Localization of cementoenamel junction in intraoral ultrasonographs with machine learning

Accuracy of intraoral ultrasound to evaluate alveolar bone level: an ex vivo study in human cadavers

OBJECTIVES

This study aimed to evaluate the reliability and accuracy of an intraoral ultrasound (US) device to evaluate alveolar bone by comparing it between different raters and to microCT (µCT) measurements.

METHODS

38 teeth distributed across three human cadavers were prepared by placing two notches on the facial enamel surface. The maxillary and mandibular teeth were imaged with a custom-designed intraoral 20 MHz ultrasound and µCT with 0.03 mm voxel size. µCT was considered the reference standard for this study. For each sample, the distance from the inferior border of the most apical notch to the tip of the alveolar bone crest on the facial aspect of the teeth was measured from the US and µCT images. Intraclass correlation coefficient (ICC) and standard deviation were calculated.

RESULTS

The intra-examiner and inter-examiner reliability for both the µCT and US alveolar bone measurements were found to be excellent (intra-examiner ICC was 0.998 for µCT and 0.997 for US, inter-examiner ICC was 0.996 for µCT and between 0.947 and 0.950 for US). The accuracy of the US was found to be good compared to µCT (ICC between 0.885 and 0.894).

CONCLUSION

The study demonstrated that intraoral ultrasound is highly reliable and accurate compared to the µCT reference standard for evaluating facial alveolar bone height.

Intraoral Ultrasound Imaging Using a Rotational Transducer with Periodontal Feature Identification by Machine Learning

Innovative intraoral ultrasound devices with smart artificial intelligence-based identification for dento-anatomy could provide crucial information for oral health diagnosis and treatment and shed light on real-time detection of developmental dentistry. However, the grand challenge is that the current ultrasound technologies are meant for external use due to their bulkiness and low frequency. We report a compact versatile ultrasound intraoral device that consists of a rotational probe head robustly pivoted around a hand-held and portable handle for real-time imaging of intraoral anatomy using high-frequency ultrasonography (up to 25 MHz). The intraoral ultrasound device that could be adjusted for various orientations of the imaging planes by rotating the head provides real-time, high-resolution ultrasonograms of intraoral structures, including dento-periodontium of most tooth types and maxillary palate. Machine learning-based algorithms are integrated to automate the identification of important structures, including alveolar bone and cementum-enamel junction. The intraoral ultrasound device smartened with artificial intelligence could innovate oral health diagnosis and treatment plans toward precision health and patient care.

Nonionizing diagnostic imaging modalities for visualizing health and pathology of periodontal and peri-implant tissues

Radiographic examination has been an essential part of the diagnostic workflow in periodontology and implant dentistry. However, radiographic examination unavoidably involves ionizing radiation and its associated risks. Clinicians and researchers have invested considerable efforts in assessing the feasibility and capability of utilizing nonionizing imaging modalities to replace traditional radiographic imaging. Two such modalities have been extensively evaluated in clinical settings, namely, ultrasonography (USG) and magnetic resonance imaging (MRI). Another modality, optical coherence tomography (OCT), has been under investigation more recently. This review aims to provide an overview of the literature and summarize the usage of USG, MRI, and OCT in evaluating health and pathology of periodontal and peri-implant tissues. Clinical studies have shown that USG could accurately measure gingival height and crestal bone level, and classify furcation involvement. Due to physical constraints, USG may be more applicable to the buccal surfaces of the dentition even with an intra-oral probe. Clinical studies have also shown that MRI could visualize the degree of soft-tissue inflammation and osseous edema, the extent of bone loss at furcation involvement sites, and periodontal bone level. However, there was a lack of clinical studies on the evaluation of peri-implant tissues by MRI. Moreover, an MRI machine is very expensive, occupies much space, and requires more time than cone-beam computed tomography (CBCT) or intraoral radiographs to complete a scan. The feasibility of OCT to evaluate periodontal and peri-implant tissues remains to be elucidated, as there are only preclinical studies at the moment. A major shortcoming of OCT is that it may not reach the bottom of the periodontal pocket, particularly for inflammatory conditions, due to the absorption of near-infrared light by hemoglobin. Until future technological breakthroughs finally overcome the limitations of USG, MRI and OCT, the practical imaging modalities for routine diagnostics of periodontal and peri-implant tissues remain to be plain radiographs and CBCTs.

