Imaging characteristics of ameloblastomas and diagnostic value of computed tomography and magnetic resonance imaging in a series of 26 patients

Clinicoradiologic features of ameloblastomas: A single-centre study of 155 cases

BACKGROUND

The purpose of the current study was to report on the clinical presentation and radiologic features of 155 cases of ameloblastoma (AB), representing a detailed, large, single-centre radiologic study.

METHODS

Histologically confirmed cases were reviewed over 11 years. Demographic and clinical data were retrieved from the patient's records. Radiologic information was analysed from available radiographs. The radiologic features of ABs were assessed according to the mean age of presentation and the mean duration of the lesion. The distinguishing radiologic features between adults/children and sex were also evaluated.

RESULTS

A statistically significant correlation existed between loss of border demarcation and advanced mean age. Multilocular lesions were markedly more common in adults compared to children. Multilocular ABs were associated with increased lesion duration and advanced mean age. Radiologic signs of reactive bony changes associated with the tumour presented at the highest mean duration of all bony effects. Bony expansion and cortical destruction were statistically correlated with lesion duration. Tooth impaction was more common in children. Some mandibular lesions reached a significant size, resulting in impingement of the maxillary sinus, zygoma, orbit and pterygoid plates.

CONCLUSION

Due to unfortunate healthcare access constraints, ABs grow to significant sizes and exhibit features not often reported in the literature. The findings of this analysis highlighted the radiologic features of ABs expressed through the mean age and duration of the lesion. This emphasises the significance of timely management of these lesions.

Diagnosing oral and maxillofacial diseases using deep learning

The classification and localization of odontogenic lesions from panoramic radiographs is a challenging task due to the positional biases and class imbalances of the lesions. To address these challenges, a novel neural network, DOLNet, is proposed that uses mutually influencing hierarchical attention across different image scales to jointly learn the global representation of the entire jaw and the local discrepancy between normal tissue and lesions. The proposed approach uses local attention to learn representations within a patch. From the patch-level representations, we generate inter-patch, i.e., global, attention maps to represent the positional prior of lesions in the whole image. Global attention enables the reciprocal calibration of path-level representations by considering non-local information from other patches, thereby improving the generation of whole-image-level representation. To address class imbalances, we propose an effective data augmentation technique that involves merging lesion crops with normal images, thereby synthesizing new abnormal cases for effective model training. Our approach outperforms recent studies, enhancing the classification performance by up to 42.4% and 44.2% in recall and F1 scores, respectively, and ensuring robust lesion localization with respect to lesion size variations and positional biases. Our approach further outperforms human expert clinicians in classification by 10.7 % and 10.8 % in recall and F1 score, respectively.

Ameloblastoma of the jaws in adult: A retrospective review of local recurrent lesions based on the resection margin in the adjacent apparent healthy tissues

Background

The surgical treatment of ameloblastoma of the jaws remains contentious due to the variable recurrence rate amongst its variants, the tumor's local invasive behavior, and the lack of consensus among surgeons concerning the extent of resection in the contiguous healthy tissues.

Objective

To determine the recurrence rate of ameloblastoma and its association with the resection margins.

Materials and methods

This is a retrospective cohort study of the medical records of patients who underwent surgical resection of the jaws as the primary modality of treatment for ameloblastoma. Clinical data over the 26 years were analyzed for age, gender, site of the lesion, size, radiographic appearance, histopathological sub-type, and the incidence of recurrence post-treatment. Descriptive and bivariate statistics were computed.

Results

A retrospective audit of 234 cases was included in the study that was typical (solid/multicystic) ameloblastoma. The age of patients ranged from 20 to 66 years with a mean age of 33.4 ± 9.6 years, and a male-to-female ratio of 1.2: 1 (P = 0.52). The follicular and plexiform types accounted for the majority of histopathological variants (89.8%; P = 0.000). Overall, 6.8% of cases relapsed after the initial primary surgery. The rate of recurrence was high with a resection margin of 1.0 or 1.5 cm than 2.0 cm (P = 0.001). No case of recurrence was seen with a resection margin of 2.5 cm margin.

