Developing a deep learning model for sleep stage prediction in obstructive sleep apnea cohort using 60 GHz frequency-modulated continuous-wave radar.

Given the significant impact of sleep on overall health, radar technology offers a promising, non-invasive, and cost-effective avenue for the early detection of sleep disorders, even prior to relying on polysomnography (PSG)-based classification. In this study, we employed an attention-based bidirectional long short-term memory (Attention Bi-LSTM) model to accurately predict sleep stages using 60 GHz frequency-modulated continuous-wave (FMCW) radar. Our dataset comprised 78 participants from an ongoing obstructive sleep apnea (OSA) cohort, recruited between July 2021 and November 2022, who underwent overnight polysomnography alongside radar sensor monitoring. The dataset encompasses comprehensive polysomnography recordings, spanning both sleep and wakefulness states. The predictions achieved a Cohen's kappa coefficient of 0.746 and an overall accuracy of 85.2% in classifying wakefulness, rapid-eye-movement (REM) sleep, and non-REM (NREM) sleep (N1 + N2 + N3). The results demonstrated that the models incorporating both Radar 1 and Radar 2 data consistently outperformed those using only Radar 1 data, indicating the potential benefits of utilising multiple radars for sleep stage classification. Although the performance of the models tended to decline with increasing OSA severity, the addition of Radar 2 data notably improved the classification accuracy. These findings demonstrate the potential of radar technology as a valuable screening tool for sleep stage classification.

Electron density dual-energy CT can improve the detection of lumbar disc herniation with higher image quality than standard and virtual non-calcium images.

OBJECTIVES: To compare the diagnostic performance and image quality of dual-energy computed tomography (DECT) with electron density (ED) image reconstruction with those of DECT with standard CT (SC) and virtual non-calcium (VNCa) image reconstructions, for diagnosing lumbar disc herniation (L-HIVD). METHODS: A total of 59 patients (354 intervertebral discs from T12/L1 to L5/S1; mean age, 60 years; 30 women and 29 men) who underwent DECT with spectral reconstruction and 3-T MRI within 2 weeks were enrolled between March 2021 and February 2022. Four radiologists independently assessed three image sets of randomized ED, SC, and VNCa images to detect L-HIVD at 8-week intervals. The coefficient of variance (CV) and the Weber contrast of the ROIs in the normal and diseased disc to cerebrospinal fluid space (NCR-normal/-diseased, respectively) were calculated to compare the image qualities of the noiseless ED and other series. RESULTS: Overall, 129 L-HIVDs were noted on MRI. In the detection of L-HIVD, ED showed a higher AUC and sensitivity than SC and VNCa; 0.871 vs 0.807 vs 833 (p = 0.002) and 81% vs 70% vs 74% (p = 0.006 for SC), respectively. CV was much lower in all measurements of ED than those for SC and VNCa (p < 0.001). Furthermore, NCR-normal and NCR-diseased were the highest in ED (ED vs SC in NCR-normal and NCR-diseased, p = 0.001 and p = 0.004, respectively; ED vs VNCa in NCR-diseased, p = 0.044). CONCLUSION: Compared to SC and VNCa images, DECT with ED reconstruction can enhance the AUC and sensitivity of L-HIVD detection with a lower CV and higher NCR. CLINICAL RELEVANCE STATEMENT: To our knowledge, this is the first study to quantify the image quality of noiseless ED images. ED imaging may be helpful for detecting L-HIVD in patients who cannot undergo MRI. KEY POINTS: ED images have diagnostic potential, but relevant quantitative analyses of image quality are limited. ED images detect disc herniation, with a better coefficient of variance and normalized contrast ratio values. ED images could detect L-HIVD when MRI is not an option.

Diagnostic performance of electron-density dual-energy CT in detection of cervical disc herniation in comparison with standard gray-scale CT and virtual non-calcium images.