Effects of occlusal wear on 16-year progression of non-carious cervical lesions-Results of the study of health in Pomerania (SHIP-START)

BACKGROUND

It is still discussed whether occlusal wear (OW) affects the formation of non-carious cervical lesions (NCCLs).

OBJECTIVE

To estimate effects of OW on the presence and development of NCCLs, using 16-year follow-up data from a cohort study.

METHODS

Occlusal and cervical defects were measured in 728 cast models (one from the upper jaw and one from the lower jaw) of 364 participants. Adjusted mixed-effects ordinal logistic models analysing estimated cross-sectional (N = 1308 teeth/291 subjects) and longitudinal (N = 718 teeth/226 subjects) associations of OW with NCCLs using tooth level data.

RESULTS

OW size was cross-sectionally (OR = 1.74; 95% CI: 1.27-2.38 for OW size; OR = 0.97; 95% CI: 0.94-0.99 for squared OW size), but not longitudinally (OR = 1.14; 95% CI: 0.99-1.30) associated with odds of higher NCCL sizes. For cross-sectional analyses, predicted probabilities of an NCCL size of 0 decreased from about 0.996 to 0.010 for OW sizes of 0 to 25.

CONCLUSION

Results suggest an association between OW and NCCL size. However, as longitudinal results were non-significant, while consistent in direction, large-scaled cohort studies are demanded to more precisely estimate effect strength.

The feasibility of ultrasonography for the measurement of periodontal and peri-implant phenotype: A systematic review and meta-analysis

BACKGROUND

Screening ultrasonography was proposed for monitoring periodontal soft tissues in the early 1960s, owing to its nonionizing, real-time, and cost-effective properties. Studies have provided convincing preliminary evidence for the use of ultrasound (US) in implant dentistry.

PURPOSE

To assess the feasibility of ultrasonography (US) for measuring the buccal thickness of periodontal and peri-implant tissues. The secondary objective was to evaluate the reliability of US measurements compared to classic techniques, such as CBCT and directly measurements.

MATERIALS AND METHODS

An electronic literature search was conducted by three independent reviewers through February 2023. The inclusion criteria were articles investigating at least five patients/cadavers with US measurements in periodontal or peri-implant buccal tissues. Compliance with methodological reporting standards and risk of bias was assessed using EULAR and QUADAS-C tools, respectively. Random-effects meta-analysis was conducted, using Bland-Altman analysis. Certainty of the evidence was assessed using GRADE.

RESULTS

The final selection included 12 studies examining 458 patients and 13 cadavers, with a total of 226 implants, 1958 teeth and 60 edentulous sites. The body of evidence was assessed as partially compliant with methodological reporting standards for US studies and had an unclear to high risk of bias. Meta-analysis of five comparative studies showed no evidence of clinically significant bias between US and direct measurements (very low certainty), and between US and CBCT (very low certainty) for soft-tissue thickness. Likewise, for bone thickness, there is no evidence of clinically significant bias between US and CBCT (low certainty).

CONCLUSIONS

Compared to the CBCT and direct measurements, ultrasonography might be a reliable approach for monitoring on periodontal and peri-implant phenotype. However, there is uncertainty about estimates of the actual effect, so further standardized and larger sample size of clinical research is needed.

Overview of Ultrasound in Dentistry for Advancing Research Methodology and Patient Care Quality with Emphasis on Periodontal/Peri-implant Applications

BACKGROUND

Ultrasound is a non-invasive, cross-sectional imaging technique emerging in dentistry. It is an adjunct tool for diagnosing pathologies in the oral cavity that overcomes some limitations of current methodologies, including direct clinical examination, 2D radiographs, and cone beam computerized tomography. Increasing demand for soft tissue imaging has led to continuous improvements on transducer miniaturization and spatial resolution. The aims of this study are (1) to create a comprehensive overview of the current literature of ultrasonic imaging relating to dentistry, and (2) to provide a view onto investigations with immediate, intermediate, and long-term impact in periodontology and implantology.