Conclusion

A low recurrence rate of 6.8% was noted in our series of cases. A wide 2.5 cm resection margin is recommended in the adjacent healthy tissues.

Diagnostic efficacy of apparent diffusion coefficient, texture features, and their combination for differential diagnosis of odontogenic cysts and tumors

OBJECTIVE

This study assessed the diagnostic efficacy of apparent diffusion coefficient (ADC), texture features, and their combination for the differential diagnosis of odontogenic cysts and tumors with cyst-like features.

STUDY DESIGN

In total, 14 dentigerous cysts (DCs), 12 odontogenic keratocysts (OKCs), and 6 unicystic ameloblastomas (UABs) were used as predictor variables in 32 outpatients who underwent magnetic resonance imaging. The outcome variables were ADC, texture features, and their combination for each lesion. Texture features including histogram and gray-level co-occurrence matrix (GLCM) were measured on ADC maps. Ten features were selected by using the Fisher coefficient method. The Kruskal-Wallis test and post hoc Mann-Whitney test with Bonferroni adjustment were used to analyze trivariate statistics. Statistical significance was established at P < .05. Receiver operating characteristic analysis was used to evaluate the diagnostic effect of ADC, texture features, and their combination in distinguishing the lesions from each other.

RESULTS

Apparent diffusion coefficient, 1 histogram feature, 9 GLCM features, and their combination demonstrated significant differences between DC, OKC, and UAB (P ≤ .01). Receiver operating characteristic analysis revealed a high area under the curve of .95 to 1.00 for ADC, 10 texture features, and their combination. Sensitivity, specificity, and accuracy ranged from .86 to 1.00.

CONCLUSIONS

Apparent diffusion coefficient and texture features, alone or in combination, can be clinically important in facilitating the distinction between these odontogenic lesions.

A Rare Case of Deceptive Maxillary Ameloblastoma

Ameloblastoma (AM) is a rare epithelium-derived odontogenic tumor, mostly involving the mandible and less often the maxilla. Most AMs are benign and characterized by indolence and local invasiveness, with a high recurrence rate. Herein, we present a case of maxillary AM in a 42-year-old female suffering from left nasal congestion and facial swelling for almost one month after endoscopic surgery at a local hospital. The mass was completely resected by a transnasal functional endoscopic sinus surgery based on radiographic examination. Subsequently, postsurgical histopathological examinations were conducted, and she was diagnosed with a plexiform AM pattern. Immunohistochemical staining revealed that the tumor was positive for PCK, P63, CK5/6, and CK14 but negative for S100, ER, and Ki67. Based on these findings, the patient was diagnosed with maxillary AM.

Contrast-enhanced multidetector computed tomography features and histogram analysis can differentiate ameloblastomas from central giant cell granulomas

BACKGROUND

No qualitative or quantitative analysis of contrast-enhanced computed tomography (CT) images has been reported for the differentiation between ameloblastomas and central giant cell granulomas (CGCGs).

AIM

To describe differentiating multidetector CT (MDCT) features in CGCGs and ameloblastomas and to compare differences in enhancement of these lesions qualitatively and using histogram analysis.

METHODS

MDCT of CGCGs and ameloblastomas was retrospectively reviewed to evaluate qualitative imaging descriptors. Histogram analysis was used to compare the extent of enhancement of the soft tissue. Fisher’s exact tests and Mann–Whitney U test were used for statistical analysis (P < 0.05).

RESULTS

Twelve CGCGs and 33 ameloblastomas were reviewed. Ameloblastomas had a predilection for the posterior mandible with none of the CGCGs involving the angle. CGCGs were multilocular (58.3%), with a mixed lytic sclerotic appearance (75%). Soft tissue component was present in 91% of CGCGs, which showed hyperenhancement (compared to surrounding muscles) in 50% of cases, while the remaining showed isoenhancement. Matrix mineralization was present in 83.3% of cases. Ameloblastomas presented as a unilocular (66.7%), lytic (60.6%) masses with solid components present in 81.8% of cases. However, the solid component showed isoenhancement in 63%. No matrix mineralization was present in 69.7% of cases. Quantitatively, the enhancement of soft tissue in CGCG was significantly higher than in ameloblastoma on histogram analysis (P < 0.05), with a minimum enhancement of > 49.05 HU in the tumour providing 100% sensitivity and 85% specificity in identifying a CGCG.

