Published papers

Background: Cervical segmental rotation has rarely been quantified in physiological upright, weight-bearing positions. This study aimed to characterize cervical rotational kinematics using upright multidetector CT (MDCT) and to evaluate the independent associations between rotational parameters, age, sex, and cervical sagittal alignment. Methods: Forty-nine healthy volunteers—24 younger (21–40 years) and 25 elderly (61–79 years)—underwent upright MDCT in an upright standing position and maximal right cervical rotation. Rotational angles (O–C7) were measured relative to T1. Parameters included segmental rotational range of motion (ROM) (rsd: intervertebral rotation change from neutral to maximal rotation), rotated segmental position (rd: intervertebral rotation at maximal rotation), and percentage segmental contribution (%rsd). Results: The largest segmental ROM occurred at C1–2 (median rsdC1: 34.3°), accounting for 66.1% of total rotation. Compared with younger participants, the elderly group showed significantly reduced total rotation and rsdC1 (30.3° vs. 37.0°, p = 0.008). Correlation analysis showed that age correlated negatively with rsdC1 (r = −0.31) and positively with rsdC7 (r = 0.29). Multivariable regression identified elderly age as a significant negative predictor of rsdC1 (β = −5.59, p = 0.004). In contrast, rdC1 was independently associated with age (β = −6.05, p < 0.001), male sex (β = −5.30, p = 0.003), and smaller C2–7 angle (β = −0.14, p = 0.020). Conclusions: Under physiological weight-bearing conditions, cervical rotation is predominantly generated at C1–2. Aging reduces upper cervical segmental ROM, potentially leading to compensatory motion at the cervicothoracic junction. While rotational ROM is primarily dictated by age, the intervertebral position at maximal rotation is independently influenced by age, sex, and sagittal alignment. These normative data may serve as a reference for future studies examining cervical disorders and postoperative rotational function.

Understanding the transmission of forces through the joints of the lower limb under weight-bearing conditions is critical for assessing the progression of knee and ankle osteoarthritis, as well as for developing effective surgical and preventive strategies. This study presents the first simultaneous quantification of three-dimensional skeletal posture and ground reaction force vectors during quiet standing, utilizing upright computed tomography and a custom-made force plate. We constructed the force plate using two six-axis force sensors, and the coordinate system of the CT was aligned with that of the force plate using metal spheres attached to the edges of the force plate. This system enabled precise measurement of ground reaction force vectors relative to the skeletal posture. The present study demonstrated that the load-bearing axis (the line connecting the femoral head center to the midpoint of the distal tibial plafond) in healthy subjects during quiet standing is not identical to the actual line of force application. The analysis revealed that the force vectors generally pass slightly medioposterior to the femoral head center. By providing a detailed view of how force is transmitted through the lower limb joints, this system potentially offers new insights into joint loading patterns that possibly contribute to osteoarthritis development.

BACKGROUND: Hand-behind-back (HBB) motion is commonly used to assess shoulder internal rotation; however, it involves multiple joint movements, including scapulothoracic, glenohumeral, and elbow joint motions, thus complicating the interpretation of the HBB position. This study aimed to investigate 3-dimensional joint alignment during HBB motion of the bilateral shoulders in patients with frozen shoulder and to identify which joint motions primarily contribute to motion limitation. METHODS: Seventeen patients with unilateral frozen shoulder underwent bilateral shoulder computed tomography scans in the neutral and HBB positions using an upright multidetector computed tomography. Bone surface models of the thorax, scapula, humerus, and forearm were created to calculate the rotation of the scapulothoracic, glenohumeral, thoracohumeral, and elbow joints. The affected and unaffected sides were compared, and the correlation between the HBB reach level (vertebral level) and the thoracohumeral internal rotation angle was analyzed. RESULTS: No significant differences in joint rotation were observed between the sides in the neutral position. The affected side demonstrated significantly reduced glenohumeral internal rotation (31° vs. 66°, P < .001), abduction (10° vs. 22°, P < .001), and elbow flexion (74° vs. 115°, P < .001) during HBB motion. The scapulothoracic joint on the affected side had less internal rotation (18° vs. 21°; P = .045) and more anterior tilt (24° vs. 19°; P = .003). The HBB reach level was negatively correlated with the thoracohumeral internal rotation angle in the unaffected and affected sides (r = -0.518 and r = -0.675, respectively). CONCLUSION: This study clarified that limited internal rotation of the glenohumeral joint is the main cause of restricted HBB motion in frozen shoulder, which supports the use of the thumb-to-spinous process as a valid measure of thoracohumeral internal rotation in patients with frozen shoulder.

