A deep transfer learning approach for the detection and diagnosis of maxillary sinusitis on panoramic radiographs.

To investigate the use of transfer learning when applying a deep learning source model from one institution (institution A) to another institution (institution B) for creating effective models (target models) for the detection of maxillary sinuses and diagnosis of maxillary sinusitis on panoramic radiographs. In addition, to determine appropriate numbers of training data for the transfer learning. Source model was created using 350 panoramic radiographs from institution A as training data. Transfer learning was performed by adding 25, 50, 100, 150, or 225 panoramic radiographs as training data from institution B to the source model; this yielded the target models T25, T50, T100, T150 and T225. Each model was then evaluated using test data that comprised 40 images from institution A, 30 images from institution B. The performance indices (recall, precision and F1 score) for detecting the maxillary sinuses by the source model exceeded 0.98 when using test data A from institution A, but they deteriorated when using test data B from institution B. In the evaluation of target models using test data B, model T25 showed improved detection performance (recall of 0.967). The diagnostic performance of model T50 for maxillary sinusitis exceeded 0.9 in sensitivity. Transfer learning, which involves applying a small amount of data to the source model, yielded high performances in detecting the maxillary sinuses and diagnosing the maxillary sinusitis on panoramic radiographs. This study serves as a reference when adapting source models to other institutions.

Accuracy and head positioning effects on measurements of anterior tooth length using 3-dimensional and conventional dental panoramic radiography.

INTRODUCTION: The purposes of this study were to examine the accuracy and the head positioning effects on measurements of anterior tooth length using 3-dimensional (3D) and conventional dental panoramic radiography and to investigate whether 3D panoramic radiography is suitable for the evaluation of anterior tooth length. METHODS: A simulated human head was radiographed at 4, 8, and 12 mm displaced positions, and at 5°, 10°, and 15° tilted positions from the standard head position using 3D and conventional panoramic radiography, and also using cone-beam computed tomography. Anterior tooth lengths were measured on the panoramic and cone-beam computed tomography images. The values for the standard head position in the panoramic radiographs were defined as the standard values. Measurement error was defined as the standard value minus the cone-beam computed tomography value on each panoramic radiograph. The head position ratio of the measurement value to the standard value at each head position was calculated. RESULTS: Measurement errors for the 3D panoramic radiographs were significantly smaller than those for the conventional panoramic radiographs. In the 3D panoramic radiographs, the head position ratios at the 4, 8, and 12 mm displaced positions and at the 5° tilted position were within ±5% of the standard value. CONCLUSIONS: Three-dimensional panoramic radiography is suitable for the quantitative evaluation of anterior tooth length with high accuracy.

Can sonographic features be efficacy predictors of robotic massage treatment for masseter and temporal muscle in patients with temporomandibular disorder with myofascial pain?

OBJECTIVE: The aim of this study was to detect sonographic predictors for the efficacy of massage treatment of masseter and temporal muscle in temporomandibular disorders (TMDs) patients with myofascial pain. METHODS: Thirty-seven TMD patients with myofascial pain (6 men and 31 women, a median age of 45 years) were enrolled. An oral rehabilitation robot massaged the patient's masseter and temporal muscles with a standard massage pressure of 10 N for 16 min. The standard treatment protocol was set five sessions every 2 weeks. The median total duration of treatment was 9.5 weeks. Efficacy of treatment was evaluated based on maximum mouth opening and visual analog scale scores of muscle pain and daily life impediments. The intramuscular echogenic bands and elasticity index ratios of the masseter muscles were evaluated on sonographic or sonoelastographic images obtained before treatment and after the third and last treatment sessions. RESULTS: The sonographic features detected different changes after the third treatment session between the therapy-effective and therapy-ineffective groups: in the therapy-effective group, the frequency of visibility of the distinct echogenic bands increased, and the elasticity index ratio decreased. CONCLUSION: Sonographic features after the third treatment session may be useful as predictors of therapeutic efficacy.