CONCLUSION

A multilocular, lytic sclerotic lesion with significant hyperenhancement in soft tissue, which spares the angle of the mandible and has matrix mineralization, should indicate prospective diagnosis of CGCG.

Artificial intelligence in dentomaxillofacial radiology

Artificial intelligence (AI) has the potential to revolutionize healthcare and dentistry. Recently, there has been much interest in the development of AI applications. Dentomaxillofacial radiology (DMFR) is within the scope of these applications due to its compatibility with image processing methods. Classification and segmentation of teeth, automatic marking of anatomical structures and cephalometric analysis, determination of early dental diseases, gingival, periodontal diseases and evaluation of risk groups, diagnosis of certain diseases, such as; osteoporosis that can be detected in jaw radiographs are among studies conducted by using radiological images. Further research in the field of AI will make great contributions to DMFR. We aim to discuss most recent AI-based studies in the field of DMFR.

A retrospective cohort study on predictors associated with skull base invasion of maxillary ameloblastomas

PURPOSE

To identify factors associated with skull base involvement (SBI) of maxillary ameloblastomas (MA).

METHODS

This retrospective cohort study was composed of MA patients treated during a 7-year period. Demographic, radiographic, and nine immunohistopathologic predictor variables were included. The outcome variable was presence of SBI (yes/no). Descriptive, bi- and multivariate statistics were computed, and P ≤ .05 in multivariate analyses was considered statistically significant.

RESULTS

The sample comprised 23 subjects (34.8% females; 21.7% with SBI) with a mean age of 50.3 ± 18.2 years. Candidate predictors of an SBI in MAs were 1) male gender, 2) a low Karnofsky Performance Status score (KPS), 3) multilocular radiolucency, 4) ill-defined margins, 5) cortical perforation, 6) inclusion of an unerupted tooth, 7) moderate to strong reactivity to p53, Ki-67, CD10, astrocyte elevated gene-1 (AEG-1) protein, carbonic anhydrase IX (CA IX), calretinin (calbindin2; CALB2), and BRAF-V600E, and 8) negative to low immunopositivity to α-smooth muscle actin (α-SMA) and syndecan-1 (CD138). However, multivariate analyses confirmed the significant associations of SBI with negative/low syndecan-1 reactivity (P = .003; adjusted odds ratio [OR_adj.], 4.04; 95% confidence interval [95% CI], -.89 to -.48; Pearson's Correlation Coefficient [r] = -.74) and with KPS (P = .003; OR_adj., 4.04; 95% CI, -.78 to -.17; r = -.54) only.

CONCLUSIONS

Our findings suggest an aggressive approach to MAs with negative to low syndecan-1 immunopositivity and/or in multi-morbid patients (who may have difficulty in access to health care). Otherwise, health care inequalities due to low KPS scores should be minimized or eliminated.

Deep learning based diagnosis for cysts and tumors of jaw with massive healthy samples

We aimed to develop an explainable and reliable method to diagnose cysts and tumors of the jaw with massive panoramic radiographs of healthy peoples based on deep learning, since collecting and labeling massive lesion samples are time-consuming, and existing deep learning-based methods lack explainability. Based on the collected 872 lesion samples and 10,000 healthy samples, a two-branch network was proposed for classifying the cysts and tumors of the jaw. The two-branch network is firstly pretrained on massive panoramic radiographs of healthy peoples, then is trained for classifying the sample categories and segmenting the lesion area. Totally, 200 healthy samples and 87 lesion samples were included in the testing stage. The average accuracy, precision, sensitivity, specificity, and F1 score of classification are 88.72%, 65.81%, 66.56%, 92.66%, and 66.14%, respectively. The average accuracy, precision, sensitivity, specificity, and F1 score of classification will reach 90.66%, 85.23%, 84.27%, 93.50%, and 84.74%, if only classifying the lesion samples and healthy samples. The proposed method showed encouraging performance in the diagnosis of cysts and tumors of the jaw. The classified categories and segmented lesion areas serve as the diagnostic basis for further diagnosis, which provides a reliable tool for diagnosing jaw tumors and cysts.