OBJECTIVES: To assess the diagnostic performance of dual-energy CT (DECT) with electron-density (ED) image reconstruction compared with standard CT (SC) and virtual non-calcium (VNCa) image CT reconstruction for detecting cervical disc herniation. METHODS: This cross-sectional study was approved by the IRB. We enrolled 64 patients (336 intervertebral discs from C2/3 to C7/T1; mean age, 55 years; 17 women and 47 men) who underwent DECT with spectral reconstruction and 3-T MRI within 2 weeks between January 2018 and June 2020. Four radiologists independently evaluated the first image set of randomized SC, VNCa, and ED images to detect cervical disc herniation. After 8 weeks, the readers re-evaluated the second and the last image sets with an 8-week interval. MRI evaluations performed by two other experienced served as the reference standard. Comparing diagnostic performance between each images set was evaluated by a generalized estimating equation. RESULTS: A total of 233 cervical disc herniations were noted on MRI. For detecting cervical disc herniation, electron-density images showed higher sensitivity (94% [219/233; 95% CI, 90-97] vs. 76% [177/233; 70-81] vs. 69% [160/233; 62-76]) (p < 0.001) and similar specificity (90% [93/103; 83-95] vs. 89% [92/103; 82-96] vs. 90% [93/103; 83-95]) (p > 0.05) as SC and VNCa images, respectively. Inter-reader agreement for cervical disc herniation calculated among the four readers was moderate for all image sets (κ = 0.558 for ED, κ = 0.422 for SC, and κ = 0.449 for VNCa). CONCLUSION: DECT with ED reconstruction can improve cervical disc herniation detection and diagnostic confidence compared with SC and VNCa images. KEY POINTS: • Intervertebral discs with high material density are well visualized on electron-density images obtained from dual-energy CT. • Electron-density images showed much higher sensitivity and diagnostic accuracy than standard CT and virtual non-calcium images for the detection of cervical disc herniation. • Electron-density images can have false-negative results, especially for disc herniation with high signal intensity on T2W images and can show pseudo-disc extrusion at the lower cervical spine.

Non-Contact VITAL Signs Monitoring of a Patient Lying on Surgical Bed Using Beamforming FMCW Radar.

Respiration and heartrates are important information for surgery. When the vital signs of the patient lying prone are monitored using radar installed on the back of the surgical bed, the surgeon's movements reduce the accuracy of these monitored vital signs. This study proposes a method for enhancing the monitored vital sign accuracies of a patient lying on a surgical bed using a 60 GHz frequency modulated continuous wave (FMCW) radar system with beamforming. The vital sign accuracies were enhanced by applying a fast Fourier transform (FFT) for range and beamforming which suppress the noise generated at different ranges and angles from the patient's position. The experiment was performed for a patient lying on a surgical bed with or without surgeon. Comparing a continuous-wave (CW) Doppler radar, the FMCW radar with beamforming improved almost 22 dB of signal-to-interference and noise ratio (SINR) for vital signals. More than 90% accuracy of monitoring respiration and heartrates was achieved even though the surgeon was located next to the patient as an interferer. It was analyzed using a proposed vital signal model included in the radar IF equation.

Assessment of rapidly advancing bone age during puberty on elbow radiographs using a deep neural network model.

OBJECTIVES: Bone age is considered an indicator for the diagnosis of precocious or delayed puberty and a predictor of adult height. We aimed to evaluate the performance of a deep neural network model in assessing rapidly advancing bone age during puberty using elbow radiographs. METHODS: In all, 4437 anteroposterior and lateral pairs of elbow radiographs were obtained from pubertal individuals from two institutions to implement and validate a deep neural network model. The reference standard bone age was established by five trained researchers using the Sauvegrain method, a scoring system based on the shapes of the lateral condyle, trochlea, olecranon apophysis, and proximal radial epiphysis. A test set (n = 141) was obtained from an external institution. The differences between the assessment of the model and that of reviewers were compared. RESULTS: The mean absolute difference (MAD) in bone age estimation between the model and reviewers was 0.15 years on internal validation. In the test set, the MAD between the model and the five experts ranged from 0.19 to 0.30 years. Compared with the reference standard, the MAD was 0.22 years. Interobserver agreement was excellent among reviewers (ICC: 0.99) and between the model and the reviewers (ICC: 0.98). In the subpart analysis, the olecranon apophysis exhibited the highest accuracy (74.5%), followed by the trochlea (73.7%), lateral condyle (73.7%), and radial epiphysis (63.1%). CONCLUSIONS: Assessment of rapidly advancing bone age during puberty on elbow radiographs using our deep neural network model was similar to that of experts. KEY POINTS: • Bone age during puberty is particularly important for patients with scoliosis or limb-length discrepancy to determine the phase of the disease, which influences the timing and method of surgery. • The commonly used hand radiographs-based methods have limitations in assessing bone age during puberty due to the less prominent morphological changes of the hand and wrist bones in this period. • A deep neural network model trained with elbow radiographs exhibited similar performance to human experts on estimating rapidly advancing bone age during puberty.

