Deep learning for detection and 3D segmentation of maxillofacial bone lesions in cone beam CT.

OBJECTIVES: To develop an automated deep-learning algorithm for detection and 3D segmentation of incidental bone lesions in maxillofacial CBCT scans. METHODS: The dataset included 82 cone beam CT (CBCT) scans, 41 with histologically confirmed benign bone lesions (BL) and 41 control scans (without lesions), obtained using three CBCT devices with diverse imaging protocols. Lesions were marked in all axial slices by experienced maxillofacial radiologists. All cases were divided into sub-datasets: training (20,214 axial images), validation (4530 axial images), and testing (6795 axial images). A Mask-RCNN algorithm segmented the bone lesions in each axial slice. Analysis of sequential slices was used for improving the Mask-RCNN performance and classifying each CBCT scan as containing bone lesions or not. Finally, the algorithm generated 3D segmentations of the lesions and calculated their volumes. RESULTS: The algorithm correctly classified all CBCT cases as containing bone lesions or not, with an accuracy of 100%. The algorithm detected the bone lesion in axial images with high sensitivity (95.9%) and high precision (98.9%) with an average dice coefficient of 83.5%. CONCLUSIONS: The developed algorithm detected and segmented bone lesions in CBCT scans with high accuracy and may serve as a computerized tool for detecting incidental bone lesions in CBCT imaging. CLINICAL RELEVANCE: Our novel deep-learning algorithm detects incidental hypodense bone lesions in cone beam CT scans, using various imaging devices and protocols. This algorithm may reduce patients' morbidity and mortality, particularly since currently, cone beam CT interpretation is not always preformed. KEY POINTS: • A deep learning algorithm was developed for automatic detection and 3D segmentation of various maxillofacial bone lesions in CBCT scans, irrespective of the CBCT device or the scanning protocol. • The developed algorithm can detect incidental jaw lesions with high accuracy, generates a 3D segmentation of the lesion, and calculates the lesion volume.

Integrated therapy of intraductal irrigations and sialoendoscopies of salivary glands to improve mouth dryness.

OBJECTIVE: To assess the combination of salivary gland intraductal irrigations (IG) followed by sialoendoscopy irrigations (SI) of the parotid gland on the improvement of salivary gland secretory dysfunction (SGSD). METHODS: We retrospectively analyzed the records of SGSD patients who underwent major salivary gland IG followed by SI during 2014-2020. Records included demographics, systemic background, signs, and symptoms. Improvement was assessed by comparing the mean unstimulated and stimulated whole salivary flow rate (UWSF, SWSF) from the baseline point (before IG procedure) to the last point (after SI) using repeated measures. The between-subjects effects of various factors and covariants were analyzed using repeated measures ANCOVA. RESULTS: One hundred patients were included with an age range of 15-83 years (mean age of 60.1 ± 13.1 years). Improvement was detected on UWSF measurements (p = 0.031, F = 3.83), but not on SWSF measurements (p = 0.165, F = 1.85). The between-subjects effects on UWSF measurements were statistically significant for sex (p = 0.003, F = 9.526) and salivary gland manipulators use (p < 0.001, F = 15.107) and for the interaction between sex and salivary gland manipulators use (p- = 0.002, F = 9.709). Results of long-term follow-up for 10.87 ± 11.79 months after the SI procedure demonstrated sustained improvement in UWSF measurements (p = 0.011, F = 4.91). CONCLUSIONS: The combination of IG followed by SI increases UWSF salivary secretion in SGSD patients for a relatively extended duration.

Iodinated contrast media allergy is rare in patients undergoing sialography.

