Objective To evaluate the accuracy of different convolutional neural networks (CNN),representative deep learning models,in the differential diagnosis of ameloblastoma and odontogenic keratocyst,and subsequently compare the diagnosis results between models and oral radiologists. Methods A total of 1000 digital panoramic radiographs were retrospectively collected from the patients with ameloblastoma (500 radiographs) or odontogenic keratocyst (500 radiographs) in the Department of Oral and Maxillofacial Radiology,Peking University School of Stomatology.Eight CNN including ResNet (18,50,101),VGG (16,19),and EfficientNet (b1,b3,b5) were selected to distinguish ameloblastoma from odontogenic keratocyst.Transfer learning was employed to train 800 panoramic radiographs in the training set through 5-fold cross validation,and 200 panoramic radiographs in the test set were used for differential diagnosis.Chi square test was performed for comparing the performance among different CNN.Furthermore,7 oral radiologists (including 2 seniors and 5 juniors) made a diagnosis on the 200 panoramic radiographs in the test set,and the diagnosis results were compared between CNN and oral radiologists. Results The eight neural network models showed the diagnostic accuracy ranging from 82.50% to 87.50%,of which EfficientNet b1 had the highest accuracy of 87.50%.There was no significant difference in the diagnostic accuracy among the CNN models (P=0.998,P=0.905).The average diagnostic accuracy of oral radiologists was (70.30±5.48)%,and there was no statistical difference in the accuracy between senior and junior oral radiologists (P=0.883).The diagnostic accuracy of CNN models was higher than that of oral radiologists (P<0.001). Conclusion Deep learning CNN can realize accurate differential diagnosis between ameloblastoma and odontogenic keratocyst with panoramic radiographs,with higher diagnostic accuracy than oral radiologists.
Humans ; Ameloblastoma/diagnostic imaging* ; Deep Learning ; Diagnosis, Differential ; Radiography, Panoramic ; Retrospective Studies ; Odontogenic Cysts/diagnostic imaging* ; Odontogenic Tumors

Objective:To investigate the feasibility and clinical effect of the surgical approach and method of transnasal fenestration under nasal endoscope for the treatment of maxillary odontogenic cyst. Methods:The clinical data of 23 cases with maxillary odontogenic cysts treated by nasal endoscopy through nasal fenestration were retrospectively analyzed. All cases underwent nasal endoscopy and CT examination before the operation. The mucosal membrane of the parietal wall of the cyst was excised through fenestration of the nasal base. The cyst fluid was removed by decompression, and the bony opening of the nasal base was trimmed and enlarged to the edge of the cyst. The intraoperative and postoperative effects were observed. Results:All cases were well exposed under the direct vision of nasal endoscope. The top wall of the cyst was removed to maximize the communication between the cyst cavity and the nasal floor. There were no complications such as nasolacrimal duct injury, turbinate atrophy, necrosis, and facial numbness. All patients were followed up for 6-12 months, and the clinical symptoms gradually disappeared after surgery. The inferior turbinate was in good shape, the cyst cavity was smooth, the cyst wall was determined, and no cyst recurrence was observed. Conclusion:The treatment of odontogenic cyst of maxilla under nasal endoscope through nasal fenestration is convenient. It has less trauma, fewer complications and a satisfactory curative effect, which is worthy of clinical promotion.
Humans ; Maxilla ; Retrospective Studies ; Odontogenic Cysts/surgery* ; Endoscopy ; Turbinates/surgery* ; Endoscopes

Objective: To detect the SMO mutations in odontogenic keratocyst (OKC) and to explore the mechanism behind. Methods: Patients with OKC who received treatment in the Department of Oral and Maxillofacial Surgery, School and Hospital of Stomatology,Peking University, from September 2012 to June 2017 were enrolled. OKC samples from 10 patients diagnosed as naevoid basal cell carcinoma syndrome (NBCCS)-related OKC (4 females and 6 males) and 20 patients diagnosed as sporadic OKC (7 females and 13 males) were collected. Genomic DNAs were extracted from fibrous capsules and epithelial lining respectively. SMO mutations were detected and analyzed by Sanger sequencing. Results: Three SMO mutations were found in one NBCCS-associated OKC who carrying c.2081C>G (p.P694R) mutation) and two sporadic OKC who carrying c.907C>T (p.L303F) mutation and c.1247_1248delinsAA (p.G416E), respectively), among which the first two mutations were novel mutations that had not been reported before. Besides, two mutations in sporadic OKC were not paired with PTCH1 mutations. Conclusions: In addition to PTCH1 gene mutations, SMO gene mutations also exist in OKC which might be related to the development of OKC.
Basal Cell Nevus Syndrome/genetics* ; Female ; Humans ; Male ; Mutation ; Odontogenic Cysts/genetics* ; Odontogenic Tumors/genetics* ; Smoothened Receptor/genetics*