Imaging appearance of post-arthroscopic radiocarpal chondrolysis: comparison with osteoarthritis associated with scapholunate dissociation.

BACKGROUND: Since the diagnosis of post-arthroscopic chondrolysis is very difficult, it can be underdiagnosed and confused with other diseases in clinical practice. PURPOSE: To propose imaging features of post-arthroscopic radiocarpal chondrolysis (PRCC) and to compare these with osteoarthritis associated with scapholunate dissociation which are the most common misdiagnoses of PRCC. MATERIAL AND METHODS: To identify missed diagnoses of PRCC, 994 magnetic resonance imaging scans performed in 910 patients were retrospectively reviewed. After the identification of 73 patients who exhibited significant radiocarpal cartilage loss, 11 were diagnosed with PRCC. Since scapholunate advanced collapse was the most common incorrect diagnosis of PRCC (4/11), the imaging findings were compared among the 11 patients with PRCC and 14 patients with osteoarthritis caused by scapholunate dissociation who were diagnosed in the same period. The following imaging features were evaluated: scapholunate dissociation; the center of disease and grade of radiocarpal joint destruction; characteristics of bone marrow edema; the presence of radial styloid and distal scaphoid osteophytes; and the extent of joint effusion and synovitis. RESULTS: The imaging diagnosis of PRCC was significantly differentiated from osteoarthritis associated with scapholunate dissociation based on occurrence at a younger age, bone marrow edema crossing the joint, center of disease in the proximal radioscaphoid joint, and absence of radial styloid and scaphoid osteophytes (P < 0.05). PRCC occasionally presented with arch-shape bone marrow edema based on the proximal carpal row. CONCLUSION: The diagnosis of PRCC can be aided if its characteristic imaging findings are differentiated from other disease entities in patients with a history of arthroscopy.

Additive values of pelvic tomosynthesis in comparison to pelvic radiography alone for the diagnosis of sacroiliitis in patients with suspected axial spondyloarthritis.

OBJECTIVES: To compare inter-reader agreement and diagnostic confidence in detecting sacroiliitis by the modified New York criteria (mNY) on digital radiography (DR) versus digital pelvic tomosynthesis assisted DR (DR+DPT), and to evaluate changes in the presence of axial spondyloarthritis (axSpA) according to the Assessment of Spondyloarthritis International Society (ASAS) criteria. METHODS: One hundred and thirty-eight patients who underwent both DR and DPT with suspicious axSpA in our rheumatologic clinic were included from February 2017 to February 2018. Three radiologists independently graded sacroiliitis and confidence level on DR first and then re-graded them on DPT in a paired manner. Agreement, confidence, and diagnostic accuracy were evaluated for readers. Changes in the presence of disease by mNY and ASAS criteria were assessed between DR alone and DR+DPT. RESULTS: On DR alone, 73 patients were assessed with radiographic sacroiliitis, and 85 were classified into axSpA by the ASAS criteria; however, 78 and 85, respectively, were classified on DR+DPT. With the assistance of DPT, 17 and 12 patients changed to the disease positivity according to the mNY and ASAS criteria, respectively; the negative results changed to positive in 11 and six patients, respectively. For all readers, agreement improved with DPT (0.79 to 0.89). DR+DPT achieved higher diagnostic accuracy (AUC, P < 0.05). CONCLUSION: The combination of DR and DPT achieved a higher diagnostic performance than that of DR alone, with better agreement. On DR+DPT, the diagnoses of 9.0% of patients with suspicious axSpA (12 of 134) were changed to the status of disease by the ASAS criteria.

The correlation between follow-up MRI findings and laboratory results in pyogenic spondylodiscitis.

