Deep learning in the diagnosis for cystic lesions of the jaws: a review of recent progress.

Cystic lesions of the gnathic bones present challenges in differential diagnosis. In recent years, artificial intelligence (AI) represented by deep learning (DL) has rapidly developed and emerged in the field of dental and maxillofacial radiology (DMFR) Dental radiography provides a rich resource for the study of diagnostic analysis methods for cystic lesions of the jaws and has attracted many researchers. The aim of the current study was to investigate the diagnostic performance of DL for cystic lesions of the jaws. Online searches on Google Scholar, PubMed, and IEEE Xplore databases, up to September 2023, with subsequent manual screening for confirmation. The initial search yielded 1862 titles, and 44 studies were ultimately included. All studies used DL methods or tools for the identification of a variable number of maxillofacial cysts. The performance of algorithms with different models varies. Although most of the reviewed stu dies demonstrated that DL methods have better discriminative performance than clinicians, further development is still needed before routine clinical implementation due to several challenges and limitations such as lack of model interpretability, multicenter data validation, etc Considering the current limitations and challenges, future studies for the differential diagnosis of cystic lesions of the jaws should follow actual clinical diagnostic scenarios to coordinate study design and enhance the impact of artificial intelligence in the diagnosis of oral and maxillofacial diseases.

Quantitative assessment of condylar bone resorption using fused CBCT images: differentiating and diagnosing three distinct groups based on volume and thickness decrease.

OBJECTIVES: To investigate the accuracy of fused CBCT images in diagnosing three distinct groups of bone changes characterized by volume and thickness decrease in patients with temporomandibular joint osteoarthrosis (TMJ OA) during follow-up. METHODS: In this retrospective study, 109 patients (176 TMJs) with TMJ OA were included. Two consecutive CBCT images for the same patient were registered and fused. Then, three image sets were established: without fusion, fused 2D image, and fused 3D image. Three residents randomly and independently evaluated whether there was condylar resorption with the three image sets respectively. The samples diagnosed as condylar resorption by the expert panel were divided into three subgroups according to the volume and thickness decrease calculated after segmentation. The inter- and intraobserver agreement, receiver operating characteristic (ROC), and area under the curve (AUC) evaluated the diagnostic capability for different subgroups. RESULTS: For the volume decrease more than 50 mm3 and thickness decrease more than 1 mm groups, the AUC values for fused image sets were higher than those without fusion (p < 0.01). For the volume decrease within 50 mm3 and thickness decrease within 1 mm groups, the AUC values for fused 2D image sets were higher than the image sets without fusion (p < 0.05), but there was no significant difference between the fused 3D image sets and the image sets without fusion (p = 0.48 for volume decrease, p = 0.37 for thickness decrease). CONCLUSIONS: The fused images can improve the diagnostic accuracy and repeatability for the samples with at least 50 mm3 volume decrease or 1 mm thickness decrease compared with the image groups without fusion.

Relationship between development of the condylar cortex and the changes in condyle morphology: a cone-beam computed tomography (CBCT) observational study.

Background: A correct understanding of the mandibular condyle morphology may help clinicians judge the normal range of morphological variations of asymptomatic patients or the pathological conditions correctly. Hence, the aim of the present study was to evaluate the status of condyle cortication and condyle morphology, and to investigate the relationship between the development of the condylar cortex and the changes of condyle morphology. Methods: The present study was an observational study. A total of 1,010 temporomandibular joint (TMJ) cone-beam computed tomography (CBCT) images were collected retrospectively. The mandibular condyle morphology was observed in axial (concave, convex, plane and others for anterior and posterior facets), coronal (plane, convex, angled and round for superior facet) and sagittal (round and plane for superior facet) views, and the condylar cortication was grouped into three types (undeveloped, developing and developed). Analytical statistics were performed to detect a relationship between the cortication status and morphology of the condyles. Results: For males and females, the mean age was 15.11±2.71 and 14.25±2.60 years (for condylar bone without cortication), 19.45±3.92 and 18.65±3.45 years (with developing cortical bone), 23.63±3.36 and 23.86±3.73 years (with developed cortical bone), respectively. The condyle morphology with a plane form in the anterior aspect, a convex form in the posterior aspect, a convex form in the coronal view and a round form in the sagittal view was the most often recorded condyle morphologies (13.2%). After the cortical bone of condyle completely forms, the plane form was significantly increased in the superior surface in both sagittal and coronal views. Conclusions: The condylar shape gradually changes with growth and development of the condyle bone cortex. The more mature the bone cortex is, the higher the probability that the condyle will have an uneven shape, which may mean that the condyle morphology may change due to remodeling during growth and development.

