This study aimed to establish sex-specific cutoff values for sarcopenia based on the muscle parameters skeletal muscle index (SMI) and psoas muscle thickness per height (PMTH) using CT scans in a presumably healthy Caucasian study cohort. In this retrospective, cross-sectional study, CT scans of 350 Caucasian patients (mean age, 32 years ± 8; mean BMI, 23 (women) and 24 kg/m2 (men)) with polytrauma, but otherwise young and presumably healthy, were evaluated. Sex-specific PMTH at the level of the umbilicus (mm/m) and the SMI (cm2/m2) at the level of the L3 vertebral body were calculated from cross-sectional images. Two standard deviations below the mean were considered the cutoff values for sarcopenia according to the recommendations of the European Working Group on Sarcopenia in Older People (EWGSOP). To exclude random outliers, only the 5th–95th percentiles were analyzed. The sex-specific cutoff values for PMTH were 15.7 mm/m for women and 20.2 mm/m for men, while those for SMI were 33.2 and 44.0 cm2/m2, respectively. The intra-rater and inter-rater variabilities were consistently low, indicating the high reliability of the measurement method. This study established sex-specific thresholds for sarcopenia based on two different muscle parameters in a presumably healthy Caucasian study cohort according to EWGSOP recommendations. Question Establishment of sarcopenia cutoff values according to the recommendations of the European Working Group on Sarcopenia in Older People (EWGSOP). Findings Cutoff values for psoas muscle thickness per height (mm/m) were 15.7 (women) and 20.2 (men); for skeletal muscle index (cm2/m2), 33.2 and 44.0, respectively. Clinical relevance These cutoff values for two different parameters that can be determined by CT allow patients with low muscle mass to be objectively identified, thereby revealing those at high risk of manifest sarcopenia.

Abstract To externally validate the RSNA 2020 challenge 2 nd -place deep-learning (DL) algorithm for detecting pulmonary embolism (PE) on computed tomography pulmonary angiography (CTPA) scans with additional focus on subsegmental-only PEs (SSPE). Between 2015 and 2022 1,038 CTPAs were retrospectively enrolled. An experienced radiologist (> five years) labelled CTPAs for the presence, location (central, lobar and segmental, subsegmental-only) and side (right-sided, left-sided, bilateral) of PE. The DL model was tested for its ability to predict these labels by analyzing accuracy, sensitivity, specificity and area under the receiver operating characteristic curve (AUROC) using different model outcome probabilities for each analysis. Overall, 136 central, 375 peripheral PEs (303 lobar and segmental PEs, 72 SSPEs), and 527 patients without any PE were analyzed. The model correctly predicted the presence of any PE in 921/1,038 patients (88.7%), yielding a sensitivity of 0.80, specificity of 0.97 and AUROC of 0.94. No central PE was missed, whereas 100/375 (26.7%) peripheral PEs remained undetected. Using the corresponding model output probability, the central PE status was correctly identified in 999/1,038 (96.2%) patients, while peripheral PE was correctly identified in 785/1,038 (75.6%) patients. This corresponded to sensitivities of 0.94 and 0.77, specificities of 0.97 and 0.75, and AUROC values of 0.99 and 0.74, respectively. The model performed better in detecting right-sided compared to left-sided PEs (AUROC: right- vs. left-sided: 0.95 and 0.92, p < 0.05). SSPE status was correctly predicted in 796/1,038 (76.7%) patients, yielding a sensitivity of 0.94, a specificity of 0.22, and an AUROC of 0.48, respectively. The model demonstrated high performance in detecting any PE, performing best on central and slightly lower on lobar and segmental PE. Predicting the presence of SSPEs was difficult, likely due to the model not being explicitly trained for this subtype.

Sarcopenia is an important prognostic factor in gastrointestinal malignancies, but CT-based definitions vary. This study evaluated recently published computed tomography (CT)-based cut-offs for skeletal muscle index (SMI) and psoas muscle thickness normalized to height (PMTH), derived from a young, presumably healthy reference population according to recommendations of the European Working Group on Sarcopenia in Older People (EWGSOP), in patients undergoing surgery for gastric cancer. This retrospective, single-center cohort study included patients who underwent surgery for gastric cancer between 2013 and 2018. Preoperative CT was used to assess SMI and PMTH. The recently published reference cut-offs were compared with the established definitions of Prado et al. (SMI) and Gu et al. (PMTH). Overall survival (OS) and recurrence-free survival (RFS) were the primary and secondary endpoints, respectively. Data from 212 patients with complete 5-year follow-up were analyzed. The reference cut-offs significantly discriminated OS and RFS (SMI, both p < 0.001; PMTH, both p = 0.002), whereas the established cut-offs showed weaker or no discrimination. Patients with both low SMI and low PMTH had the worst OS and RFS (both p < 0.001). Combined low SMI and low PMTH was associated with worse OS in univariable analysis (HR 2.57, 95

