Synthetic magnetic resonance imaging for quantitative parameter evaluation of temporomandibular joint disorders

Synthesis of T2-weighted images from proton density images using a generative adversarial network in a temporomandibular joint magnetic resonance imaging protocol

Purpose

This study proposed a generative adversarial network (GAN) model for T2-weighted image (WI) synthesis from proton density (PD)-WI in a temporomandibular joint (TMJ) magnetic resonance imaging (MRI) protocol.

Materials and Methods

From January to November 2019, MRI scans for TMJ were reviewed and 308 imaging sets were collected. For training, 277 pairs of PD- and T2-WI sagittal TMJ images were used. Transfer learning of the pix2pix GAN model was utilized to generate T2-WI from PD-WI. Model performance was evaluated with the structural similarity index map (SSIM) and peak signal-to-noise ratio (PSNR) indices for 31 predicted T2-WI (pT2). The disc position was clinically diagnosed as anterior disc displacement with or without reduction, and joint effusion as present or absent. The true T2-WI-based diagnosis was regarded as the gold standard, to which pT2-based diagnoses were compared using Cohen's ĸ coefficient.

Results

The mean SSIM and PSNR values were 0.4781(±0.0522) and 21.30(±1.51) dB, respectively. The pT2 protocol showed almost perfect agreement (ĸ=0.81) with the gold standard for disc position. The number of discordant cases was higher for normal disc position (17%) than for anterior displacement with reduction (2%) or without reduction (10%). The effusion diagnosis also showed almost perfect agreement (ĸ=0.88), with higher concordance for the presence (85%) than for the absence (77%) of effusion.

Conclusion

The application of pT2 images for a TMJ MRI protocol useful for diagnosis, although the image quality of pT2 was not fully satisfactory. Further research is expected to enhance pT2 quality.

Synthetic MRI in the detection and quantitative evaluation of sacroiliac joint lesions in axial spondyloarthritis

Objective

Our primary objective was to verify the hypothesis that synthetic magnetic resonance imaging (MRI) is similar to conventional MRI in detecting sacroiliac joint lesions in patients with axial spondyloarthritis (axSpA). A secondary objective was to assess the quantitative value of synthetic mapping in bone marrow edema (BME) and fat metaplasia.

Methods

A total of 132 axSpA patients who underwent synthetic and conventional MRI from October 2019 to March 2021 were included in this prospective study. Two independent readers visually evaluated active inflammatory (BME, capsulitis, enthesitis, and inflammation at site of erosion) and structural lesions (erosion, sclerosis, ankylosis, and fat metaplasia) of the sacroiliac joints on conventional and synthetic magnetic resonance (MR) images. In addition, T1, T2, and proton density (PD) values, which were generated by synthetic mapping, were used to further quantitatively evaluate BME and fat metaplasia. A McNemar test was used to compare the differences between the two methods in the detection of sacroiliac joint lesions. Intraclass correlation coefficients (ICCs) were used to assess the inter-reader consistency of quantitative values. Mann–Whitney tests were performed, and receiver operating characteristic (ROC) curves were created for all quantitative analyses.

Results

There were no statistical difference between synthetic and conventional MRI in the detection of sacroiliac joint lesions (all p-values > 0.05). A total of 103 images of BME and 111 images of fat metaplasia were quantitatively evaluated using T1, T2, and PD values. The consistency of quantitative values among readers was good (ICC 0.903–0.970). T1 and T2 values were consistently higher in BME than in normal marrow (p < 0.001), but PD values were not significantly different (p = 0.830). T2 and PD values were higher in fat metaplasia than in normal marrow, but T1 values were lower (p < 0.001). In the case of BME, T1 values had greater diagnostic efficiency [area under the curve (AUC) 0.99] than T2 values (AUC 0.78). There were no significant differences in the diagnostic efficiency of T1 (AUC 0.88), T2 (AUC 0.88), and PD (AUC 0.88) values in the case of fat metaplasia.

Conclusion

Synthetic MRI is as effective as conventional MRI in detecting sacroiliac joint lesions in patients with axSpA. Furthermore, synthetic mapping can accurately quantify BME and fat metaplasia.

Temporomandibular Disk Dislocation Impacts the Stomatognathic System: Comparative Study Based on Biexponential Quantitative T2 Maps

In this study, we aimed to assess the potential impact of temporomandibular disk displacement on anatomical structures of the stomatognathic system using biexponential T2 magnetic resonance imaging (MRI) maps. Fifty separate MRI scans of the temporomandibular joints (TMJ) of 25 patients were acquired with eight echo times. Biexponential T2 maps were created by weighted reconstruction based on Powell's conjugate direction method and divided into two groups: the TMJ without (32 images) and with (18 images) disk displacement. The disk, retrodiscal tissue, condylar bone marrow, masseter muscle, lateral and medial pterygoid muscles and dental pulp of the first and second molars were manually segmented twice. The intrarater reliability was assessed. The averages and standard deviations of the T2 times and fractions of each segmented region for each group were calculated and analysed with multiple Student's t-tests. Significant differences between groups were observed in the retrodiscal tissue, medial pterygoid muscle and bone marrow. The pulp short T2 component showed a trend toward statistical significance. The segmentation reliability was excellent (93.6%). The relationship between disk displacement and quantitative MRI features of stomatognathic structures can be useful in the combined treatment of articular disk displacement, pterygoid muscle tension and occlusive reconstruction.