BACKGROUND: Iodinated contrast media allergy is considered as a strong contra-indication for performing sialography. There is little evidence to support this approach. OBJECTIVE: To evaluate the rate of iodinated contrast media (ICM) allergy in subjects undergoing sialography and to assess the risk for allergic responses in patients with a previous diagnosis of allergy. METHODS: We retrospectively reviewed sialo-CBCT studies performed from 2014 to 2019. During the study period we implemented a protocol for performing sialo-CBCT in patients with a prior diagnosis of allergy: 1) Clinical data were collected from a questionnaire and medical records. 2) No premedication was administered but, instead, oxygen, epinephrine and a resuscitation cart were accessible. 3) Following the procedure, each patient was observed for one hour and contacted by telephone 24 hrs later. RESULTS: No allergic responses were documented in the medical records of 1515 subjects following sialo-CBCT studies, including 13 individuals previously diagnosed with ICM allergy. Investigation of the subgroup with prior allergy disclosed that the range of injected volume was between 2 ml to 6.2 ml per patient and that complete secretion of ICM was detected in 7 of 13 patients. In the remainder of subjects, retention rates of 5-50% were observed. CONCLUSIONS: Allergic reactions are exceedingly rare following sialo-CBCT studies regardless of a previous diagnosis of allergy. Pre-medication with corticosteroids and antihistamines is usually not warranted.

Novel parotid sialo-cone-beam computerized tomography features in patients with suspected Sjogren's syndrome.

OBJECTIVE: Sjogren's syndrome (SjS) causes salivary gland impairment leading to oral dryness. Parotid sialo-cone-beam computerized tomography (sialo-CBCT) demonstrates ductal architecture and to a lesser extent gland activity. This study characterizes radiographic features of patients suspected for SjS and looks for a possible correlation with the diagnosis of SjS. METHODS: The clinical and radiographic data of suspected SjS/dry mouth patients referred for sialo-CBCT in 2011-2014 were reviewed retrospectively. Two observers studied the scans for various radiographic features including duct morphology, level of branching, ductopenia and sialectasia. These features were analysed taking the specific clinical data and two sets of SjS criteria: The 2002 American-European Consensus Group (AECG) and the 2012 American College of Rheumatology (ACR) Group. RESULTS: Sialo-CBCT scans of 67-referred patients suffering from dry mouth (115 parotid glands) were included. Intraradiographic association was found between ductopenia and all other radiographic parameters. Minimal, yet important, radiographic differences were found between left and right parotid glands. AECG-confirmed-SjS patients showed strong correlation with radiographic features, whereas ACR 2012-confirmed-SjS patients did not. CONCLUSION: Sialo-CBCT demonstrates novel radiographic features which may clarify the diagnosis of SjS. Further studies are needed to determine the role of sialo-CBCT in diagnosis of SjS.

Improvement of dry mouth following intraductal irrigation of salivary glands.

OBJECTIVES: To assess the effect of major salivary gland intraductal irrigations (IGs) to relieve mouth dryness. METHODS: We retrospectively analyzed the records of patients with mouth dryness who underwent major salivary gland IG during 2013-2015. Records included demographics, medical background, dry mouth etiologies and symptomatology, and results of sialometry and sialo-cone-beam computerized tomography. Subjective improvement following the IG procedure (yes/no) and sustained subjective improvement (mouth dryness relief for ≥1 month) were recorded. Objective improvement was assessed by comparing the mean unstimulated (USF) and stimulated (SSF) whole salivary flow (WSF) rate before and after the IG. RESULTS: Seventy-four patients were included [mean age: 59.08 ± 12.46 years]. Improvement was detected in the USF (p = .027), but not in the SSF (p = .878). Fifty-five (84.6%) noted subjective improvement, while 10 (15.4%) did not. Subjective improvement was positively associated with the USF following IG (p = .037), with salivary gland swelling episodes (p = .033), and with difficulties in swallowing dry foods (p = .014). Of those with subjective improvement, 45 (81.8%) reported sustained improvement, which was positively associated with lack of a gritty eye sensation (p = .042) and abnormal sialo-CBCT findings (p = .001). CONCLUSIONS: Major salivary gland IG is a simple and safe procedure that may relieve dry mouth for a relatively extended duration. Further studies are needed to confirm these preliminary findings and assess their underlying mechanisms.

Panoramic errors in pediatric patients with special needs.