Objective Osteoarthritis (OA) is a common finding in the elderly population. Femoroacetabular Impingement (FAI)-related deformities can cause OA. Their prevalence and clinical relevance in the asymptomatic elderly population are unclear. This study aims to determine the prevalence of cam and pincer deformities in asymptomatic, non-orthopedic Caucasian patients over 60 years in computed tomography (CT) and correlate the presence of those deformities with radiological signs of OA. Design CTs of oncological patients over 60 years without any previous history of hip pathology or pain, containing a complete scan of both hips, were included. All hips were analysed for signs of OA according to the Kellgren-Lawrence (KL) classification. The acetabular anteversion, acetabular sector angles (ASA), and alpha angles were measured in clockwise positions of the hip. All angles were modeled conditionally on KL, sex, age, and obesity using random-effects median regression. Results One hundred ninety-five non-orthopedic patients (130 male, 65 female; mean age 70.4 years; 290 hips) were included. Signs of severe OA (KL 3/4) were found in 19.2 %. Patients over 80 years had significantly higher alpha angles at 9-11 o'clock, higher ASA at 9-12 o'clock, and a higher acetabular anteversion at the level of the acetabular roof. Individuals with advanced OA showed higher alpha angles, higher ASA, and a lower acetabular anteversion than those without/with moderate OA. Conclusions This study provides an estimate of the distribution of FAI-related deformities within a population of individuals over 60 years with asymptomatic hips. The prevalence of these deformities is correlated with advanced OA.

Background:Artificial intelligence (AI) is emerging as a transformative force in radiology, offering the potential to revolutionize the field by enabling sophisticated analysis of complex radiological data and uncovering previously unknown information in medical images. Materials & Methods:About a decade after the introduction of clinically applicable AI tools, this article explores the current status, opportunities, and limitations of AI integration in radiological practice. We discuss the growing demand for imaging services, increasing complexity of imaging data, and anticipated workforce shortages. Moreover, the role of large language models, computer vision, and automation in improving diagnostic accuracy, workflow efficiency, and patient communication is highlighted. We also examine the evolving European regulatory framework, including the AI Act, MDR (Medical Device Regulation), and EHDS (European Health Data Space), and their implications for the safe and ethical deployment of AI in clinical settings. Results:Radiology, as a highly digitalized and data-rich specialty, is uniquely positioned to benefit from AI-driven innovations across the entire clinical workflow - from patient scheduling to diagnosis and report generation. Challenges, such as the increasing complexity of imaging data or workforce shortages, further underscore the need for selective, well-validated AI-supported solutions. Despite its promise, current limitations such as data quality, model interpretability, or integration barriers, as well as lack of reimbursement, remain critical challenges. Conclusion:This review underscores the need for thoughtful implementation to fully realize AI's potential as an enabling infrastructure in radiology that makes imaging-based healthcare more efficient, accurate, and accessible. Key Points:· Artificial intelligence is emerging as a transformative force in diagnostic and interventional radiology.. · This article explores the status, opportunities, and limitations of clinically applicable AI tools.. · The article reviews the evolving European regulatory framework for AI deployment.. · The review highlights the need for interdisciplinary collaboration and well-planned AI implementation based on benefits and evidence.. Citation Format:· Bamberg F, Adam G, Antoch G et al. Artificial Intelligence in Radiology: Unlocking New Dimensions of Value. Rofo 2026; DOI 10.1055/a-2794-9496.

Background Volumetric segmentation in CT and MRI is valuable for artificial intelligence workflows in radiology, yet creating the large, precisely annotated datasets required for training segmentation models remains laborious. Methods Here, we tested in simulation whether the foundation model "Segment Anything Model 2" (SAM 2) can reduce expert annotation workload. In our workflow, annotators provide a single box at the object's center, and SAM 2 automatically segments the object across slices; annotators then review and correct the masks as needed. Workload reduction was defined as the proportion of SAM 2's predicted segmentation masks that were accepted without modification. Downstream segmentation models were then trained on the SAM 2-assisted masks and compared with reference models trained on ground truth masks. Results For femoral bone segmentation in MRI and liver tumor segmentation in CT, 36,614 sagittal and 16,311 axial slices were annotated, with 30% and 53% of SAM 2-generated masks accepted as is, respectively, indicating workload reduction. Crucially, segmentation models trained on SAM 2-assisted masks performed comparably to reference models, with a median dice similarity coefficient of 98.5% compared with 98.7% for femoral bone segmentation, and 77.3% compared with 77.0% for liver tumor segmentation. Conclusion Using SAM 2 could thus expedite 3D medical imaging dataset annotation and segmentation model development for both research and clinical applications.