BACKGROUND: Although MRI is the gold-standard imaging method in the diagnosis of spondylodiscitis, role of follow-up imaging is debated and there can be discrepancies with regard to the significance of bony or soft tissue responses to treatment. Purpose of our study is to test whether the MRI changes on follow-up imaging correlate with laboratory findings of treatment response. METHODS: A total of 48 patients with pyogenic spondylodiscitis who underwent baseline and follow-up MRI were retrospectively reviewed. The extent of bone marrow edema, paravertebral soft tissue inflammation, and disc height were compared on baseline and follow-up MRIs with the C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) levels obtained from the medical records at baseline and on follow-up. Relationships between the MRI and laboratory changes were analyzed using the Spearmann correlation test. RESULTS: The mean MRI follow-up period was 42.25 days. Based on the CRP (resolved: n = 19, resolving: n = 19, and aggravated: n = 10), there was significant correlation between the laboratory results and the changes in the bone and soft tissues (p < 0.01, both). The correlation was best with soft tissue changes (rho: 0.48) followed by bony changes (rho: 0.41). Based on the ESR (resolved: n = 8, resolving: n = 22, and worsened: n = 18), the correlation was stronger with bone changes (rho: 0.45, p < 0.01) than it was with soft tissue changes (rho: 0.39, p = 0.01). CONCLUSION: Follow-up MRI findings of pyogenic spondylodiscitis show variable tissue responses. CRP was best correlated with soft tissue changes, while ESR showed the best association with bony changes.

Sensor Fault Detection and Signal Restoration in Intelligent Vehicles.

This paper presents fault diagnosis logic and signal restoration algorithms for vehicle motion sensors. Because various sensors are equipped to realize automatic operation of the vehicle, defects in these sensors lead to severe safety issues. Therefore, an effective and reliable fault detection and recovery system should be developed. The primary idea of the proposed fault detection system is the conversion of measured wheel speeds into vehicle central axis information and the selection of a reference central axis speed based on this information. Thus, the obtained results are employed to estimate the speed for all wheel sides, which are compared with measured values to identify fault and recover the fault signal. For fault diagnosis logic, a conditional expression is derived with only two variables to distinguish between normal and fault; further, an analytical redundancy structure and a simple diagnostic logic structure are presented. Finally, an off-line test is conducted using test vehicle information to validate the proposed method; it demonstrates that the proposed fault detection and signal restoration algorithm can satisfy the control performance required for each sensor failure.

Multiparametric MR Imaging of Age-related Changes in Healthy Thigh Muscles.

Purpose To use multiparametric magnetic resonance (MR) imaging to assess for and establish age-related differences in healthy thigh muscles. Materials and Methods Ninety-five subjects (47 men, 48 women; median age, 47 years) with healthy body mass index were grouped according to age: 30-39 years (n = 25), 40-49 years (n = 25), 50-59 years (n = 25), and 60-69 years (n = 20). Multiparametric MR imaging (intravoxel incoherent motion diffusion-weighted, diffusion-tensor, multiecho Dixon, and dynamic contrast material-enhanced MR imaging) was performed at 3.0 T. Two radiologists independently evaluated parametric maps of the anterior, medial, and posterior compartments. Welch-modified one-way analysis of variance and post hoc Dunnet T3 test were used to evaluate differences in apparent diffusion, true diffusion, and pseudodiffusion coefficients; perfusion fraction; fractional anisotropy (FA); fat percentage; volume transfer constant; constant efflux rate from the extravascular-extracellular space to plasma; volume fraction of the extravascular-extracellular space (Ve); incremental area under the curve; and Pearson and Spearman correlation coefficients were used to evaluate relationship strength. Multiple regression analysis was performed to identify independent predictors of age, and interrater reliability was assessed with intraclass correlation coefficients. Results There were significant differences among the age groups in apparent diffusion coefficients (P = .010), true diffusion coefficients (P = .045), FA (P < .001), Ve (P = .029) of the anterior compartment muscles, and fat percentages of all three compartments (P ≤ .001). Moreover, FA (Pearson r = 0.428, Spearman ρ = 0.431; P < .001) and Ve (r = 0.226, P = .030 and ρ = 0.309, P = .003) in the anterior compartment and fat percentages in all three compartments (r = 0.481, 0.475, and 0.573; ρ = 0.515, 0.487, and 0.667; respectively; P < .001) were positively associated with age. Multiple regression analysis showed that age was predictive of fat percentage in the posterior compartment (ß = 0.500, P < .001) and of FA in the anterior compartment (ß = 0.194, P = .042). Interrater reliability was excellent (intraclass correlation coefficient, 0.745-0.992). Conclusion Multiple MR imaging parameters were significantly associated with age in thigh muscles. © RSNA, 2017 Online supplemental material is available for this article.