Diagnostic accuracy of fused CBCT images in the evaluation of temporomandibular joint condylar bone resorption.

OBJECTIVES: To evaluate the diagnostic accuracy of fused CBCT images for patients with condylar bone resorption of temporomandibular joint (TMJ) osteoarthrosis. MATERIALS AND METHODS: Forty-two TMJs from twenty-one patients were included. Bone resorption of condyles evaluated by three experts was used as the reference standard. Three oral and maxillofacial radiology residents evaluated the resorption of condyles with a five-point scale for the four sets of images (two consecutive CBCT images without fusion, fused 2D cross-sectional images, fused 3D images, and combining fused 2D cross-sectional images and fused 3D images) randomly and independently. Each set of images was evaluated at least 1 week apart, and a second evaluation was performed 4 weeks later. Intraclass correlation coefficients were calculated to assess the intra- and inter-observer agreement. The areas under the ROC curves (AUCs) were compared among the four image sets using the Z test. RESULTS: Twenty-four TMJs were determined as condylar bone resorption, and eighteen were determined as no obvious change. The average AUC values from the three observers for the three fused image sets (0.94, 0.93, 0.93) were significantly higher than the image set without fusion (p < 0.01). The intra- and inter-observer agreement on the three fused image sets (0.70-0.89, 0.91-0.92) was higher than the image set without fusion (0.37-0.63, 0.75). CONCLUSIONS: Fused CBCT images of TMJ osteoarthrosis patients can intuitively display the condylar bone resorption and significantly improve the diagnostic accuracy. CLINICAL RELEVANCE: Fused CBCT images can help clinicians intuitively observe bone changes of the condyle in TMJ osteoarthrosis patients.

Extracapsular dissection by the sternocleidomastoid muscle-parotid space approach reduces the risks of postparotidectomy sialocele and salivary fistula.

BACKGROUND: Postoperative sialoceles and fistulas are frequent surgical complications of parotid tumor resection. Extracapsular dissection by the sternocleidomastoid muscle-parotid space approach (ECD-SMPSA) is a minimally invasive technique. To our knowledge, the characteristics of sialoceles and fistulas secondary to ECD-SMPSA have not been reported. METHODS: This prospective study enrolled 52 patients who underwent ECD-SMPSA without sialocele/fistula prevention measures. Postoperative sialoceles and fistulas were evaluated during 2 months of follow-up. RESULTS: Among the 52 patients, only one male patient developed a mild sialocele. No salivary fistulas occurred. The overall rate of sialocele/fistula formation was 1.92%. CONCLUSIONS: When treating clinically benign tumors that involve the sternocleidomastoid muscle-parotid space, ECD-SMPSA may prevent postoperative formation of sialoceles and salivary fistulas.

Diagnostic efficacy of CBCT, MRI and CBCT-MRI fused images in determining anterior disc displacement and bone changes of temporomandibular joint.