Objectives:To assess the agreement of the three methods of coronary plaque burden assessment approved by the Coronary Artery Disease-Reporting and Data System classification (CAD-RADS 2.0): coronary artery calcium (CAC) testing, segment involvement score (SIS), and visual estimate. Methods:Patients with stable chest pain who underwent coronary computed tomography angiography (CCTA) and showed signs of coronary artery disease were included in the current retrospective study. The CCTAs were independently evaluated by two board-certified radiologists. Differing results were resolved in consensus. Three separate estimates of plaque burden (P) were performed for each examination: P(CAC), P(SIS) and P(visual). Linearly weighted kappa was used to assess the agreement of the three methods. Results:The study included 132 patients (mean age 66.6 years, 68.2% male). The mean CAC score was 503.6 ± 1023.5 [SD]. CAD-RADS was 1 in 47/132 patients (35.6%), 2 in 40/132 patients (30.3%), 3 in 18/132 patients (13.6%), 4 in 24/132 cases (18.2%) and 5 in 3/132 cases (2.3%). Weighted Kappa values for method agreement were κ= 0.56 (95% CI: 0.48-0.65) for agreement between P(CAC) and P(SIS), κ= 0.67 (95% CI: 0.59-0.75) for agreement between P(CAC) and P(visual), and κ= 0.72 (95% CI: 0.64-0.80) for agreement between P(SIS) and P(visual), signifying moderate to substantial agreement. Conclusion:Results for plaque burden in CAD-RADS 2.0 are clearly influenced by the choice of assessment method. When reporting plaque burden in CAD-RADS 2.0, the method of assessment should be unequivocally named, as the three proposed methods are not interchangeable and describe different aspects of plaque burden.

Abstract:The increasing complexity of interventional radiology procedures, combined with limited anesthesiology resources, makes it essential for interventional radiologists to independently provide safe and effective sedation and analgesia. To ensure responsible and timely patient care, a structured and comprehensive development of these competencies within interventional radiology is necessary. While established guidelines and training curricula exist in fields like gastroenterology and cardiology, comparable standards are still lacking in German-speaking countries for radiology. This position paper highlights the urgent need to develop a unified curriculum for sedation and pain management tailored to interventional radiology. This curriculum should be integrated into specialist and assistant training programs. Beyond clinical education, the availability of adequate infrastructure, standardized workflows, clear documentation, and emergency protocols are critical for ensuring patient safety. Collaboration with anesthesiology departments, particularly in complex or deep sedation cases, remains indispensable. Sedation performed by radiology teams is not intended to compete with anesthesiology, but rather to augment patient care capacity in a structured and safe manner. Expanding these competencies strengthens the autonomy of interventional radiology, supports the broader trend toward outpatient care, and enables timely, efficient, and comfortable procedures. This represents an important step in establishing interventional radiology as an independent clinical discipline actively involved in patient treatment. Key Points:· Growing need for autonomous analgosedation in IR - The increasing complexity of interventional radiology procedures combined with limited anesthesiology capacity makes competent sedation and analgesia performed by the IR team itself essential for timely and patient-centered care.. · International evidence supports safety and feasibility - Multiple guidelines (CIRSE, ASA, ESAIC, RCR) and recent studies - including CT-guided brachytherapy and percutaneous tumor ablation - confirm that non-anesthesiologist-administered sedation can be performed safely and effectively.. · Clear risk stratification and sedation boundaries - Minimal and moderate sedation in ASA I/II patients can be independently performed by trained interventional radiologists; deep sedation remains the domain of anesthesiologists or intensivists. Standardized patient selection and pre-procedural risk assessment are mandatory.. · Infrastructure, training, and documentation as minimum requirements - Standardized operating procedures, continuous monitoring, dedicated sedation personnel, regular simulation-based training, and structured post-procedural care - including a recovery unit and discharge management - are non-negotiable prerequisites.. · Urgent need for binding standards in the DACH region - While internationally established guidelines exist, comparable mandatory curricula and standards for IR are currently absent in German-speaking countries. This position paper provides the first practice-oriented recommendations and calls for the rapid development of competency frameworks within IR departments.. Citation Format:· Minko P, Auer TA, Gebauer B et al. Analgosedation in interventional radiology. Rofo 2026; DOI 10.1055/a-2863-2198.