This study investigates the effects of fall configurations on hip fracture risk with a focus on pelvic soft tissue shape. This was done by employing a whole-body finite element (FE) model. Soft tissue thickness around the pelvis was measured using a standing CT system, revealing a trend of increased trochanteric soft tissue thickness with higher BMI and younger age. In the lateroposterior region from the greater trochanter, the soft tissues of elderly females were thin with a concave shape. Based on the THUMS 5F model, an elderly female FE model with a low BMI was developed by morphing the soft tissue shape around the pelvis based on the CT data. FE simulation results indicated that the lateroposterior fall led to a higher femoral neck force for the elderly female model compared to the lateral fall. One reason may be related to the thin soft tissue of the pelvis in the lateroposterior region. Additionally, the effectiveness of interventions that can help mitigating hip fractures in lateroposterior falls on the thigh-hip and hip region was assessed using the elderly female model. The attenuation rate of the femoral neck force by the hip protector was close to zero in the thigh-hip fall and high in the hip fall, whereas the attenuation rate of the compliant floor was high in both falls. This study highlights age-related changes in the soft tissue shape of the pelvis in females, particularly in the lateroposterior regions, which may influence force mitigation for the hip joint during lateroposterior falls.

BACKGROUND: Coronary artery calcification (CAC) detected through chest computed tomography (CT) strongly predicts cardiovascular events in asymptomatic individuals undergoing primary prevention. Few studies with limited sample sizes have investigated the predictive value of CAC for cardiovascular complications in COVID-19. This study examined the impact of CAC on cardiovascular complications using a large-scale COVID-19 database. METHODS AND RESULTS: This multicenter retrospective cohort study used data from the Japan COVID-19 Task Force database. After exclusion based on missing information, 1,109 patients with COVID-19 were included. The Agatston score was used to evaluate CAC, dividing the population into 3 groups based on calcification degree (no, moderate, and severe CAC). The primary outcome was cardiovascular complications; the secondary outcome was critical outcomes. The severe CAC group had a higher rate of cardiovascular complications than the other groups. Multivariable analysis, considering COVID-19 severity factors, identified severe CAC as independently associated with cardiovascular complications but not with critical outcomes. Subgroup analysis revealed that, in patients without hypertension, diabetes, cardiovascular disease, or chronic kidney disease, severe CAC was significantly correlated with cardiovascular complications, whereas this association was not observed in patients with these underlying conditions. CONCLUSIONS: Patients with COVID-19 and severe CAC had increased cardiovascular complications, and identifying cardiovascular and pulmonary findings on chest CT is essential. Measuring CAC via non-electrocardiogram-gated CT helps predict patient risk.

PURPOSE: MR cholangiopancreatography (MRCP) often suffers from signal loss because of concentrated bile. Gradient-and-spin-echo (GRASE) MRCP reduces this effect but is limited by breath-holding. We developed MRCP with centric order encoding (MRCP-CORE), a novel free-breathing sequence with shorter TE using centric k-space trajectory, T2 preparation, and velocity encoding, and evaluated its ability to visualize gallbladder. METHODS: In this prospective, single-center study, we enrolled 100 consecutive patients who underwent 3 MRCP sequences (MRCP-CORE, conventional MRCP, and GRASE MRCP) on a 3T system. Two blinded radiologists independently assessed the visibility of anatomical structures, artifacts, and overall image quality. SNR, contrast-to-noise ratio, and contrast ratio of the periductal tissue and gallbladder were measured. The Friedman test was used for comparisons; Jonckheere-Terpstra analysis tested the association between the gallbladder visibility and quantitative metrics. RESULTS: Ninety-nine patients (mean age, 68.5 ± 10.9 years; 52 men, 47 women) were evaluated. MRCP-CORE showed significantly higher SNR than conventional and GRASE MRCP (20.9 vs. 17.3 vs. 15.8; P < 0.001). Gallbladder visibility was significantly higher with MRCP-CORE than via conventional MRCP (P = 0.02) and comparable to that via GRASE MRCP (P = 1.00). No significant differences in artifact and overall image quality were observed among sequences. Trend analysis revealed gallbladder visibility was significantly influenced by bile concentration in conventional MRCP (P < 0.001), but not in MRCP-CORE (P = 0.12) or GRASE MRCP (P = 0.43). CONCLUSION: MRCP-CORE enhanced gallbladder depiction compared with conventional MRCP, offering a robust free-breathing alternative to breath-hold GRASE MRCP.