Potential clinical application of masseter and temporal muscle massage treatment using an oral rehabilitation robot in temporomandibular disorder patients with myofascial pain.

OBJECTIVES: To investigate the safety, suitable treatment regimen, and efficacy of masseter and temporal muscle massage treatment using an oral rehabilitation robot. METHODS: Forty-one temporomandibular disorder (TMD) patients with myofascial pain (8 men, 33 women, median age: 46 years) were enrolled. The safety, suitable massage regimen, and efficacy of this treatment were investigated. Changes in masseter muscle thickness were evaluated on sonograms. RESULTS: No adverse events occurred with any of the treatment sessions. Suitable massage was at pressure of 10 N for 16 minutes. Five sessions were performed every 2 weeks. Total duration of treatment was 9·5 weeks in median. Massage treatment was effective in 70·3% of patients. Masseter muscle thickness decreased with treatment in the therapy-effective group. CONCLUSION: This study confirmed the safety of massage treatment, and established a suitable regimen. Massage was effective in 70·3% of patients and appeared to have a potential as one of the effective treatments for myofascial pain.

[Improvements to an automated method for detecting carotid artery calcifications by adopting a positional feature and feature selection].

The purpose of this study was to improve an automated scheme for detecting carotid artery calcification (CAC) in dental panoramic radiographs (DPRs). Using 100 DPRs, the sensitivity of CAC detection employing our previous method was 90.0% with 5.0 false positives (FPs) per image. This study describes two enhancements. One is the adoption of a new feature for the position of CACs in addition to previous features. The other is feature selection employing the support vector machine using all combinations. Five of 12 features were selected. Using our proposed method, the average sensitivity for the same database proved to be 90.0%, with only 2.5 FPs per image. These results indicate the potential effectiveness of the new positional feature and feature selection.

Sex differences and age-related changes in the mandibular alveolar bone mineral density using a computer-aided measurement system for intraoral radiography.

This study aimed to conduct a cross-sectional data analysis of the alveolar bone mineral density (al-BMD) in 225 patients of various ages and different sexes. The al-BMD value in the mandibular incisor region was calculated using a computer-aided measurement system (DentalSCOPE) for intraoral radiography. All participants with intact teeth (101 males and 124 females; age range, 25-89 years) were divided into three age-segregated groups (25-49, 50-74, and > 75 years). Statistical differences were evaluated using the Mann-Whitney U or Kruskal-Wallis test. Males exhibited significantly greater al-BMD than females (p < 0.001). The highest means were observed in the 25-49 age group, regardless of sex (1007.90 mg/cm2 in males, 910.90 mg/cm2 in females). A 9.8% decrease in al-BMD was observed with the increase in age in males (25-49 to 50-74 years; p = 0.004); however, no further changes were seen thereafter. In females, a decreasing trend was seen throughout the lifespan, with values reaching up to 76.0% of the initial peak value (p < 0.001). Similar to other skeletal sites, the alveolar bone exhibits sex differences and undergoes a reduction in BMD via the normal aging process.

A blurring correction method suitable to analyze quantitative x-ray images derived from energy-resolving photon counting detector.