OBJECTIVES: To evaluate the diagnostic efficacy of CBCT-MRI fused image for anterior disc displacement and bone changes of temporomandibular joint (TMJ), which are the main imaging manifestations of temporomandibular disorders (TMD). METHODS: Two hundred and thirty-one TMJs of 120 patients who were diagnosed with TMD were selected for the study. The anterior disc displacement, bone defect and bone hyperplasia evaluated by three experts were used as a reference standard. Three residents individually evaluated all the three sets of images, which were CBCT images, MRI images and CBCT-MRI fused images from individual CBCT and MRI images in a random order for the above-mentioned three imaging manifestations with a five-point scale. Each set of images was observed at least 1 week apart. A second evaluation was performed 4 weeks later. Intra- and interobserver agreements were assessed using the intraclass correlation coefficient (ICC). The areas under the ROC curves (AUCs) of the three image sets were compared with a Z test, and p < 0.05 was considered statistically significant. RESULTS: One hundred and forty-five cases were determined as anterior disc displacement, 84 cases as bone defect and 40 cases as bone hyperplasia. The intra- and interobserver agreements in the CBCT-MRI fused image set (0.76-0.91) were good to excellent, and the diagnostic accuracy for bone changes was significantly higher than that of MRI image set (p＜0.05). CONCLUSIONS: CBCT-MRI fused images can display the disc and surrounding bone structures simultaneously and significantly improve the observers' reliability and diagnostic accuracy, especially for inexperienced residents.

Application of Metabolomics to the Discovery of Biomarkers for Ischemic Stroke in the Murine Model: a Comparison with the Clinical Results.

Ischemic stroke (IS) is a major cause of mortality and disability worldwide. However, the pathogenesis of IS remains unknown, and methods for early prediction and diagnosis of IS are lacking. Metabolomics can be applied to biomarker discovery and mechanism exploration of IS by exploring metabolic alterations. In this review, 62 IS metabolomics studies in the murine model published from January 2006 to December 2020 in the PubMed and Web of Science databases were systematically reviewed. Twenty metabolites (e.g., lysine, phenylalanine, methionine, tryptophan, leucine, lactate, serine, N-acetyl-aspartic acid, and glutathione) were reported consistently in more than two-third murine studies. The disturbance of metabolic pathways, such as arginine biosynthesis; alanine, aspartate and glutamate metabolism; aminoacyl-tRNA biosynthesis; and citrate cycle, may be implicated in the development of IS by influencing the biological processes such as energy failure, oxidative stress, apoptosis, and glutamate toxicity. The transient middle cerebral artery occlusion model and permanent middle cerebral artery occlusion model exhibit both common and distinct metabolic patterns. Furthermore, five metabolites (proline, serine, LysoPC (16:0), uric acid, glutamate) in the blood sample and 7 metabolic pathways (e.g., alanine, aspartate, and glutamate metabolism) are shared in animal and clinical studies. The potential biomarkers and related pathways of IS in the murine model may facilitate the biomarker discovery for early diagnosis of IS and the development of novel therapeutic targets.

Diagnostic efficacy of CBCT, MRI, and CBCT-MRI fused images in distinguishing articular disc calcification from loose body of temporomandibular joint.

OBJECTIVES: To evaluate the diagnostic efficacy of CBCT-MRI fused images for articular disc calcification of temporomandibular joint (TMJ). MATERIALS AND METHODS: Twenty patients (24 TMJs) whose image examinations showed dense bodies in the TMJ space were included in the study. The locations of dense bodies evaluated by the three experts were used as a reference standard. Three oral and maxillofacial radiology residents evaluated whether the dense bodies were disc calcification or not, with a five-point scale for four sets of images (CBCT alone, MRI alone, both CBCT and MRI observed at a time, and CBCT-MRI fused images) randomly and independently. Each set of images was observed at least 1 week apart. A second evaluation was performed after 4 weeks. Intraclass correlation coefficients were calculated to assess the intra- and inter-observer agreement. The areas under the ROC curves (AUCs) were compared between the four image sets using Z test. RESULTS: Ten cases were determined as articular disc calcifications, and fourteen cases were recognized as loose bodies in the TMJ spaces. The average AUC index for the CBCT-MRI fused images was 0.95 and significantly higher than the other sets (p < 0.01). The intra- and inter-observer agreement in the CBCT-MRI fused images (0.90-0.91, 0.93) was excellent and higher than those in the other images. CONCLUSIONS: CBCT-MRI fused images can significantly improve the observers' reliability and accuracy in determining articular disc calcification of the TMJ. CLINICAL RELEVANCE: The multimodality image fusion is feasible in detecting articular disc calcification of the TMJ which are hard to define by CBCT or MRI alone. It can be utilized especially for inexperienced residents to shorten the learning curve and improve diagnostic accuracy.