This study aimed to analyze the types and frequencies of panoramic technical errors in pediatric patients with special needs, as compared to matching images of patients with normal developmental patterns. Panoramic images of 100 pediatric patients, with equal numbers of either special needs or healthy controls, referred to our Oral Maxillofacial Imaging unit, were retrospectively examined by four blinded observers for the presence of four common technical errors: palatoglossal air space, movement, positioning errors, and foreign bodies appearance. In addition, they subjectively determined the image quality on an ordinal scale. The statistical examination included inter-observer correlation and correlations between demographic factors (age, gender, developmental status) and the number and types of errors. The frequencies of demonstrated errors were, in descending order: movement, positioning, palatoglossal air space, and foreign bodies' appearance. The special needs group images showed significantly more errors and were rated as low-quality radiographs. Younger patients in both groups showed more movement and positioning errors. Technical errors in panoramic images of patients with special needs were more frequently found. Therefore, increased awareness of the staff and appropriate pre-imaging instructions to the patients, are required when imaging pediatric patients, especially those with special needs.

Radiographic findings of space-occupying lesions in sialo-CBCT of the major salivary glands.

OBJECTIVES: When performing CBCT sialography (sialo-CBCT), space-occupying lesions may be identified incidentally. The objective was to describe their radiologic-clinical-histopathologic correlations. METHOD AND MATERIALS: The archive of sialo-CBCT scans was retrospectively searched for suspected space-occupying lesions. Based on the scan and clinical-histopathologic data, the cases were divided into "pathologic" vs "normal," "intra-parenchymal" vs "extra-parenchymal," and "benign" vs "malignant." Two precalibrated, blinded radiologists performed a survey of the radiographic features of each scan. Cohen kappa, chi-square, Kruskal-Wallis, and Mann-Whitney tests assessed inter-observer agreement and radiologic-clinical-histopathologic correlations. RESULTS: In total, 27 (1.5%) suspected space-occupying lesions were found in 1,758 reports. Full follow-up data were available for 15 cases: four were "malignant," six were "benign," and the remaining five were "normal." Kappa showed substantial inter-observer agreement (0.8 to 1.0). Constant swelling correlated with "pathologic" cases (P = .003). Lesion diameter was greater in "pathologic" than "normal" (P < .001) cases, with a cut-off of 12.6 mm. Clinical and radiographic features were similar in "benign" and "malignant" lesions. "Intra-parenchymal" and "extra-parenchymal" space-occupying lesions correlated with "no-fill-region" (P = .01) and "main-duct-displacement" (P = .002), respectively. CONCLUSIONS: Suspected space-occupying lesions in sialo-CBCT with a diameter greater than 12.6 mm are likely to be "pathologic." No radiographic features were able to differentiate between "malignant" and "benign" space-occupying lesions.

Microbiota-dependent and -independent postnatal development of salivary immunity.

While saliva regulates the interplay between the microbiota and the oral immune system, the mechanisms establishing postnatal salivary immunity are ill-defined. Here, we show that high levels of neutrophils and neonatal Fc receptor (FcRn)-transferred maternal IgG are temporarily present in the neonatal murine salivary glands in a microbiota-independent manner. During weaning, neutrophils, FcRn, and IgG decrease in the salivary glands, while the polymeric immunoglobulin receptor (pIgR) is upregulated in a growth arrest-specific 6 (GAS6)-dependent manner independent of the microbiota. Production of salivary IgA begins following weaning and relies on CD4-help, IL-17, and the microbiota. The weaning phase is characterized by a transient accumulation of dendritic cells capable of migrating from the oral mucosa to the salivary glands upon exposure to microbial challenges and activating T cells. This study reveals the postnatal mechanisms developed in the salivary glands to induce immunity and proposes the salivary glands as an immune inductive site.

Standardization of terminology, imaging features, and interpretation of CBCT sialography of major salivary glands: a clinical review.

Sialography combined with cone beam computerized tomography (sialo-CBCT) is an imaging technique that demonstrates the ductal system of the major salivary glands and allows evaluation of gland function. This review describes the sialo-CBCT technique, terminology, common pitfalls and limitations, as well as radiographic features and suggested pathogenicity of various salivary gland disorders, based on 1,758 sialo-CBCT examinations conducted over the last decade in one institution, and the current literature. The adoption of standardized terminology is proposed to prevent miscommunication, facilitate formulation of differential diagnoses, and thereby promote patient management: (1) Sialo-CBCT requires specific training, and operator experience is required for adequate glandular filling with minimal extravasation; (2) Limit injection-to-scan time to avoid pre-mature emptying; (3) The sialo-CBCT report should include a description of the morphology of the primary duct as well as the secondary, tertiary, and descending branches, the maximal branching level, the presence of sialectasis, overall glandular size, and parenchymal findings; (4) Functional evaluation is based on assessment of iodine clearance in the post evacuation image; (5) Sialectasis and ductopenia are the main findings in Sjogren syndrome and recurrent juvenile parotitis; (6) Sialodochitis with or without fillings defects or hyperdense calcifications characterize obstructive sialadenitis and sialolithiasis; (7) The findings following radioactive-iodine-induced damage are similar to obstructive sialadenitis, with atrophy in late stages; (8) In chronic graft-versus-host disease (cGVHD), variable presentations of ductopenia, sialectasis, and sialodochitis may be evident; (9) The red flags indicating a space-occupying lesion include areas of no filling, splaying of ducts, and primary duct deviation.

