Diagnostic value of panoramic radiography in predicting inferior alveolar nerve injury after mandibular third molar extraction: a meta-analysis

The diagnostic performance of impacted third molars in the mandible: A review of deep learning on panoramic radiographs

Background

Mandibular third molar is prone to impaction, resulting in its inability to erupt into the oral cavity. The radiographic examination is required to support the odontectomy of impacted teeth. The use of computer-aided diagnosis based on deep learning is emerging in the field of medical and dentistry with the advancement of artificial intelligence (AI) technology. This review describes the performance and prospects of deep learning for the detection, classification, and evaluation of third molar-mandibular canal relationships on panoramic radiographs.

Methods

This work was conducted using three databases: PubMed, Google Scholar, and Science Direct. Following the literature selection, 49 articles were reviewed, with the 12 main articles discussed in this review.

Results

Several models of deep learning are currently used for segmentation and classification of third molar impaction with or without the combination of other techniques. Deep learning has demonstrated significant diagnostic performance in identifying mandibular impacted third molars (ITM) on panoramic radiographs, with an accuracy range of 78.91% to 90.23%. Meanwhile, the accuracy of deep learning in determining the relationship between ITM and the mandibular canal (MC) ranges from 72.32% to 99%.

Conclusion

Deep learning-based AI with high performance for the detection, classification, and evaluation of the relationship of ITM to the MC using panoramic radiographs has been developed over the past decade. However, deep learning must be improved using large datasets, and the evaluation of diagnostic performance for deep learning models should be aligned with medical diagnostic test protocols. Future studies involving collaboration among oral radiologists, clinicians, and computer scientists are required to identify appropriate AI development models that are accurate, efficient, and applicable to clinical services.

Application of Cone Beam Computed Tomography in Risk Assessment of Lower Third Molar Surgery

Risks of lower third molar surgery like the inferior alveolar nerve injury may result in permanent consequences. Risk assessment is important prior to the surgery and forms part of the informed consent process. Traditionally, plain radiographs like orthopantomogram have been used routinely for this purpose. Cone beam computed tomography (CBCT) has offered more information from the 3D images in the lower third molar surgery assessment. The proximity of the tooth root to the inferior alveolar canal, which harbours the inferior alveolar nerve, can be clearly identified on CBCT. It also allows the assessment of potential root resorption of the adjacent second molar as well as the bone loss at its distal aspect as a consequence of the third molar. This review summarized the application of CBCT in the risk assessment of lower third molar surgery and discussed how it could aid in the decision-making of high-risk cases to improve safety and treatment outcomes.

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.

Association of the Inferior Alveolar Nerve Position and Nerve Injury: A Systematic Review and Meta-Analysis

Background: We aimed to compare the relationship between the buccal and lingual positions of the inferior alveolar nerve canal (IAC) relative to the lower third molar (LM3) and the rate of the inferior alveolar nerve (IAN) injury. Methods: A systematic search was performed in the following databases: PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), Web of Science, and Journals@Ovid. No language or publication status restrictions were set. The publication year was set from 2009 to 2021. The process of meta-analysis was performed by Review Manager software (Cochrane Collaboration). Results: A total of 1063 articles were initially searched and full texts of 53 articles were read, and 11 satisfactory articles were found. There was a statistical difference between the rate of IAN injury and the lingual position and buccal position of the IAC relative to the LM3 roots (OR, 4.96; 95% CI, 2.11 to 11.62; p = 0.0002), with high heterogeneity (p = 0.001, I² = 65%). Conclusion: A statistical difference was found in the rate of IAN injury between cases where the IAC was positioned buccally and lingually of the LM3 roots. The IAC was at a relatively higher risk of damage in third molar extraction when it was located on the lingual position of the LM3 roots.

Comparison of Digital OPG and CBCT in Assessment of Risk Factors Associated with Inferior Nerve Injury during Mandibular Third Molar Surgery

