthoCalc as a precise and reliable tool for assessing maxillary position changes with six degrees of freedom in orthognathic surgery, endorsing its clinical utility.

Morphological and quantitative study of the inferior alveolar nerve canal in hemifacial microsomia.

This study aimed to probe into the anatomic course of inferior alveolar nerve canal (IANC) in hemifacial microsomia (HFM) on a large scale, morphological observations and further quantitative study were performed. Patients were classified by Pruzansky-Kaban classification. The anatomic course of IANC was analyzed morphologically with three-dimensional (3D) imaging software among 248 patients. Seven distances between fixed landmarks on both sides were measured for 236 patients. The differences between affected and unaffected sides were compared. Significant differences were found in the entrance (P < 0.001), route (P < 0.001), and exit (P < 0.05) of IANC in type IIb and III HFM. The higher the degree of mandibular deformity was, the higher the incidence of IANC variation was (P < 0.05). The distances in the horizontal aspect of IANC including from mandibular foramen to mental foramen (P < 0.05) and from mental foramen to gonion (P < 0.05) were significantly shorter on the affected side. Abnormalities of the anatomical course of IANC exist in patients with Pruzansky-Kaban type IIb and type III HFM. The reduction of IANC on the affected side in the horizontal distance is more obvious. Three-dimensional imaging assessment is recommended before surgery.

Efficacy of navigation system-assisted distraction osteogenesis for hemifacial microsomia based on artificial intelligence for 3 to 18 years old: study protocol for a randomized controlled single-blind trial.

BACKGROUND: Mandibular distraction osteogenesis (MDO) is a major part of the treatment for hemifacial microsomia patients. Due to the narrow surgical field of the intraoral approach, osteotomy accuracy is highly dependent on the surgeons' experience. Electromagnetic (EM) tracking systems can achieve satisfying accuracy to provide helpful real-time surgical navigation. Our research team developed an EM navigation system based on artificial intelligence, which has been justified in improving the accuracy of osteotomy in the MDO in animal experiments. This study aims to clarify the effect of the EM navigation system in improving the MDO accuracy for hemifacial microsomia patients. METHODS: This study is designed as a single-centered and randomized controlled trial. Altogether, 22 hemifacial microsomia patients are randomly assigned to the experiment and control groups. All patients receive three-dimensional CT scans and preoperative surgical plans. The EM navigation system will be set up for those in the experiment group, and the control group will undergo traditional surgery. The primary outcome is the surgical precision by comparing the osteotomy position of pre- and postoperative CT scan images through the Geomagic Control software. The secondary outcomes include mandibular symmetry (occlusal plane deviation angle, mandibular ramus height, and body length), pain scale, and complications. Other indications, such as the adverse events of the system and the satisfactory score from patients and their families, will be recorded. DISCUSSION: This small sample randomized controlled trial intends to explore the application of an EM navigation system in MDO for patients, which has been adopted in other surgeries such as orthognathic procedures. Because of the delicate structures of children and the narrow surgical view, accurate osteotomy and protection of nearby tissue from injury are essential for successful treatment. The EM navigation system based on artificial intelligence adopted in this trial is hypothesized to provide precise real-time navigation for surgeons and optimally improve patient outcomes, including function and aesthetic results. The results of this trial will extend the application of new navigation technology in pediatric plastic surgery. TRIAL REGISTRATION: Chinese Clinical Trial Registry ChiCTR2200061565. Registered on 29 June 2022.

Automatic 3D Teeth Reconstruction from Five Intra-oral Photos Using Parametric Teeth Model.

Orthodontic treatment is a lengthy process that requires regular in-person dental monitoring, making remote dental monitoring a viable alternative when face-to-face consultation is not possible. In this study, we propose an improved 3D teeth reconstruction framework that automatically restores the shape, arrangement, and dental occlusion of upper and lower teeth from five intra-oral photographs to aid orthodontists in visualizing the condition of patients in virtual consultations. The framework comprises a parametric model that leverages statistical shape modeling to describe the shape and arrangement of teeth, a modified U-net that extracts teeth contours from intra-oral images, and an iterative process that alternates between finding point correspondences and optimizing a compound loss function to fit the parametric teeth model to predicted teeth contours. We perform a five-fold cross-validation on a dataset of 95 orthodontic cases and report an average Chamfer distance of 1.0121 mm2 and an average Dice similarity coefficient of 0.7672 on all the test samples in the cross-validation, demonstrating a significant improvement compared with the previous work. Our teeth reconstruction framework provides a feasible solution for visualizing 3D teeth models in remote orthodontic consultations.