METHODS

A rapid literature review was performed using two broad searches conducted in the PubMed database, yielding 576 and 757 citations, respectively. A rating was established within a citation software (EndNote) using a 5-star classification. The broad search with 757 citations allowed for high sensitivity whereas the subsequent rating added specificity.

RESULTS

A critical review of the clinical applications of ultrasound in dentistry was provided with a focus on applications in periodontology and implantology. The role of ultrasound as a developing dental diagnostic tool was reviewed. Specific uses such as soft and hard tissue imaging, longitudinal monitoring, as well as anatomic and physiological evaluation were discussed.

CONCLUSIONS

Future efforts should be directed towards the transition of ultrasonography from a research tool to a clinical tool. Moreover, a dedicated effort is needed to introduce ultrasonic imaging to dental education and the dental community to ultimately improve the quality of patient care.

Estimating Crestal Thickness of Alveolar Bones on Intra-oral Ultrasonograms

OBJECTIVE

Alveolar crestal bone thickness and level provide important diagnostic and prognostic information for orthodontic treatment, periodontal disease management and dental implants. Ionizing radiation-free ultrasound has emerged as a promising clinical tool in imaging oral tissues. However, the ultrasound image is distorted when the wave speed of the tissue of interest is different from the mapping speed of the scanner and, therefore, the subsequent dimension measurements are not accurate. This study was aimed at deriving a correction factor that can be applied to the measurements to correct for discrepancy caused by speed variation.

METHODS

The factor is a function of the speed ratio and the acute angle that the segment of interest makes with the beam axis perpendicular to the transducer. The phantom and cadaver experiments were designed to validate the method.

DISCUSSION

The comparisons agree well with absolute errors not more than 4.9%. Dimension measurements on ultrasonographs can be properly corrected by applying the correction factor without recourse to the raw signals.

CONCLUSION

The correction factor has reduced the measurement discrepancy on the acquired ultrasonographs for the tissue whose speed is different from the scanner's mapping speed.

Ultrasound Imaging of the Periodontium Complex: A Reliability Study

Background

Ultrasonography is a noninvasive, low-cost diagnostic tool widely used in medicine. Recent studies have demonstrated that ultrasound imaging might have the potential to be used intraorally to assess periodontal biomarkers.

Objectives

To evaluate the reliability of interlandmark distance measurements on intraoral ultrasound images of the periodontal tissues.

Materials and Methods

Sixty-four patients from the graduate periodontics (n = 33) and orthodontics (n = 31) clinics were recruited. A 20 MHz handheld intraoral ultrasound transducer was used to scan maxillary and mandibular incisors, canines, and premolars. Distances between the alveolar bone crest and cementoenamel junction (ABC-CEJ), gingival thickness (GT), and alveolar bone thickness (ABT) were measured by 3 raters. The intercorrelation coefficient (ICC) and mean absolute deviation (MAD) were calculated among and between the raters. Raters also scored images according to quality.

Results

The ICC scores for intrarater reliability were 0.940 (0.932-0.947), 0.953 (0.945-0.961), and 0.859 (0.841-0.876) for ABC-CEJ, GT, and ABT, respectively. The intrarater MAD values were 0.023 (±0.019) mm, 0.014 (±0.005) mm, and 0.005 (±0.003) mm, respectively. The ICC scores for interrater reliability were 0.872 (95% CI: 0.836-0.901), 0.958 (95% CI: 0.946-0.968), and 0.836 (95% CI: 0.789-0.873) for ABC-CEJ, GT, and ABT, respectively. The interrater MAD values were 0.063 (±0.029) mm, 0.023 (±0.018) mm, and 0.027 (±0.012) mm, respectively.

Conclusions

The present study showed the high reliability of ultrasound in both intrarater and interrater assessments. Results suggest there might be a potential use of intraoral ultrasound to assess periodontium.