Background: Pre-operative radiographic assessment of the anatomical relationship between the roots of the mandibular third molar and the inferior alveolar nerve (IAN) is a must to minimize the risk of IAN injury during surgery. Objectives: To compare the radiographic signs of digital orthopantomogram (OPG) and cone-beam computed tomography (CBCT). An additional objective was to assess the cortex status between the mandibular canal and third molar on CBCT images in relation to the demographic characteristics, region (right or left side), and angulation of mandibular molar. Methodology: In this retrospective study, a total of 350 impacted mandibular third molars with a close relationship between the inferior alveolar canal (IAC) and impacted mandibular third molars on digital OPG were further referred for CBCT imaging for assessment of the position of the mandibular canal. The study was conducted between August 2018 and February 2020. Digital OPGs were evaluated for radiographic signs like interruption of the mandibular canal wall, darkening of the roots, diversion of the mandibular canal, and narrowing of the mandibular canal. The age and sex of patients, site of impacted third molar, Winter’s classification of mandibular third molar, position of IAC relative to impacted molar, and the radiographic markers of OPG were assessed for cortical integrity using CBCT. Chi square testing was applied to study the values of difference and binomial logistic regression was done to assess the factors associated with cortication. Statistical significance was set at p ≤ 0.05. Results: Among 350 patients, 207 (59.1%) were male and 143 (40.9%) were female with a mean age of 36.8 years. The most common OPG sign was interruption of white line, seen in 179 (51.1%) cases. In total, 246 cases (70.3%) showed an absence of canal cortication between the mandibular canal and the impacted third molar on CBCT images. Cortication was observed in all cases with a combination of panoramic signs which was statistically significant (p = 0.047). Cortication was observed in 85 (50.6%) cases where IAC was positioned on the buccal side, 11 (16.9%) in cases of inferiorly positioned IAC, and just 8 (7.6%) for cases of lingually positioned IAC which was statistically significant (p = 0.003). Statistically insignificant (p > 0.05) results were noted for cortex status in CBCT images with regards to the age, sex, site, and angulation of impacted third molars. Conclusion: CBCT imaging is highly recommended for those cases where diversion of the mandibular canal is observed on OPG and when the roots are present between canals.

Artificial Intelligence Model to Detect Real Contact Relationship between Mandibular Third Molars and Inferior Alveolar Nerve Based on Panoramic Radiographs

This study aimed to develop a novel detection model for automatically assessing the real contact relationship between mandibular third molars (MM3s) and the inferior alveolar nerve (IAN) based on panoramic radiographs processed with deep learning networks, minimizing pseudo-contact interference and reducing the frequency of cone beam computed tomography (CBCT) use. A deep-learning network approach based on YOLOv4, named as MM3-IANnet, was applied to oral panoramic radiographs for the first time. The relationship between MM3s and the IAN in CBCT was considered the real contact relationship. Accuracy metrics were calculated to evaluate and compare the performance of the MM3-IANnet, dentists and a cooperative approach with dentists and the MM3-IANnet. Our results showed that in comparison with detection by dentists (AP = 76.45%) or the MM3-IANnet (AP = 83.02%), the cooperative dentist-MM3-IANnet approach yielded the highest average precision (AP = 88.06%). In conclusion, the MM3-IANnet detection model is an encouraging artificial intelligence approach that might assist dentists in detecting the real contact relationship between MM3s and IANs based on panoramic radiographs.

Reliability of Orthopantamogram in Lower Third Molar Surgery: Inter- and Intra-observer Agreement

Context:

The evaluation of relationship between the roots of impacted lower third molar (IL3M) and inferior alveolar nerve injury (IAN) with orthopantamogram (OPG) is mandatory before performing de-impaction surgery. An investigation is considered reliable if it can be reproduced by various examiners. Assessment of OPG is subjective and varies among examiners. OPG is reliable to clinicians if the interpretation is not a product of guess work.

Aim:

The aim of this study was to evaluate the magnitude of agreement among oral surgeons and oral radiologist in observing intimate relationship between IL3M and mandibular canal.

Materials and Methods:

OPGs were evaluated by two oral surgeons and one oral radiologist for nerve root relationship. All the three were from different institutions with 10–15 years of experience. The three observers were blinded from each other’s findings. A total of 127 OPGs were evaluated for inter-observer agreement. Fifty OPGs were evaluated after 60 days for intra-observer agreement.

The agreement was evaluated based on Cohen’s κ statistics.

Results:

Our results denote that the interpretation of OPG among specialists is not in good agreement. We suggest development of methods to standardize evaluation of OPG and the exposure technique to improve inter-observer agreement among the dental specialists.

Assessment of the reliability of radiographic signs on panoramic radiographs to determine the relationship between mandibular third molars and the inferior alveolar canal

OBJECTIVE

To assess the reliability of radiographic signs of the relationship between impacted mandibular third molars (IMTMs) and the inferior alveolar canal (IAC), and the vertical relationship between IMTM and IAC on panoramic radiographs (PRs) as risk indicators for IAC exposure during extraction.

STUDY DESIGN

Seven radiographic signs regarding the appearance of roots and the IAC were assessed as present or absent on PRs. The vertical relationship between IMTM and the IAC was divided into 3 levels. The correlation between radiographic data and IAC exposure during extraction was analyzed by using χ² statistics and logistic regression analysis.

RESULTS

In total, 198 IMTMs were examined. In 46 cases (23.2%), the IAC was exposed during extraction. Four radiographic signs-darkening of the root, narrowing of the root, narrowing of the IAC, and diversion of the IAC-were associated with IAC exposure (P < .05). Concurrence of 2 or more of 7 radiographic signs increased the risk of IAC exposure (P < .001). The vertical relationship between IMTM and the IAC was found to be a risk indicator for IAC exposure (P < .001). No significant correlation was found between IAC exposure and inferior alveolar nerve damage (P = .148).

