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

An uncommon case of a patient with recurrent mandibular ameloblastoma involving various adjacent soft tissues is presented with plain film radiography, computed tomography and magnetic resonance imaging(MRI) and bone scan. The tumor involved mandible and eroded several bones including foramen ovale. This case shows that although an ameloblastoma primarily affect mandible, exact radiographic evaluation of tumor extent will assist the surgeon in the progress of rational approaches to the management of ameloblastoma.
Ameloblastoma* ; Cytochrome P-450 CYP1A1 ; Diagnostic Imaging* ; Foramen Ovale ; Humans ; Mandible ; Radiography ; Recurrence ; Skull Base

Maxillofacial images must be examined to find pathologies not identified during clinical examination. Unicystic ameloblastoma (UA) extending to the mandibular body and ramus was neglected on initial panoramic radiographic examination. After orthodontic therapy, a huge lesion was observed clinically and through imaging exams. After the conservative surgery, no recurrence was observed during five years of follow-up. This case emphasized the need for careful evaluation of patient images focusing on the oral diagnosis before any dental treatment planning, including orthodontic therapy.
Ameloblastoma* ; Delayed Diagnosis* ; Dentistry ; Diagnosis, Oral ; Diagnostic Imaging ; Follow-Up Studies ; Humans ; Mandible* ; Pathology ; Recurrence

Presented here is a rare occurrence of ameloblastic carcinoma located in the maxilla of a pediatric patient. The patient's chief complaint is a large mass protruding from the right maxillary area diagnosed as ameloblastic carcinoma before consulting the nuclear medicine section for whole-body bone scintigraphy. Whole-body bone scintigraphy was performed with cranial SPECT imaging. The maxillary tumor appeared to have decreased radiotracer uptake with intense activity in the anterior and anteromedial aspects of the tumor. There was also a focus of uptake in the lateral border of the tumor, radiating to the angle of the right mandible. Mandibular X-ray series was also acquired, with findings that correlate with those of the bone scintigraphy.

Human ; Male ; Child ; Child ; Mandible ; Maxilla ; Maxillary Neoplasms ; Nuclear Medicine ; Tomography, Emission-computed, Single-photon ; Whole Body Imaging ; X-rays ; Ameloblastoma ; Neoplasms ; Diagnostic Imaging

Ameloblastoma is a benign odontogenic tumor with an aggressive biological behavior, and the surgical treatment frequently results in failure for the postoperative recurrence. The aim of this article was to investigate whether the proliferative ability and prognosis of ameloblastoma could be evaluated by the radiographic boundary. The ameloblastoma cases treated by the conservative therapy in our hospital between 1981 and 2001 were divided into three groups based on the nature of the radiographic borders of the lesions. The biologic behavior was evaluated by Ki-67 antibody immunohistochemically. Comparisons of prognosis and Ki-67 expression were carried out by statistic methods. There were 24 cases of well-defined edge with sclerosis (group I), 41 cases of well-defined edge without sclerosis (group II) and 32 cases of ill-defined edge (group III). The recurrent rates were 29.2% in group I, 43.9% in group II and 62.5% in group III (P<0.05). The cells in group III expressed the highest Ki-67 level (P<0.05). The radiographic boundary could be used as one of indicators in evaluating the proliferative ability of ameloblastoma and the patient's prognosis, which was consistent with Ki-67 expression.
Adult ; Ameloblastoma ; diagnostic imaging ; pathology ; surgery ; Cell Proliferation ; Chi-Square Distribution ; Female ; Humans ; Ki-67 Antigen ; analysis ; Male ; Mandibular Neoplasms ; diagnostic imaging ; pathology ; surgery ; Middle Aged ; Neoplasm Recurrence, Local ; Observer Variation ; Prognosis ; Radiography, Panoramic ; Retrospective Studies ; Young Adult

Primary sinonasal ameloblastoma is an extremely rare odontogenic epithelial tumor histomorphologically identical to its gnathic counterparts but with distinct epidemiologic and clinicopathologic characteristics. We present a case of a 46-year-old female with a one year history of recurrent epistaxis, nasal obstruction and frontonasal headache. Clinical examination, CT scan and subsequent surgical excsion revealed an intranasal mass attached to the lateral nasal cavity with histomorphologic features of ameloblastoma and was signed out as extragnathic soft tissue ameloblastoma of the sinonasal area. Extraosseous extragnathic primary sinonasal ameloblastoma are rare but do occur and should be distinguished from infrasellar craniopharyngiomas.

Human ; Female ; Middle Aged ; AMELOBLASTOMA ; ODONTOGENIC TUMORS ; CALCIFYING EPITHELIAL ODONTOGENIC TUMOR ; NEOPLASMS BY HISTOLOGIC TYPE ; NEOPLASMS ; EPISTAXIS ; RECURRENCE ; DIAGNOSTIC IMAGING ; MEDICAL IMAGING ; DIAGNOSTIC TECHNIQUES AND PROCEDURES ; TOMOGRAPHY SCANNERS, X-RAY COMPUTED

OBJECTIVETo assess the value of multi-slice spiral CT (MSCT) with three dimensional (3D) reconstruction in the diagnosis of neoplastic lesions in the jawbones.

METHODSThirty-three patients with neoplastic lesions of the jawbones underwent MSCT scanning with 3D reconstruction. Of these patients, 14 had ameloblastoma, 8 had hemangioma, 3 had osteosarcoma, 3 had ossifying fibroma, 2 had chondrosarcoma, 2 had fibrosarcoma, and 1 had odontogenic myxoma. Preoperative MSCT scanning was performed with the slice thickness of 2 mm, and 3D reconstruction of the images was conducted by means of multi-planar reconstruction (MPR), curved-planar reformation (CRP), and 3D volume rendering technique (VRT). The results were compared with those observed during the operations.

RESULTSIn the 33 cases, the neoplastic lesions of the jawbones were displayed by 2D or 3D imaging and confirmed by intraoperative findings. Two-dimensional imaging allowed better observation than 3D imaging of the deep structures, whereas 3D imaging was superior in visualizing the morphological changes of the compromised bones and the spatial relationship between the tumors and surrounding structures. Two-dimensional imaging and MPR were excellent in revealing the internal structures and pathological changes of tumors, having also better performance in showing the tumors involving the soft tissues. Benign tumors were most visualized as bone expansion changes with well defined ovoid or lobulated borderlines, and malignant ones often resulted in adjacent bony destruction and soft tissue masses.

CONCLUSIONMSCT examination is useful in defining the scope of tumor involvement and bony changes to help in the definite diagnosis, differential diagnosis and choice of clinical treatment. Two-dimensional imaging, MPR, VRT and CRP have their respective advantages and limitations in showing jawbone tumor, and their combination can be of great clinical value.

Adolescent ; Adult ; Ameloblastoma ; diagnosis ; Female ; Hemangioma ; diagnosis ; Humans ; Imaging, Three-Dimensional ; methods ; Jaw ; diagnostic imaging ; Jaw Neoplasms ; diagnosis ; Male ; Middle Aged ; Reproducibility of Results ; Sensitivity and Specificity ; Tomography, Spiral Computed ; methods ; Young Adult