Peripheral odontogenic keratocysts are rarely observed, and cases of odontogenic keratocysts of buccal soft tissues are even rarer. This study was performed to present two rare cases of odontogenic keratocysts in buccal soft tissues and review related literature.
Humans ; Odontogenic Cysts/diagnosis* ; Odontogenic Tumors

Among the ghost cell lesions, Dentinogenic Ghost Cell Tumors (DGCT) are among the rarest. We report a case of a 45-year-old Filipino man, who presented with a right mandibular mass. Microscopic examination showed a solid neoplasm composed of islands of odontogenic epithelium with areas showing aberrant keratinization forming ghost cells and dentinoid material. We also discuss the pertinent differential diagnosis of ghost cell-containing odontogenic tumors. We report this case due to its rarity, its morphological resemblance to ameloblastoma, and its potential for malignant transformation.
Odontogenic Cysts ; Odontogenic Tumors

OBJECTIVE: To investigate the clinicopathologic features and prognostic factors in odontogenic keratocyst (OKC), and to provide new reference for clinic treatment and management of these patients. METHODS: Clinicopathological data of 844 cases initially diagnosed as or associated with OKC at Department of Oral Pathology, Peking University Hospital of Stomatology from 2000 to 2018 were collected. The cases were divided into 4 groups: sporadic OKCs (intraosseous, cystic lesion irrelevant to nevoid basal cell carcinoma syndrome), syndromic OKCs, solid OKCs and peripheral OKCs. The patients were follow-up for 6 to 216 months and the factors that might relate to recurrence were analyzed. RESULTS: There were 805 cases (95.4%) of sporadic OKCs, 32 cases (3.8%) of syndromic OKCs, 3 cases of solid OKCs and 4 cases of peripheral OKCs. The main age of sporadic OKCs was 36.03 years with the peak at the second and third decades. Ratio of male and female was 1.27:1. The predilection site was the molar and ramus area of mandibular (56.2%). In the study, 428 cases (71.2%) were unilocular in radiography while 28.8% were multilocular. The recurrent rate of enucleation with the follow-up was 20.1% (118/588) while most of them occurred in 1-3 years after surgery. The recurrent rate of multilocular patients (39.0%) was significantly higher than that of the unilocular. Enucleation after marsupialization (43 cases) or enucleation only (545 cases) showed no difference in recurrence (P>0.05). The syndromic OKCs was younger (main 20.97) and preferred to be multiple compared with sporadic OKCs (30/32, 93.7%). The predilection site was also molar and ramus area of mandibular (41.7%). Age and gender distribution of multiple cases had no significant difference with those in sporadic OKCs. More daughter cysts and epithelial islands were seen (56.3% and 17.9%). Furthermore, the recurrent rate was significantly higher than that of the sporadic OKCs (13/29, 44.9%). But there was no evidence of recurrent-related factors. The age of solid and peripheral OKCs, averaged at 45.00 and 65.75 years, were older than others. Four of peripheral OKCs showed no recurrence after enucleation. CONCLUSION The recurrence rate of sporadic OKCs after enucleation is 20.1%. The multilocular lesions prefer to be recurrent. There is no significant difference of recurrence with enucleation only or enucleation after marsupialization. Compared with sporadic OKCs, the syndromic patients are younger and easier to be multiple. It tends to be recurrent frequently and rapidly. There are no related factors about recurrence of syndromic patients. The clinicians should considerate comprehensively and make an individual management of therapy and follow-up. Solid and peripheral OKCs are rare and older.
Female ; Humans ; Male ; Mandible ; Neoplasm Recurrence, Local ; Odontogenic Cysts ; Odontogenic Tumors ; Recurrence ; Retrospective Studies

A 70-year-old man consulted for a mass of unspecified duration in the anterior area of the mandible. Radiologic information was not provided. An incision biopsy was performed by the dentistry service. The specimen received at the laboratory was labeled “cystic lining” and consisted of two light-gray to dark brown, irregularly shaped tissue fragments measuring 0.3 cm and 0.4 cm in diameters.
Odontogenic Cysts