A Case Report on Acanthomatous Ameloblastoma of the Anterior Mandible with Brief Review on Advanced Imaging Diagnosis

Ameloblastoma is the most common benign odontogenic tumor of epithelial origin. It exhibits a locally aggressive behavior and high recurrence rate with multitude of factors involving in its molecular pathogenesis. This article reports a case of acanthomatous ameloblastoma involving the mandible in a 60-year-old male patient with peculiar imaging characteristics. The role of computed tomography and magnetic resonance imaging in diagnosis of conventional ameloblastoma has been elaborately emphasized in the discussion. Although the final diagnosis is based on histopathological features, physicians should be aware of the role of advanced imaging for diagnosis of ameloblastoma and for better surgical management.

Usefulness of MR imaging for odontogenic tumors

MRI has become an invaluable diagnostic tool in all areas of the body. However, it has not been widely used to image odontogenic tumors of the jaw. Major advantages of MRI include excellent soft tissue contrast in the absence of ionizing radiation. Furthermore, diffusion-weighted MRI and dynamic contrast-enhanced MRI can be used as functional imaging techniques for assessing tissue biology. In this review article, we present representative MR images of several types of odontogenic tumors, and discuss MR imaging characteristics useful for differential diagnosis.

Ameloblastoma of the mandible: analysis of radiographic and histopathological features

Introduction: Ameloblastoma is the most common benign tumour of odontogenic origin in Africa and presents five times more in the mandible than the maxilla. The presentation of ameloblastoma in the mandible is unique due to its anatomical variation and hence the aim of this study was to analyse the radiographic and histopathological features of ameloblastoma involving the mandible. Materials and methods: This was a retrospective, descriptive study of all histopathologically diagnosed ameloblastoma of the mandible over a period of 45 years. Patient demographics, radiographic and histopathological features were recorded and compared to previous studies. Results: A total of 148 lesions were included. The male to female ratio was nearly equal (1.05:1). The majority of patients were below 50 years of age (83.77%) and were black African (58.8%). The posterior region was the most affected site with majority of lesions presenting with multilocular appearance (68.24%) and root resorption (66.38%). Histologically, conventional ameloblastoma was the most common variant (48.65%). Conclusions: Mandibular ameloblastoma had a higher predilection for black African patients with higher prevalence of mixed density lesions when compared to previous studies. The size of lesions in this sample was considerably larger than those reported in previous studies. In addition, lesions in this sample also exhibited marked cortical expansion as well as root resorption.

Fluorescently Labeled Cetuximab-IRDye800 for Guided Surgical Excision of Ameloblastoma: A Proof of Principle Study

PURPOSE

Fluorescently labeled epidermal growth factor receptor (EGFR) antibodies have successfully identified microscopic tumors in multiple in vivo models of human cancers with limited toxicity. The present study sought to demonstrate the ability of fluorescently labeled anti-EGFR, cetuximab-IRDye800, to localize to ameloblastoma (AB) tumor cells in vitro and in vivo.

MATERIAL AND METHODS

EGFR expression in AB cells was confirmed by quantitative real-time polymerase chain reaction and immunohistochemistry. Primary AB cells were labeled in vitro with cetuximab-IRDye800 or nonspecific IgG-IRDye800. An in vivo patient-derived xenograft (PDX) model of AB was developed. The tumor tissue from 3 patients was implanted subcutaneously into immunocompromised mice. The mice received an intravenous injection of cetuximab-IRDye800 or IgG-IRDye800 and underwent imaging to detect infrared fluorescence using a Pearl imaging system (LI-COR Biosciences, Lincoln, NE). After resection of the overlying skin, the tumor/background ratios (TBRs) were calculated and statistically analyzed using a paired t test.

RESULTS

EGFR expression was seen in all AB samples. Tumor-specific labeling was achieved, as evidenced by a positive fluorescence signal from cetuximab-IRDye800 binding to AB cells, with little staining seen in the negative controls treated with IgG-IRDye800. In the animal PDX model, imaging revealed that the TBRs produced by cetuximab were significantly greater than those produced by IgG on days 7 to 14 for AB-20 tumors. After skin flap removal to simulate a preresection state, the TBRs increased with cetuximab and were significantly greater than the TBRs with the IgG control for PDX tumors derived from the 3 patients with AB. The excised tissues were embedded in paraffin and examined to confirm the presence of tumor.