Intravoxel incoherent motion (IVIM) analysis of vertebral bone marrow changes after radiation exposure from diagnostic imaging and interventional procedures.

Background High cumulative radiation dose in cancer patients warrants systemic examination of possible changes in bone marrow. Purpose To assess retrospectively changes in vertebral bone marrow diffusion and perfusion using intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) in hepatocellular carcinoma (HCC) patients after exposure to radiation from diagnostic imaging and interventional procedures. Material and Methods A total of 21 IVIM-DWI sets in 20 HCC patients, consisting of baseline and follow-up liver magnetic resonance imaging (MRI) with an interval less than 100 days, were reviewed after varying levels of radiation exposure from transarterial chemoembolization (TACE), multiphase liver CT, and abdominal radiography. IVIM parameters (apparent diffusion coefficient [ADC], true diffusion coefficient [D], pseudodiffusion coefficient [D*], and perfusion fraction [PF]) of vertebral bone marrow were analyzed for significant differences between baseline and follow-up MRI using Wilcoxon signed-rank test, and for correlations with cumulative effective dose, as well as time interval between last radiation exposure and follow-up MRI using Spearman's correlation. Results Compared to baseline MRI, ADC, D*, and PF significantly decreased on follow-up MRI (ADC: median [interquartile range], 0.405 × 103 mm2/s [0.364-0.477] versus 0.390 [0.348-0.461]; D*: 24.011 [18.141-29.584] versus 20.815 [15.022-28.347]; PF: 10.960% [8.828-12.985] versus 9.125 [8.606-12.803]) ( P < 0.05). There was no significant difference in D ( P = 0.807). Cumulative effective dose was moderately correlated with decrease in D* (r = 0.434). In addition, longer intervals between last exposure and follow-up MRI showed negative correlations with changes in D and ADC (r = -0.352 and -0.333, respectively). Conclusion Vertebral bone marrow diffusion and perfusion parameters were significantly changed after exposure to medical radiation.

An intracortical chondromyxoid fibroma in the diaphysis of the metatarsal.

Chondromyxoid fibromas (CMFs) are rare, benign, primary tumors of bones, and occur in the metaphyses of the medullary canals of the long bones. The occurrence of intracortical CMFs is extremely rare. Very few cases of intracortical CMFs located in the long tubular bones have been reported to date. Moreover, even though the feet are the second most common site for CMF (after the knees), intracortical metatarsal CMF has not been reported previously, to our knowledge. We report an intracortical CMF occurring in the diaphysis of the metatarsal in a 17-year-old man. It showed the same imaging findings as usual intramedullary CMFs, except for its cortical location. The development and serial increase in this tumor over time are also demonstrated in this report. Additionally, we present a review of current literature on intracortical CMFs.

Intramuscular granular cell tumor: emphasizing the stripe sign.

We present a case of an intramuscular granular cell tumor in the thigh manifesting a 'stripe sign' on magnetic resonance imaging (MRI). This MR finding may correspond with the entrapped muscle fibers within the tumor, which demonstrated parallel arrangement on histopathologic examination. In addition, other disease entities characterized by the presence of this sign such as proliferative myositis, heterotopic ossification, and muscular sarcoidosis will be reviewed.

Quantitative analysis of lungs and airways with CT in subjects with the chronic obstructive pulmonary disease (COPD) candidate FAM13A gene: case control study for CT quantification in COPD risk gene.

OBJECTIVE: To investigate the relationship between a chronic obstructive pulmonary disease (COPD) candidate gene, based on a genomewide association study, and computed tomographic (CT) quantitative analysis; and to find a phenotype in the COPD candidate FAM13A gene. MATERIALS AND METHODS: This study was conducted in subclinical male smokers between 2 groups with matched age and smoking status; 162 subjects (mean age, 58 years) with risk (CTGA, n = 85) and reference (TCAG, n = 77) diplotypes replicated through genomewide association study underwent chest CT for quantitative analysis of lungs and airways. We analyzed the measures in both the risk and reference groups using a 2-sample t test. RESULTS: Subjects with the risk CTGA diplotype had significantly higher total lung volume and emphysema index than the reference TCAG diplotype (P = 0.04). Mean lung density was significantly lower (P < 0.05) in the risk group. However, in the analysis of airways, wall area, luminal area, wall and lumen area ratio, and mean lung density on expiratory and inspiratory phases, no significant differences between the 2 groups were seen. CONCLUSIONS: There is a strong relationship between CT quantitative analysis and the COPD candidate gene. Furthermore, the CTGA diplotype was associated with emphysema among the phenotypes of COPD.