Diabetes mellitus is a systemic metabolic disease that causes a wide spectrum of complications affecting multiple organs. Advances in imaging modalities have substantially improved the detection, characterization, and clinical management of these conditions. However, the imaging manifestations of diabetic complications are highly diverse, often mimicking other diseases and posing diagnostic challenges for radiologists. This review aimed to provide an organ-based overview of the characteristic imaging features of diabetic complications from head to toe, integrating pathophysiological mechanisms relevant to image interpretation. Representative cases are illustrated using multimodal imaging—computed tomography (CT), magnetic resonance imaging, ultrasound, radiography, and 18F-fluorodeoxyglucose positron emission tomography/CT—to highlight key findings, differential diagnoses, and potential pitfalls. Specifically, this review focuses on conditions and diseases that are amenable to evaluation by radiological imaging. Understanding these imaging patterns is essential for early diagnosis, risk stratification, and optimal patient management. Through an educational, pictorial approach, this review aims to enhance radiologists’ recognition of diabetic complications across organ systems.

Upright multidetector CT (MDCT) enables whole-body, high-resolution, three-dimensional cross-sectional imaging in standing or sitting positions, providing detailed visualization of anatomy and biomechanics in physiologically relevant, load-bearing postures. This system maintains high spatial resolution and quantitative performance, including stable noise characteristics and consistent CT attenuation values, while improving workflow efficiency compared with conventional supine MDCT. In the shoulder and upper extremity, upright MDCT has clarified gravity-dependent alignment changes and enabled dynamic four-dimensional assessment of sternoclavicular and acromioclavicular joint motion. In the spine, it provides reproducible spinopelvic measurements, reveals posture-dependent narrowing of intervertebral foramina, and characterizes age- and sex-related differences in segmental spinal mobility. Upright MDCT of the knee allows precise evaluation of joint-line orientation, mechanical-axis migration, and rotational kinematics under weight-bearing conditions, enhancing diagnostic accuracy and preoperative planning in osteoarthritis. In the lower extremities, it captures kinetic-chain mechanics from the foot to the hip, integrating ground-reaction-force data to analyze load distribution and compensatory motion across adjacent joints. Upright MDCT also visualizes thoracic cage deformation and vertebral displacement in conditions such as pectus excavatum, demonstrating load-related changes that supine imaging may overlook. This review provides a state-of-the-art overview of upright MDCT applications in orthopedic imaging and highlights its potential to transform diagnostic strategies and deepen understanding of musculoskeletal pathophysiology under physiologically relevant weight-bearing conditions. Collectively, these studies demonstrate that upright MDCT is a quantitatively robust modality that offers new insights into functional anatomy, disease mechanisms, and biomechanical adaptation. By bridging diagnostic imaging with functional assessment, upright MDCT has the potential to refine understanding of musculoskeletal pathophysiology and to guide more personalized therapeutic strategies in orthopedic and related fields.

OBJECTIVE: Postural changes influence systemic hemodynamics and may affect cerebral perfusion; however, quantitative evaluation remains limited in vivo under physiological gravitational conditions. We aimed to assess posture-dependent alterations in cerebral perfusion using upright multidetector computed tomography perfusion (CTP), with whole-brain cerebral blood flow (CBF) as the primary outcome. MATERIALS AND METHODS: This was a single-center observational study of healthy volunteers approved by the institutional review board. Eighteen volunteers underwent CTP imaging in both the supine and seated positions on the same day using two 320-row multidetector CT scanners with identical systems. Whole-brain CBF was defined as the primary endpoint. Mean transit time (MTT) and cerebral blood volume (CBV) were evaluated as secondary outcomes. Paired comparisons were performed between postures, and exploratory analyses assessed regional changes and interindividual variability. RESULTS: Whole-brain CTP-derived CBF was significantly higher in the seated position than in the supine position (32.9 ± 6.0 [mean ± standard deviation] versus 29.2 ± 5.6 mL/100 g/min; p = 0.010), representing a 1.15 ± 0.19-fold increase. CTP-derived MTT was significantly shorter in the seated position (4.25 ± 0.61 versus 4.63 ± 0.72 s; p = 0.035), and CTP-derived CBV showed a modest but significant increase (2.23 ± 0.16 versus 2.16 ± 0.17 mL/100 g; p = 0.044). Regional perfusion changes were consistent across brain segments with interindividual variability in postural responses. CONCLUSIONS: Upright CTP detected posture-associated differences in whole-brain CTP-derived perfusion metrics, suggesting potential utility for investigating physiological and pathological alterations in cerebral hemodynamics associated with postural change. RELEVANCE STATEMENT: Upright CTP enables physiological evaluation of posture-related cerebral hemodynamics under gravitational loading and may provide new insights into cerebral circulation in both health and disease. KEY POINTS: Whole-brain cerebral blood flow increased by 15% in the seated position compared with supine using upright CT perfusion. Mean transit time shortened by 7% and cerebral blood volume increased modestly by 4% in the seated position. Perfusion responses showed marked interindividual variability, suggesting different physiological adaptations to posture.