CONCLUSIONS

Fluorescently labeled anti-EGFR demonstrated specificity for AB cells and PDX tumors. The present study is the first report of tumor-specific, antibody-based imaging of odontogenic tumors, of which AB is one of the most clinically aggressive. We expect this technology will ultimately assist surgeons treating AB by helping to accurately assess the tumor margins during surgery, leading to improved long-term local tumor control and less surgical morbidity.

Luminal, Intramural Unicystic Ameloblastoma with Marked Fluid-Fluid Level: Validity of CT and MRI Findings

We report herein a case of a luminal and intramural unicystic ameloblastoma (UA) with a marked fluid-fluid level. The validity of imaging findings in diagnosing UA in the present case is discussed in reference to the literature. The patient was a 50-year-old woman who presented with swelling of the gingiva in the region of the left mandibular third molar and numbness in the lower lip. Computed tomography (CT) and magnetic resonance imaging (MRI) revealed a large mass lesion with a unilocular appearance and a biphasic aspect, suggesting liquid content. Contrast-enhanced MRI (CE-MRI) and dynamic contrast-enhanced MRI (DCE-MRI) demonstrated that the biphasic aspect indicated a fluid-fluid level with no blood pooling/flow; it also revealed a thick rim-enhanced margin with mural protrusion. Postoperatively, the lesion was histopathologically diagnosed as a luminal and intramural UA. In conclusion, extensive imaging including both standard CT and MRI together with CE-MRI and DCE-MRI allowed mural protrusions or nodules on a thick cystic wall and liquid content to be correctly identified. This suggests that such imaging can play an important role in diagnosing a UA, even though the results were at first misleading due to the marked fluid-fluid level.

Updated WHO nomenclature of head and neck lesions and associated imaging findings

This article reviews the imaging features of head and neck lesions with updated 2017 World Health Organization (WHO) nomenclature. The major WHO changes include refined terminology of existing entities, descriptions of new tumor types, elimination of defunct categories, and updated biological characterization of various tumor types. In particular, the updates pertaining to the following conditions will be reviewed: tumors of the oral cavity and oropharynx, including HPV-positive or HPV-negative squamous cell carcinoma, small cell carcinoma; tumors of the hypopharynx, larynx, trachea, and parapharyngeal space, including nomenclature revisions for laryngeal neuroendocrine tumors; tumors of the nasal cavity and paranasal sinuses including newly added entities such as NUT carcinoma and biphenotypic sinonasal sarcoma; odontogenic and maxillofacial bone tumors, including the reversal of terminology for certain cystic lesions; tumors of the salivary glands, including updated terminology related to high-grade transformation and polymorphous adenocarcinomas tumors; temporal bone lesions including modifications of the nomenclature and classification criteria; tumor-like lesions of the neck and lymph nodes, with a discussion encompassing developmental cysts, metastases of unknown primary, and heterotopia-associated neoplasia; and mucosal melanoma. Familiarity with the proper WHO terminology for conditions that might be mentioned in differential diagnoses and a general understanding of the behavior of head and neck lesions can help optimize imaging assessment and reporting.

Radiographic estimation of the growth rate of initially underdiagnosed ameloblastomas

BACKGROUND

To evaluate the specific growth rate (SGR) of ameloblastoma.

MATERIAL AND METHODS

cases of ameloblastoma initially underdiagnosed (e.g. cases overlooked or diagnosed as reactive lesions) which had adequate radiographic documentation to evaluate their progression were retrospectively selected. Two panoramic radiographs were analyzed to determine the specific growth rate (SGR) of each tumor, defined as the logarithm of the ratio of final tumor area (when the diagnosis of ameloblastoma was made) to the initial tumor area (when the lesion was underdiagnosed), divided by the time interval between the radiographic images. The tumor area was measured using the software ImageJ.