The sonographic "coffee bean" sign helps distinguish an axillary neurofibroma from a lymphadenopathy.

Axillary masses may represent various soft tissue tumors or lymphadenopathy. Neurofibromas are benign peripheral nerve sheath tumors and, while they are very uncommon, it is important to remember that neurogenic tumors arising from brachial plexus can develop in the axilla. We describe an axillary neurofibroma arising from the brachial plexus that presented with a "coffee bean sign" on sonography that distinguished it from axillary lymphadenopathy.

T2 Black Synovitis in Musculoskeletal MRI: Disease Spectrum and Imaging Characteristics of Joint Diseases

The synovium may be affected by a wide spectrum of disorders, including inflammatory, infectious, degenerative, traumatic, hemorrhagic, and tumorous conditions. Magnetic resonance imaging (MRI) is a valuable imaging modality to characterize synovial disorders. Most abnormal lesions appear as areas of nonspecific high signal intensity on T2-weighted images (T2-WI) due to high water content or increased perfusion. However, T2 hypointensity can be attributed to blood components of varying ages, calcification, inorganic crystals, fibrosis, caseous necrosis and/or amyloid deposition. Hypointense lesions on T2-WI are infrequent and additional clinical and imaging characteristics can help to limit the list of differential diagnoses, which may include tenosynovial giant cell tumor, synovial chondromatosis, rheumatoid arthritis, tuberculous arthritis, chronic tophaceous gout, amyloid arthropathy, synovial hemangioma, lipoma arborescens and hemosiderotic synovitis. Recently, susceptibility weighted imaging has been developed and may contribute to more accurate diagnosis for deoxygenated blood and calcium. We review the MRI features of hypointense synovial lesions on MRI and emphasize the characteristic findings that suggest a correct diagnosis.

Subsidence Performance of the Bioactive Glass-Ceramic (CaO-SiO2-P2O5-B2O3) Spacer in Terms of Modulus of Elasticity and Contact Area: Mechanical Test and Finite Element Analysis.

OBJECTIVE: The objective of this study is to evaluate the subsidence performance of a bioactive glass-ceramic (CaO-SiO2-P2O5-B2O3) spacer in terms of its modulus of elasticity and contact area using mechanical tests and finite element analysis. METHODS: Three spacer three-dimensional models (Polyether ether ketone [PEEK]-C: PEEK spacer with a small contact area; PEEK-NF: PEEK spacer with a large contact area; and Bioactive glass [BGS]-NF: bioactive glass-ceramic spacer with a large contact area) are constructed and placed between bone blocks for compression analysis. The stress distribution, peak von Mises stress, and reaction force generated in the bone block are predicted by applying a compressive load. Subsidence tests are conducted for three spacer models in accordance with ASTM F2267. Three types of blocks measuring 8, 10, and 15 pounds per cubic foot are used to account for the various bone qualities of patients. A statistical analysis of the results is conducted using a one-way Analysis of variance and post hoc analysis (Tukey's Honestly Significant Difference) by measuring the stiffness and yield load. RESULTS: The stress distribution, peak von Mises stress, and reaction force predicted via the finite element analysis are the highest for PEEK-C, whereas they are similar for PEEK-NF and BGS-NF. Results of mechanical tests show that the stiffness and yield load of PEEK-C are the lowest, whereas those of PEEK-NF and BGS-NF are similar. CONCLUSIONS: The main factor affecting subsidence performance is the contact area. Therefore, bioactive glass-ceramic spacers exhibit a larger contact area and better subsidence performance than conventional spacers.

Clinical Validation of an Artificial Intelligence Model for Detecting Distal Radius, Ulnar Styloid, and Scaphoid Fractures on Conventional Wrist Radiographs.

