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.

Magnetic resonance imaging (MRI) plays a significant role in the diagnosis of lumbar disc disease. However, the use of MRI is limited because of its high cost and significant operating and processing time. More importantly, MRI is contraindicated for some patients with claustrophobia or cardiac pacemakers due to the possibility of injury. In contrast, computed tomography (CT) scans are much less expensive, are faster, and do not face the same limitations. In this paper, we propose a method for estimating lumbar spine MR images based on CT images using a novel objective function and a dual cycle-consistent adversarial network (DC ${}^2$ Anet) with semi-supervised learning. The objective function includes six independent loss terms to balance quantitative and qualitative losses, enabling the generation of a realistic and accurate synthetic MR image. DC ${}^2$ Anet is also capable of semi-supervised learning, and the network is general enough for supervised or unsupervised setups. Experimental results prove that the method is accurate, being able to construct MR images that closely approximate reference MR images, while also outperforming four other state-of-the-art methods.

Objective: The purpose of our study is to develop a spoken dialogue system (SDS) for pain questionnaire in patients with spinal disease. We evaluate user satisfaction and validated the performance accuracy of the SDS in medical staff and patients.Methods: The SDS was developed to investigate pain and related psychological issues in patients with spinal diseases based on the pain questionnaire protocol. We recognized patients’ various answers, summarized important information, and documented them. User satisfaction and performance accuracy were evaluated in 30 potential users of SDS, including doctors, nurses, and patients and statistically analyzed.Results: The overall satisfaction score of 30 patients was 5.5 ± 1.4 out of 7 points. Satisfaction scores were 5.3 ± 0.8 for doctors, 6.0 ± 0.6 for nurses, and 5.3 ± 0.5 for patients. In terms of performance accuracy, the number of repetitions of the same question was 13, 16, and 33 (13.5%, 16.8%, and 34.7%) for doctors, nurses, and patients, respectively. The number of errors in the summarized comment by the SDS was 5, 0, and 11 (5.2%, 0.0%, and 11.6 %), respectively. The number of summarization omissions was 7, 5, and 7 (7.3%, 5.3%, and 7.4%), respectively.Conclusion: This is the first study in which voice-based conversational artificial intelligence (AI) was developed for a spinal pain questionnaire and validated by medical staff and patients. The conversational AI showed favorable results in terms of user satisfaction and performance accuracy. Conversational AI can be useful for the diagnosis and remote monitoring of various patients as well as for pain questionnaires in the future.

Cho SH, et al. Korean J Spine. 2014 Jun;11(2):45-51. http://dx.doi.org/10.14245/kjs.2014.11.2.45

Bone flap resorption (BFR) after cranioplasty with an autologous bone flap (ABF) is well known. However, the prevalences and degrees of BFR remain unclear. This study aimed to evaluate changes in ABFs following cranioplasty and to investigate factors related with BFR. We retrospectively reviewed 97 patients who underwent cranioplasty with frozen ABF between January 2007 and December 2016. Brain CT images of these patients were reconstructed to form three-dimensional (3D) images, and 3D images of ABF were separated using medical image editing software. ABF volumes on images were measured using 3D image editing software and were compared between images in the immediate postoperative period and at postoperative 12 months. Risk factors related with BFR were also analyzed. The volumes of bone flaps calculated from CT images immediately after cranioplasty ranged from 55.3 cm³ to 175 cm³. Remnant bone flap volumes at postoperative 12 months ranged from 14.2% to 102.5% of the original volume. Seventy-five patients (77.3%) had a BFR rate exceeding 10% at 12 months after cranioplasty, and 26 patients (26.8%) presented severe BFR over 40%. Ten patients (10.3%) underwent repeated cranioplasty due to severe BFR. The use of a 5-mm burr for central tack-up sutures was significantly associated with BFR (p<0.001). Most ABFs after cranioplasty are absorbed. Thus, when using frozen ABF, patients should be adequately informed. To prevent BFR, making holes must be kept to a minimum during ABF grafting.

Most studies on the ossification of the posterior longitudinal ligament (OPLL) using the finite element method were conducted in the neutral state, and the resulting decompression was judged to be good. As these studies do not reflect the actual behavior of the cervical spine, this study conducted an analysis in the neutral state and a biomechanical analysis during flexion and extension behaviors. After validation via the construction of an intact cervical spine model, the focal OPLL model was inserted into the C4–C5 segment and a simulation was performed. The neutral state was shown by applying a fixed condition to the lower part of the T1 and Y-axis fixed condition of the spinal cord and simulating spinal cord compression with OPLL. For flexion and extension simulation, a ±30-degree displacement was additionally applied to the top of the C2 dens. Accordingly, it was confirmed that spinal cord decompression did not work well during the flexion and extension behaviors, but rather increased. Thus, if patients with focal OPLL inevitably need to undergo posterior decompression, additional surgery using an anterior approach should be considered.