Objective.The purpose of this study is to propose a novel blurring correction method that enables accurate quantitative analysis of the object edge when using energy-resolving photon counting detectors (ERPCDs). Although the ERPCDs have the ability to generate various quantitative analysis techniques, such as the derivations of effective atomic number (Zeff) and bone mineral density values, at the object edge in these quantitative images, accurate quantitative information cannot be obtained. This is because image blurring prevents the gathering of accurate primary x-ray attenuation information.Approach.We developed the following procedure for blurring correction. A 5 × 5 pixels masking region was set as the processing area, and the pixels affected by blurring were extracted from the analysis of pixel value distribution. The blurred pixel values were then corrected to the proper values estimated by analyzing minimum and/or maximum values in the set mask area. The suitability of our correction method was verified by a simulation study and an experiment using a prototype ERPCD.Main results. WhenZeffimage of aluminum objects (Zeff= 13) were analyzed without applying our correction method, regardless of raw data or correction data applying a conventional edge enhancement method, the properZeffvalues could not be derived for the object edge. In contrast, when applying our correction method, 82% of pixels affected by blurring were corrected and the properZeffvalues were calculated for those pixels. As a result of investigating the applicability limits of our method through simulation, it was proven that it works effectively for objects with 4 × 4 pixels or more.Significance. Our method is effective in correcting image blurring when the quantitative image is calculated based on multiple images. It will become an in-demand technology for putting a quantitative diagnosis into actual medical examinations.

External Validation of the Effect of the Combined Use of Object Detection for the Classification of the C-Shaped Canal Configuration of the Mandibular Second Molar in Panoramic Radiographs: A Multicenter Study.

INTRODUCTION: The purposes of this study were to evaluate the effect of the combined use of object detection for the classification of the C-shaped canal anatomy of the mandibular second molar in panoramic radiographs and to perform an external validation on a multicenter dataset. METHODS: The panoramic radiographs of 805 patients were collected from 4 institutes across two countries. The CBCT data of the same patients were used as "Ground-truth". Five datasets were generated: one for training and validation, and 4 as external validation datasets. Workflow 1 used manual cropping to prepare the image patches of mandibular second molars, and then classification was performed using EfficientNet. Workflow 2 used two combined methods with a preceding object detection (YOLOv7) performed for automated image patch formation, followed by classification using EfficientNet. Workflow 3 directly classified the root canal anatomy from the panoramic radiographs using the YOLOv7 prediction outcomes. The classification performance of the 3 workflows was evaluated and compared across 4 external validation datasets. RESULTS: For Workflows 1, 2, and 3, the area under the receiver operating characteristic curve (AUC) values were 0.863, 0.861, and 0.876, respectively, for the AGU dataset; 0.935, 0.945, and 0.863, respectively, for the ASU dataset; 0.854, 0.857, and 0.849, respectively, for the ODU dataset; and 0.821, 0.797, and 0.831, respectively, for the ODU low-resolution dataset. No significant differences existed between the AUC values of Workflows 1, 2, and 3 across the 4 datasets. CONCLUSIONS: The deep learning systems of the 3 workflows achieved significant accuracy in predicting the C-shaped canal in mandibular second molars across all test datasets.

Deep learning and artificial intelligence in dental diagnostic imaging.

The application of artificial intelligence (AI) based on deep learning in dental diagnostic imaging is increasing. Several popular deep learning tasks have been applied to dental diagnostic images. Classification tasks are used to classify images with and without positive abnormal findings or to evaluate the progress of a lesion based on imaging findings. Region (object) detection and segmentation tasks have been used for tooth identification in panoramic radiographs. This technique is useful for automatically creating a patient's dental chart. Deep learning methods can also be used for detecting and evaluating anatomical structures of interest from images. Furthermore, generative AI based on natural language processing can automatically create written reports from the findings of diagnostic imaging.

Detection of aspiration from images of a videofluoroscopic swallowing study adopting deep learning.

OBJECTIVES: A videofluoroscopic swallowing study (VFSS) is conducted to detect aspiration. However, aspiration occurs within a short time and is difficult to detect. If deep learning can detect aspirations with high accuracy, clinicians can focus on the diagnosis of the detected aspirations. Whether VFSS aspirations can be classified using rapid-prototyping deep-learning tools was studied. METHODS: VFSS videos were separated into individual image frames. A region of interest was defined on the pharynx. Three convolutional neural networks (CNNs), namely a Simple-Layer CNN, Multiple-Layer CNN, and Modified LeNet, were designed for the classification. The performance results of the CNNs were compared in terms of the areas under their receiver-operating characteristic curves (AUCs). RESULTS: A total of 18,333 images obtained through data augmentation were selected for the evaluation. The different CNNs yielded sensitivities of 78.8%-87.6%, specificities of 91.9%-98.1%, and overall accuracies of 85.8%-91.7%. The AUC of 0.974 obtained for the Simple-Layer CNN and Modified LeNet was significantly higher than that obtained for the Multiple-Layer CNN (AUC of 0.936) (p < 0.001). CONCLUSIONS: The results of this study show that deep learning has potential for detecting aspiration with high accuracy.