This study aimed to assess the feasibility and performance of an artificial intelligence (AI) model for detecting three common wrist fractures: distal radius, ulnar styloid process, and scaphoid. The AI model was trained with a dataset of 4432 images containing both fractured and non-fractured wrist images. In total, 593 subjects were included in the clinical test. Two human experts independently diagnosed and labeled the fracture sites using bounding boxes to build the ground truth. Two novice radiologists also performed the same task, both with and without model assistance. The sensitivity, specificity, accuracy, and area under the curve (AUC) were calculated for each wrist location. The AUC for detecting distal radius, ulnar styloid, and scaphoid fractures per wrist were 0.903 (95% C.I. 0.887-0.918), 0.925 (95% C.I. 0.911-0.939), and 0.808 (95% C.I. 0.748-0.967), respectively. When assisted by the AI model, the scaphoid fracture AUC of the two novice radiologists significantly increased from 0.75 (95% C.I. 0.66-0.83) to 0.85 (95% C.I. 0.77-0.93) and from 0.71 (95% C.I. 0.62-0.80) to 0.80 (95% C.I. 0.71-0.88), respectively. Overall, the developed AI model was found to be reliable for detecting wrist fractures, particularly for scaphoid fractures, which are commonly missed.

Dorsal sural neuropathy: electrophysiological, ultrasonographic, and magnetic resonance imaging findings.

Dorsal sural neuropathy occurs infrequently as a result of compression of the nerve due to various etiologies.We describe a 52-year-old diabetic woman with dorsal sural neuropathy. The diagnosis was based on clinical, electrophysiological, ultrasonographic, and magnetic resonance imaging (MRI) findings. This case report demonstrates the usefulness of electrophysiological and imaging studies, including ultrasonography, in the diagnosis of dorsal sural neuropathy.

Lumbar Spine Computed Tomography to Magnetic Resonance Imaging Synthesis Using Generative Adversarial Network: Visual Turing Test.

(1) Introduction: Computed tomography (CT) and magnetic resonance imaging (MRI) play an important role in the diagnosis and evaluation of spinal diseases, especially degenerative spinal diseases. MRI is mainly used to diagnose most spinal diseases because it shows a higher resolution than CT to distinguish lesions of the spinal canals and intervertebral discs. When it is inevitable for CT to be selected instead of MR in evaluating spinal disease, evaluation of spinal disease may be limited. In these cases, it is very helpful to diagnose spinal disease with MR images synthesized with CT images. (2) Objective: To create synthetic lumbar magnetic resonance (MR) images from computed tomography (CT) scans using generative adversarial network (GAN) models and assess how closely the synthetic images resembled the true images using visual Turing tests (VTTs). (3) Material and Methods: Overall, 285 patients aged ≥ 40 years who underwent lumbar CT and MRI were enrolled. Based on axial CT and T2-weighted axial MR images from 285 patients, an image synthesis model using a GAN was trained using three algorithms (unsupervised, semi-supervised, and supervised methods). Furthermore, VTT to determine how similar the synthetic lumbar MR images generated from lumbar CT axial images were to the true lumbar MR axial images were conducted with 59 patients who were not included in the model training. For the VTT, we designed an evaluation form comprising 600 randomly distributed axial images (150 true and 450 synthetic images from unsupervised, semi-supervised, and supervised methods). Four readers judged the authenticity of each image and chose their first- and second-choice candidates for the true image. In addition, for the three models, structural similarities (SSIM) were evaluated and the peak signal to noise ratio (PSNR) was compared among the three methods. (4) Results: The mean accuracy for the selection of true images for all four readers for their first choice was 52.0% (312/600). The accuracies of determining the true image for each reader's first and first + second choices, respectively, were as follows: reader 1, 51.3% and 78.0%; reader 2, 38.7% and 62.0%, reader 3, 69.3% and 84.0%, and reader 4, 48.7% and 70.7%. In the case of synthetic images chosen as first and second choices, supervised algorithm-derived images were the most often selected (supervised, 118/600 first and 164/600 second; semi-supervised, 90/600 and 144/600; and unsupervised, 80/600 and 114/600). For image quality, the supervised algorithm received the best score (PSNR: 15.987 ± 1.039, SSIM: 0.518 ± 0.042). (5) Conclusion: This was the pilot study to apply GAN to synthesize lumbar spine MR images from CT images and compare training algorithms of the GAN. Based on VTT, the axial MR images synthesized from lumbar CT using GAN were fairly realistic and the supervised training algorithm was found to provide the closest image to true images.