Yeom JA, et al. Korean J Radiol. 2016 Sep-Oct;17(5):565-580. https://doi.org/10.3348/kjr.2016.17.5.565

To generate synthetic spine magnetic resonance (MR) images from spine computed tomography (CT) using generative adversarial networks (GANs), as well as to determine the similarities between synthesized and real MR images. GANs were trained to transform spine CT image slices into spine magnetic resonance T2 weighted (MRT2) axial image slices by combining adversarial loss and voxel-wise loss. Experiments were performed using 280 pairs of lumbar spine CT scans and MRT2 images. The MRT2 images were then synthesized from 15 other spine CT scans. To evaluate whether the synthetic MR images were realistic, two radiologists, two spine surgeons, and two residents blindly classified the real and synthetic MRT2 images. Two experienced radiologists then evaluated the similarities between subdivisions of the real and synthetic MRT2 images. Quantitative analysis of the synthetic MRT2 images was performed using the mean absolute error (MAE) and peak signal-to-noise ratio (PSNR). The mean overall similarity of the synthetic MRT2 images evaluated by radiologists was 80.2%. In the blind classification of the real MRT2 images, the failure rate ranged from 0% to 40%. The MAE value of each image ranged from 13.75 to 34.24 pixels (mean, 21.19 pixels), and the PSNR of each image ranged from 61.96 to 68.16 dB (mean, 64.92 dB). This was the first study to apply GANs to synthesize spine MR images from CT images. Despite the small dataset of 280 pairs, the synthetic MR images were relatively well implemented. Synthesis of medical images using GANs is a new paradigm of artificial intelligence application in medical imaging. We expect that synthesis of MR images from spine CT images using GANs will improve the diagnostic usefulness of CT. To better inform the clinical applications of this technique, further studies are needed involving a large dataset, a variety of pathologies, and other MR sequence of the lumbar spine.

Objectives: The purpose of this study was to present the CT and MRI findings of patients with fibrous dysplasia (FD) of the spine. Methods: Among the patients with pathologically proven skeletal FD, 12 (8 males and 4 females; mean age, 43 years) who were evaluated with either spine CT or MRI were included. The number and location of the involved vertebral segments, the presence of lytic lesions, ground-glass opacity (GGO), an expansile nature, cortical disruption, a sclerotic rim, a decrease in body height and contour deformity were examined on CT scans (n=12), while signal intensity, enhancement patterns and the presence of a dark signal rim on the lesion were examined using MRI (n=9). Results: Nine patients had polyostotic FD, including one with an isolated spinal localisation, while three had monostotic FD. An expansile nature (n=3) and osteolytic lesions with GGO (n=3) were seen. On CT images, GGO was noted in all patients. An expansile nature (n=11) and presence of lytic lesions (n=11) were noted. A ...

Avulsion injuries result from the application of a tensile force to a musculoskeletal unit or ligament. Although injuries tend to occur more commonly in skeletally immature populations due to the weakness of their apophysis, adults may also be subject to avulsion fractures, particularly those with osteoporotic bones. The most common sites of avulsion injuries in adolescents and children are apophyses of the pelvis and knee. In adults, avulsion injuries commonly occur within the tendon due to underlying degeneration or tendinosis. However, any location can be involved in avulsion injuries. Radiography is the first imaging modality to diagnose avulsion injury, although advanced imaging modalities are occasionally required to identify subtle lesions or to fully delineate the extent of the injury. Ultrasonography has a high spatial resolution with a dynamic assessment potential and allows the comparison of a bone avulsion with the opposite side. Computed tomography is more sensitive for depicting a tiny osseous fragment located adjacent to the expected attachment site of a ligament, tendon, or capsule. Moreover, magnetic resonance imaging is the best imaging modality for the evaluation of soft tissue abnormalities, especially the affected muscles, tendons, and ligaments. Acute avulsion injuries usually manifest as avulsed bone fragments. In contrast, chronic injuries can easily mimic other disease processes, such as infections or neoplasms. Therefore, recognizing the vulnerable sites and characteristic imaging features of avulsion fractures would be helpful in ensuring accurate diagnosis and appropriate patient management. To this end, familiarity with musculoskeletal anatomy and mechanism of injury is necessary.