Childhood interstitial lung disease (chILD) comprises a diverse group of rare disorders, which were previously poorly categorized on clinical, histologic, and pathogenic grounds. The classification of chILD introduced by the chILD Research Network has led to a systematic approach to identifying and categorizing these rare conditions. A multidisciplinary approach integrating clinical, radiologic, pathologic, and genetic findings enables accurate diagnosis in children with "pediatric interstitial pneumonia of uncertain cause." It has become apparent that chILDs are collectively frequent in pediatric practice, and prompt diagnosis can improve the outcomes of affected children. Since the clinical symptoms of chILDs are nonspecific, imaging using high-resolution CT plays a central role in the diagnostic approach to this diverse group of diseases. In most disorders, cellular infiltration and abnormal exudates commonly cause acute-phase imaging findings, such as centrilobular ground-glass opacities that are multifocal, geographic, and diffuse, as well as a crazy-paving appearance. Persistent inflammatory reactions lead to chronic-phase findings, including reticulation, lung cysts, and honeycombing with traction bronchiectasis. Small-airway compromise causes air trapping and a mosaic attenuation pattern. These imaging findings are individually nonspecific; however, meticulous integration of the diverse imaging findings allows a specific diagnosis to be reached. The authors review the current knowledge on the imaging findings associated with chILDs based on recent investigations of these conditions. ©RSNA, 2026 Supplemental material is available for this article.

A dual imaging pattern of hyperattenuation on non-enhanced computed tomography and hypointensity on T2-weighted imaging is frequently encountered in various lesions. In the present review, we aimed to explore this imaging combination across a spectrum of abdominal and pelvic lesions to clarify its clinical implications.This pattern is commonly observed in lesions rich in fibrous or smooth muscle components, making it more frequently encountered in benign entities, including fat-poor angiomyolipomas, uterine leiomyomas, and tumors with predominant fibrous stroma. However, similar imaging findings may also be encountered in rare high-cellularity malignancies, selected systemic diseases, and lesions containing melanin, metal deposition, or thyroid tissue. Although not specific to a single disease, this imaging pattern offers valuable diagnostic clues for lesion characterization. When contrast-enhanced imaging is unavailable or contraindicated, combining non-enhanced computed tomography and T2-weighted imaging findings may help narrow the differential diagnosis. Despite its general association with benign lesions, this imaging phenotype requires careful exclusion of important pathologic mimics. Overall, integration with clinical information and other imaging findings remains essential for accurate interpretation.

Background: Carbon ion radiotherapy (CIRT) in the upright posture is a treatment technique which delivers carbon ion beams from a fixed direction to patients who sit or stand on a rotating chair. Although various potential advantages of the upright positioning have been discussed, feasibility of upright CIRT for prostate cancer has not been quantitatively evaluated. Purpose: This study demonstrated prostate cancer CIRT in the upright posture and evaluated its feasibility by comparing anatomy and dosimetric metrics between the supine and upright postures. Methods: A total of 12 pairs of computed tomography (CT) images of asymptomatic volunteers in the supine and upright postures were retrospectively analyzed. Based on a clinical treatment protocol, clinical target volume (CTV) was defined as being a prostate gland and proximal seminal vesicle. Other target and organ-at-risk (OAR) contours within a planning volume were also delineated. A CIRT plan was calculated for each CT volume with the same dose prescription and evaluation criteria. The delineated contour volumes, 16 dosimetric metrics, dose-volume histograms (DVHs) of CTV and OARs, and robustness against setup and range uncertainties were compared between the supine and upright postures. Results: No significant differences in the contour volumes were found between the two postures (p > 0.05). The CTV and rectum contour volumes were 34.30 ± 7.76 and 36.35 ± 13.53 cm3 for the supine posture and 35.89 ± 7.37 and 32.96 ± 8.08 cm3 for the upright posture, respectively. Moreover, mean absolute differences of five CTV dosimetric metrics (D2%, D5%, D50%, D95%, and D98%) were all less than 0.15 Gy (0.29% of the prescription dose of 51.6 Gy). Nine other dosimetric metrics in the upright posture were also equivalent to those in the supine posture. The DVHs and robustness in the upright posture were in agreement with those in the supine posture. Conclusions: The upright CIRT provided dose distributions with target dose coverage and OAR dose sparing equivalent to those of the conventional supine CIRT. Upright CIRT for prostate cancer is expected to have equivalent treatment effectiveness while offering reduced installation costs and simple system management of platforms.