A miniaturized ultrasound transducer for monitoring full-mouth oral health: a preliminary study

OBJECTIVE

To customize a miniaturized ultrasound transducer to access full-mouth B-mode, color Doppler, and spectral Doppler imaging for monitoring oral health.

METHODS

A customized periodontal ultrasound transducer SS-19-128 (19 MHz, 128 channels) 1.8-cm wide and 1-cm thick was developed and connected to a data acquisition (DAQ) system. B-mode, color Doppler, and spectral Doppler data could all be collected with SS-19-128. The imaging resolution and penetration capacity of SS-19-128 were characterized on phantoms. The gingival thickness was measured on 11 swine teeth by SS-19-128 for comparison with conventional transgingival probing via Bland-Altman analysis and Pearson correlation. Five human subjects were then recruited to demonstrate B-mode and Doppler imaging by SS-19-128.

RESULTS

The axial and lateral spatial resolution at 5.5 mm depth is 102.1 µm and 142.9 µm, respectively. The penetration depth in a tissue-mimicking phantom is over 30 mm. In vivo B-mode imaging of all 28 teeth was demonstrated on one human subject, and imaging of tooth #18 was accessed on five human subjects. Gingival thickness measurement compared with transgingival probing showed a bias of -0.015 mm and SD of 0.031 mm, and a r = 0.9235 (p < 0.0001) correlation. In vivo color and spectral Doppler imaging of the supraperiosteal artery in human gingiva was performed to generate hemodynamic information.

CONCLUSIONS

The small size of SS-19-128 offers important advantages over existing ultrasound technology-more specifically, whole-mouth scanning/charting reminiscent of radiography. This is nearly a two-fold increase in the number of teeth that can be assessed versus conventional transducers.

Posterior photoacoustic/ultrasound imaging of the periodontal pocket with a compact intraoral transducer

Periodontitis is a public issue and imaging periodontal pocket is important to evaluate periodontitis. Regular linear transducers have limitations in imaging the posterior teeth due to their geometry restrictions. Here we characterized a transducer that can image the posterior teeth including assessment of periodontal pockets via a combination of photoacoustic and ultrasound imaging. Unlike conventional transducer design, this device has a toothbrush-shaped form factor with a side-view transducer to image molars (total size: 1 ×1.9 cm). A laser diode was integrated as the light source to reduce the cost and size and facilitates clinical transition. The in vivo imaging of a molar of a periodontal patient demonstrated that the transducer could image in the posterior area of gum in vivo; the value determined by imaging was within 7 % of the value measured clinically.

Automatic Segmentation of Periodontal Tissue Ultrasound Images with Artificial Intelligence: A Novel Method for Improving Dataset Quality

This research aimed to evaluate Mask R-CNN and U-Net convolutional neural network models for pixel-level classification in order to perform the automatic segmentation of bi-dimensional images of US dental arches, identifying anatomical elements required for periodontal diagnosis. A secondary aim was to evaluate the efficiency of a correction method of the ground truth masks segmented by an operator, for improving the quality of the datasets used for training the neural network models, by 3D ultrasound reconstructions of the examined periodontal tissue.

METHODS

Ultrasound periodontal investigations were performed for 52 teeth of 11 patients using a 3D ultrasound scanner prototype. The original ultrasound images were segmented by a low experienced operator using region growing-based segmentation algorithms. Three-dimensional ultrasound reconstructions were used for the quality check and correction of the segmentation. Mask R-CNN and U-NET were trained and used for prediction of periodontal tissue's elements identification.

RESULTS

The average Intersection over Union ranged between 10% for the periodontal pocket and 75.6% for gingiva. Even though the original dataset contained 3417 images from 11 patients, and the corrected dataset only 2135 images from 5 patients, the prediction's accuracy is significantly better for the models trained with the corrected dataset.

CONCLUSIONS

The proposed quality check and correction method by evaluating in the 3D space the operator's ground truth segmentation had a positive impact on the quality of the datasets demonstrated through higher IoU after retraining the models using the corrected dataset.