Deep learning for preliminary profiling of panoramic images.

OBJECTIVE: This study explored the feasibility of using deep learning for profiling of panoramic radiographs. STUDY DESIGN: Panoramic radiographs of 1000 patients were used. Patients were categorized using seven dental or physical characteristics: age, gender, mixed or permanent dentition, number of presenting teeth, impacted wisdom tooth status, implant status, and prosthetic treatment status. A Neural Network Console (Sony Network Communications Inc., Tokyo, Japan) deep learning system and the VGG-Net deep convolutional neural network were used for classification. RESULTS: Dentition and prosthetic treatment status exhibited classification accuracies of 93.5% and 90.5%, respectively. Tooth number and implant status both exhibited 89.5% classification accuracy; impacted wisdom tooth status exhibited 69.0% classification accuracy. Age and gender exhibited classification accuracies of 56.0% and 75.5%, respectively. CONCLUSION: Our proposed preliminary profiling method may be useful for preliminary interpretation of panoramic images and preprocessing before the application of additional artificial intelligence techniques.

Variance of radiographical alveolar bone mineral density by the anatomical morphology of mandibular bone.

Purpose: Evaluating the bone mineral density (BMD) of alveolar bone is useful for dental treatments. The DentalSCOPE is an image analysis system developed to evaluate the BMD of alveolar bone. The aim of this study was to evaluate the relationship between cross-sectional anatomical size and BMD value. Materials and methods: Thirty-four subjects (adult dental patients and volunteers) participated in this study. Intraoral radiographs of the mandibular molar region were acquired. Using DentalSCOPE software, three to four line-shaped regions of interest (ROIs) were obtained in the alveolar septum region. Cross-sectional CT images of mandible at the same position to above mentioned line-shaped ROI was reconstructed from subject's dental CBCT images. The measurements were performed using cross-sectional CT images and compared with BMD value. Results and discussion: For stepwise multiple linear regression analysis, the buccal-lingual width of the mandibular body (mandible width) and the CT value of the cancellous bone were adopted as explanatory variables that affected the BMD of the mandible. The BMD value increased by 20 mg/mm2 when the mandible width increased by 1 mm, and the BMD value increased by 5 mg/mm2 when the CT value of the cancellous bone increased by 1%. Conclusion: In the clinical application of alveolar bone BMD, the effect of the anatomical morphology of alveolar bone should be taken into consideration.

Automatic visualization of the mandibular canal in relation to an impacted mandibular third molar on panoramic radiographs using deep learning segmentation and transfer learning techniques.

OBJECTIVE: The aim of this study was to create and assess a deep learning model using segmentation and transfer learning methods to visualize the proximity of the mandibular canal to an impacted third molar on panoramic radiographs. STUDY DESIGN: The panoramic radiographs containing the mandibular canal and impacted third molar were collected from 2 hospitals (Hospitals A and B). A total of 3200 areas were used for creating and evaluating learning models. A source model was created using the data from Hospital A, simulatively transferred to Hospital B, and trained using various amounts of data from Hospital B to create target models. The same data were then applied to the target models to calculate the Dice coefficient, Jaccard index, and sensitivity. RESULTS: The performance of target models trained using 200 or more data sets was equivalent to that of the source model tested using data obtained from the same hospital (Hospital A). CONCLUSIONS: Sufficiently qualified models could delineate the mandibular canal in relation to an impacted third molar on panoramic radiographs using a segmentation technique. Transfer learning appears to be an effective method for creating such models using a relatively small number of data sets.