Quantitative assessment of condyle positional changes before and after orthognathic surgery based on fused 3D images from cone beam computed tomography.

OBJECTIVES: To establish one method that can be used to quantitatively evaluate the condyle positional changes with 3D images in postoperative mandibular prognathism patients. MATERIALS AND METHODS: This is a retrospective observational study. Twenty-one patients who underwent bilateral sagittal split ramus osteotomy (BSSRO) were scanned with cone beam computed tomography (CBCT) for temporomandibular joints (TMJs) at 1 week preoperatively (T0), 1 to 2 weeks (T1), 3 months (T2), 6 months (T3), and 12 months (T4) postoperatively. The data were then grouped into T0T1, T1T2, T2T3, T3T4 and T0T1, T0T2, T0T3, and T0T4. Semi-automatic registration was conducted, and the condyle positional changes were measured in segmented 3D models. Inter- and intra-observer variability and the repeatability of registration were analyzed with paired t test; the repeated measurement analysis of variance was used for analyzing the repeatability of the marked points; the consistency of segmentation was analyzed with nonparametric test of multiple paired samples (Friedman test) and the independent-sample t test was applied to comparing changes between different periods of time. Differences were considered to be statistically significant when P < 0.05. RESULTS: In T0T1 and T1T2, the condylar position was changed greatly. In T2T3, the mean condylar translations were less than 0.2 mm in all directions, the mean rotational changes of condyle were less than 0.2 mm; in the period of T3T4, the mean condylar translations in all directions were less than 0.02 mm. For series 2, the condyle translational changes in axial, coronal, and sagittal views were within 0.10 mm, and the rotation direction of condyle in all three views was the same within 1 year after operation. CONCLUSIONS: Fused three-dimensional images can be used to qualitatively and quantitatively evaluate condyle positional changes. The condylar position might be stable at 3 months postoperatively. The condyles of most of patients did not fully return to their preoperative position within 1 year after the operation. CLINICAL RELEVANCE: One method for fusing images has been established to detect the condylar positional changes. This method may be applied to estimate the bony changes of condyle, even bony changes in other part of dentomaxillofacial region. Meanwhile, the data of condyle positional changes from asymptomatic patients after the surgery within 1 year can be used as a reference for further exploration of the relationship between orthognathic surgery and the occurrence of osteoarthritis postoperatively in the future. KEY POINTS: • By fused 3D images, the change of condylar position after bilateral sagittal split ramus osteotomy can be observed intuitively. • For the patients with mandibular prognathism, the condylar position would be stable at 3 months postoperatively. • The condyles of most mandibular prognathism patients did not fully return to their preoperative position within 1 year after operation.

Effect of fracture orientation on detection accuracy of vertical root fractures in non-endodontically treated teeth using cone beam computed tomography.

OBJECTIVES: This study aimed to investigate the effect of fracture orientation on the detection accuracy of vertical root fractures (VRFs) in non-endodontically treated teeth using four different cone beam computed tomography (CBCT) units. MATERIALS AND METHODS: Thirty eight out of 148 extracted human permanent teeth were chosen randomly, and VRFs were artificially induced to result in 20 mesiodistally and 18 buccolingually oriented root fractures. The fracture width was subsequently measured. All the teeth were scanned with four CBCT units. CBCT images were evaluated independently by two observers. Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were calculated for each observer and fracture orientation. The AUC between the two fracture orientations was compared using Z test. RESULTS: The mean fracture width was 140 µm (standard deviation 26.8 µm). A statistically significant difference was found between the mesiodistal and buccolingual VRFs for the AUC from the CBCT unit 3D Accuitomo 170 (p = 0.02). There were no statistically significant differences between the mesiodistal and buccolingual VRFs for AUCs from the CBCT units NewTom VGi (p = 0.21), ProMax 3D Mid (p = 0.23), and i-CAT FLX (p = 0.21). CONCLUSION: Fracture orientations of teeth with VRFs in non-endodontically treated teeth may play a role in the detection accuracy of CBCT images, but this effect seems to be dependent on the CBCT unit used. CLINICAL RELEVANCE: Although for most of the CBCT units tested, the fracture orientation of VRF in non-endodontically treated teeth seems not to play a role for the diagnosis, clinical data is needed to further assess the impact of different devices on VRF detection.