Odontogenic keratocysts (OKCs) are common cystic lesions of odontogenic epithelial origin that can occur sporadically or in association with naevoid basal cell carcinoma syndrome (NBCCS). OKCs are locally aggressive, cause marked destruction of the jaw bones and have a propensity to recur. PTCH1 mutations (at ∼80%) are frequently detected in the epithelia of both NBCCS-related and sporadic OKCs, suggesting that PTCH1 inactivation might constitutively activate sonic hedgehog (SHH) signalling and play a major role in disease pathogenesis. Thus, small molecule inhibitors of SHH signalling might represent a new treatment strategy for OKCs. However, studies on the molecular mechanisms associated with OKCs have been hampered by limited epithelial cell yields during OKC explant culture. Here, we constructed an isogenic PTCH1 cellular model of PTCH1 inactivation by introducing a heterozygous mutation, namely, c.403C>T (p.R135X), which has been identified in OKC patients, into a human embryonic stem cell line using the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated 9 (Cas9) system. This was followed by the induction of epithelial differentiation. Using this in vitro isogenic cellular model, we verified that the PTCH1 heterozygous mutation causes ligand-independent activation of SHH signalling due to PTCH1 haploinsufficiency. This activation was found to be downregulated in a dose-dependent manner by the SHH pathway inhibitor GDC-0449. In addition, through inhibition of activated SHH signalling, the enhanced proliferation observed in these induced cells was suppressed, suggesting that GDC-0449 might represent an effective inhibitor of the SHH pathway for use during OKC treatment.
Anilides ; pharmacology ; Basal Cell Nevus Syndrome ; Hedgehog Proteins ; genetics ; pharmacology ; Humans ; Molecular Targeted Therapy ; Odontogenic Cysts ; genetics ; physiopathology ; therapy ; Odontogenic Tumors ; genetics ; physiopathology ; therapy ; Pyridines ; pharmacology

PURPOSE: The prevalence of supernumerary teeth has been reported to be between 0.1% and 3.8%. The aim of this study was to determine the prevalence, clinical significance, and associated pathologies of fourth molars based on a retrospective study and a literature review. MATERIALS AND METHODS: A 5-year retrospective prevalence study was conducted at the Department of Oral Diagnosis and Dentomaxillofacial Radiology of Okayama University Hospital, Okayama, Japan. The study involved extracting data from the digital records of patients from January 1, 2013 through December 31, 2017. The sampling frame included all patients who had panoramic radiographs, cone-beam computed tomography (CT), and multislice CT images during the period under review. RESULTS: A total of 26,721 cases were reviewed and 87 fourth molars were identified. The prevalence of fourth molars in the 5-year study at Okayama was calculated as 0.32%. The mean age of patients with a fourth molar was 30.43 years, and the male-to-female ratio was 1:0.98. The vast majority of cases were in the maxilla (92%) and had normal shapes (89.7%); furthermore, 82.8% of cases were unerupted. CONCLUSION: The prevalence of fourth molars in the study population was found to be 0.32%, and fourth molars occurred with approximately equal frequency in males and females. Fourth molars were more common in the maxilla and were predominantly unerupted and small.
Cone-Beam Computed Tomography ; Cross-Sectional Studies ; Diagnosis, Oral ; Female ; Humans ; Japan ; Male ; Maxilla ; Odontogenic Cysts ; Odontoma ; Pathology ; Prevalence ; Retrospective Studies ; Tooth, Supernumerary

OBJECTIVES: This study was aimed to analyze the reducing pattern of decompression on mandibular odontogenic keratocyst and to determine the proper time for secondary enucleation. MATERIALS AND METHODS: Seventeen patients with OKC of the mandible were treated by decompression. Forty-five series of CT data were taken during decompression and measured by using InVivo software (Anatomage, San Jose, Calif) and were analyzed. RESULTS: The expected relative volume during decompression is calculated using the following formula: V(t) = V initial × exp.(at + 1/2bt 2) (t = duration after decompression (day)). There was no significant directional indicator in the rate of reduction between buccolingual and mesiodistal widths. CONCLUSION: The volume reduction rate gradually decreased, and 270 days were required for 50% volume reduction following decompression of OKC. The surgeon should be aware of this pattern to determine the timing for definitive enucleation. CLINICAL RELEVANCE: The volume reduction rate and pattern of decompression of the OKC can be predicted and clinicians should be considered when treating OKC via decompression.
Decompression ; Humans ; Mandible ; Odontogenic Cysts