A preliminary deep learning study on automatic segmentation of contrast-enhanced bolus in videofluorography of swallowing.

Although videofluorography (VFG) is an effective tool for evaluating swallowing functions, its accurate evaluation requires considerable time and effort. This study aimed to create a deep learning model for automated bolus segmentation on VFG images of patients with healthy swallowing and dysphagia using the artificial intelligence deep learning segmentation method, and to assess the performance of the method. VFG images of 72 swallowing of 12 patients were continuously converted into 15 static images per second. In total, 3910 images were arbitrarily assigned to the training, validation, test 1, and test 2 datasets. In the training and validation datasets, images of colored bolus areas were prepared, along with original images. Using a U-Net neural network, a trained model was created after 500 epochs of training. The test datasets were applied to the trained model, and the performances of automatic segmentation (Jaccard index, Sørensen-Dice coefficient, and sensitivity) were calculated. All performance values for the segmentation of the test 1 and 2 datasets were high, exceeding 0.9. Using an artificial intelligence deep learning segmentation method, we automatically segmented the bolus areas on VFG images; our method exhibited high performance. This model also allowed assessment of aspiration and laryngeal invasion.

Automatic segmentation of the temporomandibular joint disc on magnetic resonance images using a deep learning technique.

OBJECTIVES: The aims of the present study were to construct a deep learning model for automatic segmentation of the temporomandibular joint (TMJ) disc on magnetic resonance (MR) images, and to evaluate the performances using the internal and external test data. METHODS: In total, 1200 MR images of closed and open mouth positions in patients with temporomandibular disorder (TMD) were collected from two hospitals (Hospitals A and B). The training and validation data comprised 1000 images from Hospital A, which were used to create a segmentation model. The performance was evaluated using 200 images from Hospital A (internal validity test) and 200 images from Hospital B (external validity test). RESULTS: Although the analysis of performance determined with data from Hospital B showed low recall (sensitivity), compared with the performance determined with data from Hospital A, both performances were above 80%. Precision (positive predictive value) was lower when test data from Hospital A were used for the position of anterior disc displacement. According to the intra-articular TMD classification, the proportions of accurately assigned TMJs were higher when using images from Hospital A than when using images from Hospital B. CONCLUSION: The segmentation deep learning model created in this study may be useful for identifying disc positions on MR images.

Performance of deep learning technology for evaluation of positioning quality in periapical radiography of the maxillary canine.

OBJECTIVES: The aim of the present study was to create and test an automatic system for assessing the technical quality of positioning in periapical radiography of the maxillary canines using deep learning classification and segmentation techniques. METHODS: We created and tested two deep learning systems using 500 periapical radiographs (250 each of good- and bad-quality images). We assigned 350, 70, and 80 images as the training, validation, and test datasets, respectively. The learning model of system 1 was created with only the classification process, whereas system 2 consisted of both the segmentation and classification models. In each model, 500 epochs of training were performed using AlexNet and U-net for classification and segmentation, respectively. The segmentation results were evaluated by the intersection over union method, with values of 0.6 or more considered as success. The classification results were compared between the two systems. RESULTS: The segmentation performance of system 2 was recall, precision, and F measure of 0.937, 0.961, and 0.949, respectively. System 2 showed better classification performance values than those obtained by system 1. The area under the receiver operating characteristic curve values differed significantly between system 1 (0.649) and system 2 (0.927). CONCLUSIONS: The deep learning systems we created appeared to have potential benefits in evaluation of the technical positioning quality of periapical radiographs through the use of segmentation and classification functions.

Tooth recognition of 32 tooth types by branched single shot multibox detector and integration processing in panoramic radiographs.