Maturation of the neonatal oral mucosa involves unique epithelium-microbiota interactions.

Postnatal host-microbiota interplay governs mucosal homeostasis and is considered to have life-long health consequences. The intestine monolayer epithelium is critically involved in such early-life processes; nevertheless, the role of the oral multilayer epithelium remains ill defined. We demonstrate that unlike the intestine, the neonate oral cavity is immensely colonized by the microbiota that decline to adult levels during weaning. Neutrophils are present in the oral epithelium prenatally, and exposure to the microbiota postnatally further recruits them to the preamble neonatal epithelium by γδT17 cells. These neutrophils virtually disappear during weaning as the epithelium seals. The neonate and adult epithelium display distinct turnover kinetics and transcriptomic signatures, with neonate epithelium reminiscent of the signature found in germ-free mice. Microbial reduction during weaning is mediated by the upregulation of saliva production and induction of salivary antimicrobial components by the microbiota. Collectively, unique postnatal interactions between the multilayer epithelium and microbiota shape oral homeostasis.

An artificial intelligence system using machine-learning for automatic detection and classification of dental restorations in panoramic radiography.

OBJECTIVES: The aim of this study was to develop a computer vision algorithm based on artificial intelligence, designed to automatically detect and classify various dental restorations on panoramic radiographs. STUDY DESIGN: A total of 738 dental restorations in 83 anonymized panoramic images were analyzed. Images were automatically cropped to obtain the region of interest containing maxillary and mandibular alveolar ridges. Subsequently, the restorations were segmented by using a local adaptive threshold. The segmented restorations were classified into 11 categories, and the algorithm was trained to classify them. Numerical features based on the shape and distribution of gray level values extracted by the algorithm were used for classifying the restorations into different categories. Finally, a Cubic Support Vector Machine algorithm with Error-Correcting Output Codes was used with a cross-validation approach for the multiclass classification of the restorations according to these features. RESULTS: The algorithm detected 94.6% of the restorations. Classification eliminated all erroneous marks, and ultimately, 90.5% of the restorations were marked on the image. The overall accuracy of the classification stage in discriminating between the true restoration categories was 93.6%. CONCLUSIONS: This machine-learning algorithm demonstrated excellent performance in detecting and classifying dental restorations on panoramic images.

Demographics and referral patterns of a university-based oral maxillofacial radiology clinic over a 20 year period.

OBJECTIVE: The aim of this study was to review the referral patterns, distribution of interpretations, and type of diagnostic imaging used in a university based oral maxillofacial radiology clinic for patients referred for consultation from both dental and nondental clinicians. STUDY DESIGN: The database of the Special Procedures clinic in the oral radiology department of the University of Toronto, Faculty of Dentistry, containing over 5000 entries, from 1993 to 2013, was queried. Using descriptive categorical analysis, the results were analyzed to describe patient demographic characteristics, the specialty of the referring clinicians, the imaging modalities used, and the interpretation provided. RESULTS: Most referrals were from oral and maxillofacial surgeons and general dentists. Approximately 25% of referrals were interpreted as variations of normal anatomic structures. The most common reasons for referral were intraosseous lesions (42%), temporomandibular joints (39%), and sialography. Ten percent of all referrals were for recall examinations. The distribution of image modalities has changed through the years covered by this study. CONCLUSIONS: This study reflects a lower proportion of referrals reported as normal structures and their variations, than reported in a previous comparable study.

Ameloblastoma incidentally detected in cone-beam computed tomography sialography: A case report and review of the literature.