Application of fused image in detecting abnormalities of temporomandibular joint.

OBJECTIVES:: To present a method for image fusion of cone beam CT (CBCT)/CT and MRI and to explore whether the image data sets fused in such a way could aid the detection of temporomandibular joint (TMJ) anatomical structures and lesions. METHODS:: There were five cases included in this study. One case was space occupying lesion giant cell tumour of tendon sheaths, one case was chronic inflammation in the condyle, one case was articular disc calcification of the bilateral TMJs, and the other two cases were TMJ disorders (anterior disc displacement without reduction). The digital imaging and communications in medicine format data of CT/CBCT and MRI of the cases were collected, and then imported to the Amira visual software where the registration process was conducted. Based on the different scan model, the registration process could be separated into automatic registration of CT/CBCT with quadrature slice MR images and the semi-automatic registration of CT/CBCT with oblique slice MR images by altering the registration parameters. Rigid transform model and the similarity metric of normalization mutual information was used for registration in the present study. RESULTS:: The relationship between the soft mass and hard tissue was shown clearly in the fused images when compared to sole observation of CBCT/CT or MR images. The fused images could define the tumour outline and the destructive bone clearly in the same image. The fused results helped the observers to ensure uncertain defects which could not be confirmed only by one image data set. CONCLUSIONS:: The CT/CBCT and MR images could be fused to aid detection of TMJ anatomical structures and related lesions.

Three-dimensional CBCT images registration method for TMJ based on reconstructed condyle and skull base.

OBJECTIVES: A method was introduced for three-dimensional (3D) cone-beamCT (CBCT) images registration of temporomandibular joint (TMJ). This study aimed to provide quantitative and qualitative analysis of TMJ bone changes in two-dimensional (2D) and 3D and to provide the technique for computer-aided diagnosis of temporomandibular joint disorders in the future. METHODS: 10 TMJ samples of six patients were obtained from Peking University Hospital of Stomatology. Four of the six patients imaged bilateral TMJs and the other two patients only imaged unilateral TMJ. Each sample consisted of two images from the same TMJ taken at different times. First, condyle and skull base were segmented semi-automatically for 3D model reconstruction. Then the segmented condyle and skull base were registered separately. Registration process can be divided into two processes of rough registration and fine registration. Rough registration step was achieved by selecting corresponding points manually and initialized fine registration. Condyle and skull base were fine registered by minimizing mean square error of condyle (MSEcondyle) and skull base (MSEskull) respectively. Qualitative assessment of osseous component changes utilized 2D color-fused model and 3D surface-fused model and quantitative analyses the convergence of this method used the mean square error of the model (MSEmodel). Independent repeated experiments were carried out to test the stability of our 3D registration method. RESULTS: Sufficiently alignment was achieved. Osseous abnormality and morphology changes were displayed using fusion model. MSEmodel of condylar registration and skull base registration declined 51.80% and 64.58% compared with that before registration. Quantitative analysis verified the stability of the method. CONCLUSIONS: The proposed method completed 3D TMJ registration for different physiological structure. The result of this method was accurate, reproducible and not relied on the experience of operators.

The detection accuracy of cone beam CT for osseous defects of the temporomandibular joint: a systematic review and meta-analysis.

