3D imaging for hindfoot alignment assessment: a comparative study between non-weight-bearing MRI and weight-bearing CT.

PURPOSE: The authors compared measurements of hindfoot alignment on MR imaging with weight-bearing CT (WB-CT) to establish the degree of correlation. Forty-seven feet in 44 patients had weight-bearing CT and MRI studies performed on the same day. MATERIALS AND METHODS: Hindfoot alignment on MRI was assessed by two radiologists who calculated tibiocalcaneal angle (TCA) and calcaneofibular ligament angle (CFLA). On WB-CT, foot ankle offset (FAO), calcaneal offset (CO) and hindfoot angle (HA) were assessed by a senior Foot and Ankle Surgeon using dedicated software. Pearson correlation coefficient was used to evaluate the correlation between these measurements. RESULTS: The study group comprised 27 males and 17 females with a mean age of 45 years (range 13-79 years). A statistically significant positive correlation was identified between TCA on MRI and all measurements of hindfoot alignment on WB-CT (p = 0.001-0.005). The CFLA on MRI only had significant correlation with CO on WB-CT (p = 0.03). A significant negative correlation was observed between both MRI parameters (p < 0.001). CONCLUSION: A highly significant correlation between tibiocalcaneal angle on non-weight-bearing ankle MR imaging and hindfoot alignment measurements on weight-bearing CT was identified.

The awareness of hindfoot malalignment on non-weight-bearing ankle MRI.

OBJECTIVE: Hindfoot malalignment is a relatively common clinical finding and several studies have suggested that hindfoot valgus can be identified on non-weight-bearing ankle MRI. The aim of this study was to determine the awareness of hindfoot malalignment on ankle MRI amongst consultant musculoskeletal radiologists. MATERIALS AND METHODS: All MRI studies referred by Foot and Ankle Unit Consultants reported by one of 14 consultant musculoskeletal radiologists between March 2016 and August 2019 were retrieved from the Hospital Radiology Information System. These were reviewed independently by a radiology fellow and a consultant radiologist. Tibiocalcaneal angle (TCA) was measured, and extra-articular talocalcaneal (EA-TCI) and calcaneofibular impingement (EA-CFI) were recorded. Radiology reports were then analysed for mention of hindfoot malalignment and the presence of EA-TCI and EA-CFI. RESULTS: The study group comprised 129 patients, 46 males and 83 females with a mean age of 46.8 years (range 8-84 years). Based on review, hindfoot valgus was present in 78-80 cases (60.5-62%), EA-TCI in 30-36 cases (23.2-27.9%) and EA-CFI in 18-21 cases (14-16.3%). By comparison, MRI reports mentioned hindfoot valgus in 18 cases (2 incorrectly), EA-TCI in 8 cases (1 incorrectly) and EA-CFI in 10 cases (1 incorrectly). CONCLUSION: Hindfoot valgus, EA-TCI and EA-CFI were present relatively commonly on review of ankle MRI studies in patients referred from a specialist Foot and Ankle Unit but were commonly under-reported highlighting a relative lack of awareness of hindfoot malalignment on ankle MRI amongst musculoskeletal radiologists, which could impact negatively on patient management.

Pseudo-CT generation from multi-parametric MRI using a novel multi-channel multi-path conditional generative adversarial network for nasopharyngeal carcinoma patients.

PURPOSE: To develop and evaluate a novel method for pseudo-CT generation from multi-parametric MR images using multi-channel multi-path generative adversarial network (MCMP-GAN). METHODS: Pre- and post-contrast T1-weighted (T1-w), T2-weighted (T2-w) MRI, and treatment planning CT images of 32 nasopharyngeal carcinoma (NPC) patients were employed to train a pixel-to-pixel MCMP-GAN. The network was developed based on a 5-level Residual U-Net (ResU-Net) with the channel-based independent feature extraction network to generate pseudo-CT images from multi-parametric MR images. The discriminator with five convolutional layers was added to distinguish between the real CT and pseudo-CT images, improving the nonlinearity and prediction accuracy of the model. Eightfold cross validation was implemented to validate the proposed MCMP-GAN. The pseudo-CT images were evaluated against the corresponding planning CT images based on mean absolute error (MAE), peak signal-to-noise ratio (PSNR), Dice similarity coefficient (DSC), and Structural similarity index (SSIM). Similar comparisons were also performed against the multi-channel single-path GAN (MCSP-GAN), the single-channel single-path GAN (SCSP-GAN). RESULTS: It took approximately 20 h to train the MCMP-GAN model on a Quadro P6000, and less than 10 s to generate all pseudo-CT images for the subjects in the test set. The average head MAE between pseudo-CT and planning CT was 75.7 ± 14.6 Hounsfield Units (HU) for MCMP-GAN, significantly (P-values < 0.05) lower than that for MCSP-GAN (79.2 ± 13.0 HU) and SCSP-GAN (85.8 ± 14.3 HU). For bone only, the MCMP-GAN yielded a smaller mean MAE (194.6 ± 38.9 HU) than MCSP-GAN (203.7 ± 33.1 HU), SCSP-GAN (227.0 ± 36.7 HU). The average PSNR of MCMP-GAN (29.1 ± 1.6) was found to be higher than that of MCSP-GAN (28.8 ± 1.2) and SCSP-GAN (28.2 ± 1.3). In terms of metrics for image similarity, MCMP-GAN achieved the highest SSIM (0.92 ± 0.02) but did not show significantly improved bone DSC results in comparison with MCSP-GAN. CONCLUSIONS: We developed a novel multi-channel GAN approach for generating pseudo-CT from multi-parametric MR images. Our preliminary results in NPC patients showed that the MCMP-GAN method performed apparently superior to the U-Net-GAN and SCSP-GAN, and slightly better than MCSP-GAN.