CONCLUSIONS

Panoramic radiography is useful for assessing the risk of IAC exposure during IMTM extraction.

Automated detection of third molars and mandibular nerve by deep learning

The approximity of the inferior alveolar nerve (IAN) to the roots of lower third molars (M3) is a risk factor for the occurrence of nerve damage and subsequent sensory disturbances of the lower lip and chin following the removal of third molars. To assess this risk, the identification of M3 and IAN on dental panoramic radiographs (OPG) is mandatory. In this study, we developed and validated an automated approach, based on deep-learning, to detect and segment the M3 and IAN on OPGs. As a reference, M3s and IAN were segmented manually on 81 OPGs. A deep-learning approach based on U-net was applied on the reference data to train the convolutional neural network (CNN) in the detection and segmentation of the M3 and IAN. Subsequently, the trained U-net was applied onto the original OPGs to detect and segment both structures. Dice-coefficients were calculated to quantify the degree of similarity between the manually and automatically segmented M3s and IAN. The mean dice-coefficients for M3s and IAN were 0.947 ± 0.033 and 0.847 ± 0.099, respectively. Deep-learning is an encouraging approach to segment anatomical structures and later on in clinical decision making, though further enhancement of the algorithm is advised to improve the accuracy.

Self-assessed neurological disturbances after surgical removal of impacted lower third molar: a pragmatic prospective study on 423 surgical extractions in 247 consecutive patients

OBJECTIVES

The aim of this study was to assess the combined role of current radiographic risk indicators and patient age in predicting lower lip sensitivity disturbances after surgical removal of impacted lower third molars. The question was which combinations indicate low or high risk.

MATERIALS AND METHODS

A prospective study was implemented involving 247 consecutive outpatients who underwent 423 surgical extractions. The predictor variables were patient age and risk indicators observed on panoramic radiographs. The outcome variable was the incidence of self-assessed lip sensitivity alterations. The extractions were subdivided into four groups according to the predictors.

RESULTS

Two hundred forty-five teeth were extracted in patients younger than 25 years and 178 in patients 25 years old or older; radiographic risk indicators were associated with 226 out of 423 teeth (53.43%). No permanent neurological damage was observed. Transient lip sensitivity alterations were observed in five cases (1.18%; 95% confidence interval = 0.4 to 2.7%), all in the older group with radiographic risk indicators.

CONCLUSIONS

The data indicate a low overall incidence of transient lip sensitivity impairment that occurred only in the presence of radiographic risk indicators in patients aged ≥ 25 years.

CLINICAL RELEVANCE

Informed consent should include the possibility of inferior alveolar nerve injury in mature patients with radiographic risk indicators. Prophylactic removal of impacted teeth with radiographic signs of risk may be indicated when the patient is not yet aged 25 years.

Magnetic resonance neurography in the management of peripheral trigeminal neuropathy: experience in a tertiary care centre

OBJECTIVE

This tertiary care experience examines the utility of magnetic resonance neurography (MRN) in the management of peripheral trigeminal neuropathies.

MATERIALS AND METHODS

Seventeen patients with clinically suspected peripheral trigeminal neuropathies (inferior alveolar nerve and lingual nerve) were imaged uniformly with 1.5-T examinations. MRN results were correlated with clinical and surgical findings in operated patients and the impact on clinical management was assessed.

RESULTS

Clinical findings included pain (14/17), sensory changes (15/17), motor changes (2/17) and palpable masses (3/17). Inciting events included prior dental surgery (12/17), trauma (1/17) and idiopathic incidents (4/17). Non-affected side nerves and trigeminal nerves in the intracranial and skull base course were normal in all cases. Final diagnoses on affected sides were nerve inflammation (4/17), neuroma in continuity (2/17), LN transection (1/17), scar entrapment (3/17), infectious granuloma (1/17), low-grade injuries (3/17) and no abnormality (3/17). Associated submandibular gland and sublingual gland oedema-like changes were seen in 3/17 cases because of parasympathetic effects. Moderate-to-excellent MRN-surgical correlation was seen in operated (8/17) patients, and neuroma and nerve transection were prospectively identified in all cases.

CONCLUSION

MRN is useful for the diagnostic work-up of suspected peripheral trigeminal neuropathy patients with significant impact on clinical management and moderate-to-excellent correlation with intra-operative findings.

KEY POINTS

• MRN substantially impacts diagnostic thinking and management in peripheral trigeminal neuropathy. • MRN has moderate-to-excellent correlation with intra-operative findings. • MRN should be considered in pre-surgical planning of peripheral trigeminal neuropathy subjects.