Machine learning-based medical imaging diagnosis in patients with temporomandibular disorders: a diagnostic test accuracy systematic review and meta-analysis

OBJECTIVES

Temporomandibular disorders (TMDs) are the second most common musculoskeletal condition which are challenging tasks for most clinicians. Recent research used machine learning (ML) algorithms to diagnose TMDs intelligently. This study aimed to systematically evaluate the quality of these studies and assess the diagnostic accuracy of existing models.

MATERIALS AND METHODS

Twelve databases (Europe PMC, Embase, etc.) and two registers were searched for published and unpublished studies using ML algorithms on medical images. Two reviewers extracted the characteristics of studies and assessed the methodological quality using the QUADAS-2 tool independently.

RESULTS

A total of 28 studies (29 reports) were included: one was at unclear risk of bias and the others were at high risk. Thus the certainty of evidence was quite low. These studies used many types of algorithms including 8 machine learning models (logistic regression, support vector machine, random forest, etc.) and 15 deep learning models (Resnet152, Yolo v5, Inception V3, etc.). The diagnostic accuracy of a few models was relatively satisfactory. The pooled sensitivity and specificity were 0.745 (0.660-0.814) and 0.770 (0.700-0.828) in random forest, 0.765 (0.686-0.829) and 0.766 (0.688-0.830) in XGBoost, and 0.781 (0.704-0.843) and 0.781 (0.704-0.843) in LightGBM.

CONCLUSIONS

Most studies had high risks of bias in Patient Selection and Index Test. Some algorithms are relatively satisfactory and might be promising in intelligent diagnosis. Overall, more high-quality studies and more types of algorithms should be conducted in the future.

CLINICAL RELEVANCE

We evaluated the diagnostic accuracy of the existing models and provided clinicians with much advice about the selection of algorithms. This study stated the promising orientation of future research, and we believe it will promote the intelligent diagnosis of TMDs.

Development of artificial intelligence model for supporting implant drilling protocol decision making

PURPOSE

This study aimed to develop an artificial intelligence (AI) model to support the determination of an appropriate implant drilling protocol using cone-beam computed tomography (CBCT) images.

METHODS

Anonymized CBCT images were obtained from 60 patients. For each case, after implant placement, images of the bone regions at the implant site were extracted from 20 slices of CBCT images. Based on the actual drilling protocol, the images were classified into three categories: protocols A, B, and C. A total of 1,200 images were divided into training and validation datasets (n = 960, 80%) and a test dataset (n = 240, 20%). Another 240 images (80 images for each type) were extracted from the 60 cases as test data. An AI model based on LeNet-5 was developed using these data sets. The accuracy, sensitivity, precision, F-value, area under the curve (AUC) value, and receiver operating curve were calculated.

RESULTS

The accuracy of the trained model is 93.8%. The sensitivity results for drilling protocols A, B, and C were 97.5%, 95.0%, and 85.0%, respectively, while those for protocols A, B, and C were 86.7%, 92.7%, and 100%, respectively, and the F values for protocols A, B, and C were 91.8%, 93.8%, and 91.9%, respectively. The AUC values for protocols A, B, and C are 98.6%, 98.6%, and 99.4%, respectively.

CONCLUSION

The AI model established in this study was effective in predicting drilling protocols from CBCT images before surgery, suggesting the possibility of developing a decision-making support system to promote primary stability.