RESULTS

Twelve patients with mandibular ameloblastomas were selected, including 5 males and 7 females, with a mean age of 24.9 years (range: 14-61 years). In four cases, the lesion was associated with the crown of an impacted third molar. In three cases, it was initially diagnosed as a periapical lesion. Three cases were extrafollicular and were not noticed in the initial radiographs. Two cases were initially diagnosed as ameloblastoma, but the surgery was delayed for personal reasons. The mean interval of time between the two radiographic images was 4.3 years (range: 0.4-9 years). Based on our analysis, ameloblastoma grows in average 40.4% per year (range: 14.9-88.7%).

CONCLUSIONS

Ameloblastoma is a progressively growing tumor, but its growth rate seems to be smaller than initially reported in the literature. Better understanding the radiographic progression of ameloblastoma might improve its early diagnosis, management, and prognosis.

Three-dimensional radiographic features of ameloblastoma and cystic lesions in the maxilla

OBJECTIVES

To characterize the radiographic features of maxillary ameloblastoma (AM), odontogenic keratocyst (OKC) and dentigerous cyst (DC) comparatively by using spiral CT and cone beam CT (CBCT).

METHODS

Clinical records, histopathological reports, and nonenhanced spiral CT or CBCT images of 191 consecutive patients with primary maxillary AMs, OKCs, or DCs were retrospectively acquired, and radiographic features were analyzed.

RESULTS

The study included 118 males and 73 females (age: 5-84 years). 72.0% of AMs and 84.3% of OKCs originated from the posterior maxilla, while 69.6% of DCs occurred in the anterior maxilla. Among 25 AMs, 44.0% were of desmoplastic type, with honey-combed appearance. 84.0% of AMs were circular or oval in shape, 84.0% expanded buccally, and 36.0% invade the nasal floor. Among 89 OKCs of 88 patients, 61.8% were circular or oval, 58.4% expanded buccally, 49.4% were dentigerous, 41.6% nearly filled the maxillary sinus, and 13.5% invaded the nasal floor. 93.7% (74/79) of DCs enveloped a single tooth, and the tooth-cyst relationship was centripetal in 35, eccentric in 30, and circumferential in 9. Moreover, 98.2% (55/56) of the cysts enveloping a supernumerary tooth were DCs, while 80.9% (38/47) of the cysts enveloping the third molar were OKCs.

CONCLUSIONS

Maxillary AMs tend to grow with buccal expansion and invade the nasal floor, and DAs with honey-combed lobularity are common. Maxillary OKCs have variant shapes and tend to invaginate the maxillary sinus. The tooth-cyst relationship of dentigerous OKCs and DCs can be centripetal, eccentric, or circumferential.

Oral Rehabilitation of Patients with Ameloblastoma of the Mandible. Clinical Results in Three Patients with Different Bone Reconstruction Techniques

Background:

Ameloblastoma is the second most common odontogenic tumor. It shows a locally aggressive behavior, with a high level of recurrence. Wide resection of the jaw is recommended for treatment of ameloblastoma. However, radical surgery causes an abnormal mandibular movement, facial asymmetry, and masticatory dysfunction.

Methods:

Three cases of different types of ameloblastoma is presented, with different reconstruction techniques including Non-Vascularized Bone Graft (NVBG), Osteocutaneous Fibula Free Flap (OFFF), and Deep Circumflex Iliac Artery flap (DCIA).

Results:

In all three cases the tumor site was successfully reconstructed to obtain very good esthetic results as well as functional oral rehabilitation with implants and fixed prosthetics for optimal masticatory function.

Conclusion:

For reconstruction of the mandible, we prefer bone grafts from the iliac crest. The natural curvature and variable bone height offer a very good reconstruction of the defect.