The purpose of this review was to evaluate whether cone-beam computed tomography (CBCT) is reliable for the detection of bone changes of the temporomandibular joint (TMJ). Studies collected from the PubMed, Web of Science, Cochrane Library, ScienceDirect, Embase, Wanfang and CNKI databases were searched, and the publishing time was limited from January 1990 to December 2015. Eight studies (23 experimental research groups) were eventually included for further analysis. The pooled sensitivity was 0.67 and the pooled specificity was 0.87, which leads to a relatively large area (0.84) under the Receiver Operating Characteristic (ROC) curve. The related pooled positive likelihood ratio (+LR) and the pooled negative likelihood ratio (-LR) were 5.2 and 0.38, respectively. The subgroup analysis was conducted for four subgroups categorized by voxel size (≤0.2; >0.2, ≤0.3; >0.3, ≤0.4; >0.4, and ≤0.5 (mm)), and the ">0.4, ≤0.5" subgroup had a higher pooled sensitivity and pooled specificity than the other groups. The present study demonstrates that CBCT has a relatively high diagnostic accuracy for TMJ bone changes, although its reliability is limited. Voxel size did not play a role in the accuracy of CBCT.

A six-site method for the evaluation of periodontal bone loss in cone-beam CT images.

OBJECTIVES: In contrast to two-dimensional planar images, a measuring point is hardly repeatedly determined in a CBCT image when alveolar bone loss is assessed. Thus, the aim of the present study was to propose a six-site measuring method, which is closely related to anatomical structure, for the evaluation of alveolar bone loss in CBCT images. METHODS: 150 measuring points in 11 molars and 14 premolars from 6 patients (2 males and 4 females) were included. CBCT images of the teeth were acquired prior to periodontal surgery. Four observers measured the distances between cemento-enamel junctions and the apical bases of the periodontal bone defect at the mesio-buccal, mid-buccal, disto-buccal, mesio-lingual/palatal, mid-lingual/palatal and disto-lingual/palatal sites in CBCT images. Direct measurements of the six sites were correspondingly obtained in the subsequent periodontal surgeries. Differences between the distances measured in the CBCT images and during the surgery were analysed. Interobserver and intraobserver variances were tested. RESULTS: No statistically significant difference was found between the surgical and CBCT measurements (p = 0.84). Diagnostic coincidence rates of four observers were 86.7%, 87.3%, 88.7% and 88.0%, respectively. The interobserver (p = 0.95) and intraobserver (p = 0.30) variances were not significant. CONCLUSIONS: The six-site measuring method validated in the present study may be a useful three-dimensional measuring method for the evaluation of periodontal disease.

Age estimation based on pulp chamber volume of first molars from cone-beam computed tomography images.

AIM: To establish a method that can be used for human age estimation on the basis of pulp chamber volume of first molars and to identify whether the method is good enough for age estimation in real human cases. MATERIALS AND METHODS: CBCT images of 373 maxillary first molars and 372 mandibular first molars were collected to establish the mathematical model from 190 female and 213 male patients whose age between 12 and 69 years old. The inclusion criteria of the first molars were: no caries, no excessive tooth wear, no dental restorations, no artifacts due to metal restorative materials present in adjacent teeth, and no pulpal calcification. All the CBCT images were acquired with a CBCT unit NewTom VG (Quantitative Radiology, Verona, Italy) and reconstructed with a voxel-size of 0.15mm. The images were subsequently exported as DICOM data sets and imported into an open source 3D image semi-automatic segmenting and voxel-counting software ITK-SNAP 2.4 for the calculation of pulp chamber volumes. A logarithmic regression analysis was conducted with age as dependent variable and pulp chamber volume as independent variables to establish a mathematical model for the human age estimation. To identify the precision and accuracy of the model for human age estimation, another 104 maxillary first molars and 103 mandibular first molars from 55 female and 57 male patients whose age between 12 and 67 years old were collected, too. Mean absolute error and root mean square error between the actual age and estimated age were used to determine the precision and accuracy of the mathematical model. The study was approved by the Institutional Review Board of Peking University School and Hospital of Stomatology. RESULTS: A mathematical model was suggested for: AGE=117.691-26.442×ln (pulp chamber volume). The regression was statistically significant (p=0.000<0.01). The coefficient of determination (R(2)) was 0.564. There is a mean absolute error of 8.122 and root mean square error of 5.603 between the actual age and estimated age for all the tested teeth. CONCLUSION: The pulp chamber volume of first molar is a useful index for the estimation of human age with reasonable precision and accuracy.