Purpose: The purpose of our study was to analyze dental panoramic radiographs and contribute to dentists' diagnosis by automatically extracting the information necessary for reading them. As the initial step, we detected teeth and classified their tooth types in this study. Approach: We propose single-shot multibox detector (SSD) networks with a side branch for 1-class detection without distinguishing the tooth type and for 16-class detection (i.e., the central incisor, lateral incisor, canine, first premolar, second premolar, first molar, second molar, and third molar, distinguished by the upper and lower jaws). In addition, post-processing was conducted to integrate the results of the two networks and categorize them into 32 classes, differentiating between the left and right teeth. The proposed method was applied to 950 dental panoramic radiographs obtained at multiple facilities, including a university hospital and dental clinics. Results: The recognition performance of the SSD with a side branch was better than that of the original SSD. In addition, the detection rate was improved by the integration process. As a result, the detection rate was 99.03%, the number of false detections was 0.29 per image, and the classification rate was 96.79% for 32 tooth types. Conclusions: We propose a method for tooth recognition using object detection and post-processing. The results show the effectiveness of network branching on the recognition performance and the usefulness of post-processing for neural network output.

Features of mandibular cortical bone morphology in osteoporotic fracture patients.

PURPOSE: The diagnostic criteria for osteoporosis are based on the bone mineral density (BMD) level in the lumbar spine and femur bone. Patients with osteoporotic fractures were diagnosed with osteoporosis. While systemic BMD and mandibular cortical bone morphology are correlated, this has not been studied in patients with a history of osteoporotic fractures. Therefore, purpose of this study was researching the mandibular cortical bone morphology in patients with osteoporotic fractures. METHODS: The subjects were 55 female and 20 male patients with osteoporotic fractures. Patients were divided into 30 primary osteoporosis patients and 45 secondary osteoporosis patients according to the medical history. Patients underwent BMD and panoramic radiography examinations during orthopedic treatment for fractures. A dual-energy X-ray absorptiometry system was used to measure BMD. Mandibular cortex index (MCI) and mandibular cortex width (MCW) were evaluated using machine-learning measurement software. RESULTS: In the analysis of MCI, the ratio of class 2 and 3 was 73% of both primary osteoporosis and secondary osteoporosis. The average MCW was 2.19 mm for primary osteoporosis and 2.30 mm for secondary osteoporosis. The sensitivity values by MCI and MCW were 73% and 76% for both primary and secondary osteoporosis, which were similar detection powers. In addition, the false-negative rates by MCI and MCW were 27% and 24%. CONCLUSION: We suggested that MCI and MCW are indicators of osteoporotic conditions in patients with primary and secondary osteoporosis. Our results show that MCI and MCW are non-inferior to the sensitivity values for lumbar BMD in patients with osteoporotic fractures.

Videofluorographic evaluation of mastication and swallowing of Japanese udon noodles and white rice.

A videofluorographic (VF) swallowing study was performed on 22 healthy volunteers to observe the complete mastication and swallowing phases for Japanese udon noodles and white rice. The hardness, stickiness, and cohesiveness of food samples were measured using a food texture analyzing system. VF images were acquired using a versatile fluoroscopic unit and barium sulfate was used as a contrast medium. Udon noodles had a harder and smoother food texture than white rice. Fewer chewing movements and more stage 2 transport were seen during the consumption of udon noodles than for white rice.

A novel algorithm for extracting soft-tissue and bone images measured using a photon-counting type X-ray imaging detector with the help of effective atomic number analysis.

Most of the objects targeted for X-ray examination are composed of soft-tissue and bone. We aimed to develop an algorithm for generating X-ray images which can give quantitative information of soft-tissue and bone using an energy-resolving photon-counting type imaging detector. We used polychromatic X-rays for analysis in which both the beam hardening effect and detector response were properly corrected and then succeeded in virtually treating the amount of measured X-ray attenuation as if it were measured using monochromatic X-rays.