OBJECTIVE: This study aimed to evaluate the effectiveness of deep learning method for denoising and artifact reduction (AR) in zero-TE (ZTE) magnetic resonance imaging (MRI). Also, Clinical applicability was evaluated by comparing image diagnosis to the temporomandibular joint (TMJ) cone-beam computed tomography (CBCT). METHODS: For thirty patients CBCT and routine ZTE-MRI data was collected, and an additional reduced scan time-ZTE-MRI was also obtained. Scan time-reduced image sets were processed into denoised and AR image based on deep learning technique. The image quality of routine sequence, de-noised and AR image sets were compared in quantitative evaluation using signal-to-noise ratio (SNR), and in qualitative using 3-point grading system (0, poor
1, good
2, excellent). The presence of osteoarthritis was assessed in each imaging protocol. Diagnostic accuracy of each protocol was compared against the CBCT results, which served as the reference standard. The SNR and the qualitative scores was compared using analysis of variance test and Kruskal-Wallis test, respectively. The diagnostic accuracy was assessed using the Cohen κ (<
0.5 = poor
0.5 to <
0.75 = moderate
0.75 to <
0.9 = good
≥0.9 = excellent). RESULTS: Both denoised and AR protocol resulted the significantly enhanced SNR compared to routine protocol and AR protocol showed higher SNR than denoised one. The qualitative assessment also showed highest grade in AR protocol with statistical significance. The osteoarthritis diagnosis showed enhanced agreement with CBCT in denoised (κ=0.928) and AR images (κ=0.929) than routine images (κ=0.707). CONCLUSIONS: A newly developed deep learning technique for both denoising and artifact reduction in ZTE-MRI presented clinical usefulness. Specifically, AR protocol showed significantly improved image quality and comparable diagnostic accuracy as CBCT. It can be expected that this novel technique would help overcome the current limitation of ZTE-MRI for replacing CBCT in bone imaging of TMJ.