BACKGROUND AND OVERVIEW: Cone-beam computerized tomographic (CBCT) imaging is increasingly used in the maxillofacial region for various purposes. The volumetric dataset created by means of CBCT increases the possibility of detecting incidental findings. CASE DESCRIPTION: In this case report, the authors describe an otherwise healthy 45-year-old man who was referred from the Sjogren's Syndrome Center in Jerusalem, Israel, for bilateral parotid CBCT sialographic imaging owing to xerostomia lasting for 4 months. CBCT sialographic imaging using iodine contrast material showed normal glandular structure and activity according to the postcannulation panoramic imaging. CBCT sialographic imaging routine volume interpretation revealed a lesion in the mandibular area causing destruction and perforation of the cortical plates, partial disappearance of adjacent lamina dura, and resorption of adjacent teeth, suggesting an aggressive benign tumor. Histopathologic examination confirmed mural cystic ameloblastoma, which was resected with preservation of the mental nerve and the lower mandibular border. The patient received bone graft and implants, with regular follow-ups and no recurrence after 4 years. CONCLUSIONS AND PRACTICAL IMPLICATIONS: This case strengthens the need for thorough examination of the whole CBCT imaging volume within and outside the region of interest to detect incidental findings with clinical importance.

Anterior mandibular rehabilitation causing chronic obstructive sialadenitis: ?A series of sialo-CBCT cases and literature review.

OBJECTIVE: The aim of this study was to assess and present, using sialo-cone beam computed tomography (sialo-CBCT) examination, cases of submandibular gland (SM) chronic obstructive sialadenitis (COS) caused by dental rehabilitation. DATA SOURCES: Clinical and radiographic data of all patients referred for SM sialo-CBCT imaging in the Oral and Maxillofacial Imaging Unit with recurrent SM swelling between January 2012 and July 2015 were reviewed. Cases with suggested iatrogenic cause were selected and described. A literature review of similar cases was also performed. RESULTS: Seventy-one patients with salivary gland (SG) swelling were referred for SM sialo-CBCT. Of these, 16 had implant-supported fixed partial prostheses and seven had full mandibular overdentures. In five patients, COS was diagnosed and a causal relationship between their dental rehabilitation and symptoms was suggested. A literature search revealed 11 cases of SM COS secondary to anterior mandibular rehabilitation. CONCLUSION: Clinicians rehabilitating the anterior edentulous mandibular region should be aware that both fixed and removable devices may block the orifices of the SM, causing COS.

Periodic persistent orofacial pain due to osteomyelitis of the mandible: A case report.

Chronic osteomyelitis of the jaws presents a diagnostic and therapeutic challenge. A 45-year-old woman presented with episodic pain in the right mandible of two and a half years' duration. During active periods, lasting for 2 to 3 weeks, the pain woke the patient from sleep at a regular time, and had features mimicking a variant of cluster headache. However, clinical, radiologic, and histologic findings confirmed the diagnosis of chronic sclerosing osteomyelitis. Conservative treatment, including 6 weeks of intravenous antibiotics, relieved the pain with no signs of recurrence at the 1-year follow-up. We emphasize the importance of including chronic osteomyelitis in the differential diagnosis of idiopathic orofacial pain disorders, particularly in the mandibular area.

Preserving formerly impaired salivary gland function in radioiodine-treated thyroid papillary carcinoma.

Salivary gland impairment after radioiodine therapy (RIT) is a well-known iatrogenic artifact that includes recurrent episodes of sialadenitis, hyposalivation, dysgeusia, and dysphagia. Several therapy modalities have been suggested over the years to decrease these effects on the salivary glands. However, in cases where the salivary gland system is impaired prior to RIT, preserving its function becomes more challenging. We present three patients who underwent RIT with pre-existing insult to the salivary glands and discuss the treatment modality employed. A protocol for treating patients with pre-existing salivary gland impairment prior to RIT is proposed. The dental professional personnel should be aware of RIT insult to the salivary glands in order to preserve more maliciously the oral health in these patients. Management of impaired salivary glands subjected to additional RIT is mandatory to reduce further damage to the glands' function primarily by combining interventional sialoendoscopy for recurrent sialadenitis and sialogogues for hyposalivation.