High-resolution ultrasonography of gingival biomarkers for periodontal diagnosis in healthy and diseased subjects

OBJECTIVE

To determine the capacity of ultrasonographic image-based measurements of gingival height and alveolar bone level for monitoring periodontal health and disease.

METHODS

Sixteen subjects were recruited from patients scheduled to receive dental care and classified as periodontally healthy (n = 10) or diseased (n = 6) according to clinical guidelines. A 40-MHz ultrasound system was used to measure gingival recession, gingival height, alveolar bone level, and gingival thickness from 66 teeth for comparison to probing measurements of pocket depth and clinical attachment level. Interexaminer variability and comparison between ultrasound measurements and probing measurements was performed via Bland-Altman analysis.

RESULTS

Gingival recession and its risk in non-recessed patients could be determined via measurement of the supra- and subgingival cementoenamel junction relative to the gingival margin. Interexaminer bias for ultrasound image analysis was negligible (<0.10 mm) for imaged gingival height (iGH) and 0.45 mm for imaged alveolar bone level (iABL). Diseased subjects had significantly higher imaging measurements (iGH, iABL) and clinical measurements (probing pocket depth, clinical attachment level) than healthy subjects (p < 0.05). Subtraction of the average biologic width from iGH resulted in 83% agreement (≤1 mm difference) between iGH and probing pocket depth measurements.

CONCLUSIONS

Ultrasonography has an equivalent diagnostic capacity as gold-standard physical probing for periodontal metrics while offering more detailed anatomical information.

Photoacoustic imaging of posterior periodontal pocket using a commercial hockey-stick transducer

SIGNIFICANCE

Photoacoustic imaging has shown advantages over the periodontal probing method in measuring the periodontal probing depth, but the large size of conventional photoacoustic transducers prevents imaging of the more posterior teeth.

AIM

Our aim is to develop a photoacoustic imaging system to image the more posterior periodontal pocket.

APPROACH

We report a clinical "hockey-stick"-style transducer integrated with fibers for periodontal photoacoustic imaging. Cuttlefish ink labeled the periodontal pocket as the photoacoustic contrast agent.

RESULTS

We characterized the imaging system and then measured the pocket depth of 35 swine teeth. Three raters evaluated the performance of the hockey-stick transducer. The measurements between the Williams probing (gold standard) and the photoacoustic methods were blinded but highly correlated. We showed a bias of ∼0.3  mm for the imaging-based technique versus Williams probing. The minimum inter-reliability was over 0.60 for three different raters of varying experience, suggesting that this approach to measure the periodontal pocket is reproducible. Finally, we imaged three pre-molars of a human subject. We could access more upper and posterior teeth than conventional linear transducers.

CONCLUSIONS

The unique angle shape of the hockey-stick transducer allows it to image more posterior teeth than regular linear transducers. This study demonstrated the ability of a hockey-stick transducer to measure the periodontal pocket via photoacoustic imaging.

A hybrid mask RCNN-based tool to localize dental cavities from real-time mixed photographic images

Nearly 3.5 billion humans have oral health issues, including dental caries, which requires dentist-patient exposure in oral examinations. The automated approaches identify and locate carious regions from dental images by localizing and processing either colored photographs or X-ray images taken via specialized dental photography cameras. The dentists' interpretation of carious regions is difficult since the detected regions are masked using solid coloring and limited to a particular dental image type. The software-based automated tools to localize caries from dental images taken via ordinary cameras requires further investigation. This research provided a mixed dataset of dental photographic (colored or X-ray) images, instantiated a deep learning approach to enhance the existing dental image carious regions' localization procedure, and implemented a full-fledged tool to present carious regions via simple dental images automatically. The instantiation mainly exploits the mixed dataset of dental images (colored photographs or X-rays) collected from multiple sources and pre-trained hybrid Mask RCNN to localize dental carious regions. The evaluations performed by the dentists showed that the correctness of annotated datasets is up to 96%, and the accuracy of the proposed system is between 78% and 92%. Moreover, the system achieved the overall satisfaction level of dentists above 80%.