Ameloblastomas of the mandible and maxilla

Ameloblastoma is a histologically benign but locally aggressive tumor of the jaws. We conducted a retrospective cohort study to review the clinical, radiologic, and pathologic features of patients with ameloblastoma of the mandible and maxilla and to report the outcomes of treatment. Our study population was made up of a series of 30 consecutively presenting patients-15 men and 15 women, aged 19 to 81 years (median: 61.5)-who had undergone their primary treatment of ameloblastoma of the mandible or maxilla at Memorial Sloan Kettering Cancer Center from January 1987 through December 2012. In addition to demographic data, we compiled information on clinical characteristics, imaging findings, the type of surgery, surgical margins, adjuvant treatments, histologic patterns, length of follow-up, time to recurrence, treatment of recurrence, and factors that had an influence on recurrence. All but 2 patients with negative margins were cured. Favorable outcomes were associated with the administration of adjuvant postoperative radiotherapy for patients with positive margins and a repeat resection for patients with recurrences. Complete excision with negative margins, however, remains the gold standard for curative treatment.

Ameloblastoma-Clinical, radiological, and therapeutic findings

Ameloblastoma are the most common odontogenic tumor. As they usually do not form metastasis, they are considered as benign tumors with a locally invasive growth pattern and destruction of the jaws and the surrounding tissue (Oral Diseases, 23, 2017, 199). This article focuses on clinical, radiological, and therapeutic findings, which may influence diagnosis and treatment of ameloblastoma in the future.

Application of Convolutional Neural Network in the Diagnosis of Jaw Tumors

Objectives

Ameloblastomas and keratocystic odontogenic tumors (KCOTs) are important odontogenic tumors of the jaw. While their radiological findings are similar, the behaviors of these two types of tumors are different. Precise preoperative diagnosis of these tumors can help oral and maxillofacial surgeons plan appropriate treatment. In this study, we created a convolutional neural network (CNN) for the detection of ameloblastomas and KCOTs.

Methods

Five hundred digital panoramic images of ameloblastomas and KCOTs were retrospectively collected from a hospital information system, whose patient information could not be identified, and preprocessed by inverse logarithm and histogram equalization. To overcome the imbalance of data entry, we focused our study on 2 tumors with equal distributions of input data. We implemented a transfer learning strategy to overcome the problem of limited patient data. Transfer learning used a 16-layer CNN (VGG-16) of the large sample dataset and was refined with our secondary training dataset comprising 400 images. A separate test dataset comprising 100 images was evaluated to compare the performance of CNN with diagnosis results produced by oral and maxillofacial specialists.

Results

The sensitivity, specificity, accuracy, and diagnostic time were 81.8%, 83.3%, 83.0%, and 38 seconds, respectively, for the CNN. These values for the oral and maxillofacial specialist were 81.1%, 83.2%, 82.9%, and 23.1 minutes, respectively.

Conclusions

Ameloblastomas and KCOTs could be detected based on digital panoramic radiographic images using CNN with accuracy comparable to that of manual diagnosis by oral maxillofacial specialists. These results demonstrate that CNN may aid in screening for ameloblastomas and KCOTs in a substantially shorter time.

The accuracy of ultrashort echo time MRI sequences for medical additive manufacturing

OBJECTIVES

Additively manufactured bone models, implants and drill guides are becoming increasingly popular amongst maxillofacial surgeons and dentists. To date, such constructs are commonly manufactured using CT technology that induces ionizing radiation. Recently, ultrashort echo time (UTE) MRI sequences have been developed that allow radiation-free imaging of facial bones. The aim of the present study was to assess the feasibility of UTE MRI sequences for medical additive manufacturing (AM).

METHODS

Three morphologically different dry human mandibles were scanned using a CT and MRI scanner. Additionally, optical scans of all three mandibles were made to acquire a "gold standard". All CT and MRI scans were converted into Standard Tessellation Language (STL) models and geometrically compared with the gold standard. To quantify the accuracy of the AM process, the CT, MRI and gold-standard STL models of one of the mandibles were additively manufactured, optically scanned and compared with the original gold-standard STL model.

RESULTS

Geometric differences between all three CT-derived STL models and the gold standard were <1.0 mm. All three MRI-derived STL models generally presented deviations <1.5 mm in the symphyseal and mandibular area. The AM process introduced minor deviations of <0.5 mm.

CONCLUSIONS

This study demonstrates that MRI using UTE sequences is a feasible alternative to CT in generating STL models of the mandible and would therefore be suitable for surgical planning and AM. Further in vivo studies are necessary to assess the usability of UTE MRI sequences in clinical settings.