RATIONALE AND OBJECTIVES:To investigate the impact of operator experience and the overall expertise of the medical center on technical success and procedural parameters of transjugular intrahepatic portosystemic shunt (TIPS) in a longitudinal setting. MATERIALS AND METHODS:TIPS procedures conducted at a German tertiary hospital from 2017 to 2023 were enrolled retrospectively. The impact of the center's expertise on technical success and procedural parameters was assessed by logistic and linear regressions stratified by year. Comparative analyses between more and less experienced operators were performed using Mann-Whitney U, Chi-square, and Fisher's exact tests. RESULTS:A total of 245 TIPS procedures (161 men, mean age 59.8±10.9years) were performed, with a technical success rate of 95.5% (234/245). Technical success remained stable over time (2017: 91.7% vs. 2023: 94.0%, p=0.532). Over time, significant reductions were observed in total procedure time (2017: 63.4±22.1 min vs. 2023: 44.4± 29.1 min, p<0.001), fluoroscopy time (2017: 18.8±10.6 min vs. 2023: 12.8±8.6 min, p=0.002), radiation dose (2017: 114.5±113.7 Gy*cm2 vs. 2023: 51.3±33.3 Gy*cm2; p<0.001), and contrast agent usage (2017: 72.0±34.6 mL vs. 2023: 48.1±24.7 mL, p<0.001). More experienced operators had significantly shorter fluoroscopy times than less experienced operators (14.0±9.9 min vs. 16.4±9.1 min, p=0.016). CONCLUSION:TIPS is a highly technically successful procedure. Increased experience at the performing center is associated with reduced procedure times, fluoroscopy times, and radiation doses, while technical success rates remain stable. Additionally, procedures performed by more experienced operators were associated with shorter fluoroscopy times.

Trotz verschiedener Möglichkeiten und technischer Verbesserungen der chirurgischen Dialysefistelanlage sind die Ergebnisse hinsichtlich Offenheit und notwendiger Reinterventionen suboptimal. Hauptfaktor für das Versagen von chirurgischen Fistelanlagen ist die neointimale Hyperplasie, welche juxta-anastomotisch durch Gefäßmanipulation während der Anlage sowie der verwendeten Anastomosenart ausgelöst wird. Neue endovaskuläre Verfahren erlauben die Anlage einer Dialysefistel mit minimalem Gefäßtrauma. Aktuell gibt es zwei Systeme zur Anlage einer endovaskulären Dialysefistel auf dem Markt. Beide Verfahren zeigten in den bisherigen Studien vielversprechende Ergebnisse hinsichtlich technischer Erfolgsraten, Komplikationen, Reinterventionen sowie kurzfristiger Offenheitsraten. Bei aktuell noch geringem Evidenzgrad bei den endovaskulären Dialysefisteln bleibt die Cimino-Brecia-Fistel weiterhin der Goldstandard.

BACKGROUND:Sarcopenia, characterized by loss of skeletal muscle mass and strength, is prevalent in patients undergoing treatment for colorectal cancer. Sarcopenia's prevalence in patients with cancer can reach up to 50% and is known to exacerbate postsurgical complications and affect long-term oncological outcomes. This study examined whether minimally invasive surgery (MIS) offers protective benefits against postoperative sarcopenia compared with open surgery in patients undergoing rectal cancer resection. METHODS:This retrospective analysis included 145 patients who underwent open or minimally invasive (laparoscopic or robot-assisted) rectal resections at the University Hospital Dresden between 2013 and 2021. Confounding variables were adjusted using propensity score matching. The skeletal muscle index (SMI) and psoas muscle thickness per height (PMTH) were analysed in preoperative and postoperative computed tomography scans to measure changes in skeletal muscle mass. Potential risk factors for muscle loss were evaluated, and oncological long-term outcome was analysed. RESULTS:The results indicate that oncological rectal resection did not result in pronounced postoperative muscle loss. No significant difference between the open and MIS groups in terms of postoperative muscle loss over 3 years postoperatively could be detected. Wound healing disorders were identified as the most significant independent risk factors for muscle loss (SMI loss > 10%). In contrast, neither the type of surgical technique nor the presence of a protective loop ileostomy significantly influenced the development of postoperative muscle loss. Patients who experienced a > 10% SMI loss within the first year had significantly poorer overall and disease-free survival. The 1-year survival rate was 93.3% in the group with high SMI loss compared with 100.0% in the group with low SMI loss (p = 0.435). The 3-year (66.7% vs. 95.6%, HR 8.75, 95% CI 1.855-41.286, p = 0.006) and 5-year (44.4% vs. 93.3%, HR 11.072, 95% CI 2.414-50.782, p = 0.002) survival rates were significantly lower in patients with high SMI loss. Patients with high SMI loss had an increased likelihood of recurrence and metastasis. CONCLUSIONS:Although MIS did not confer a protective advantage against postoperative muscle loss in patients with rectal cancer, the findings highlight the critical role of maintaining muscle mass in improving survival outcomes. Postoperative muscle loss appears to be a marker of aggressive tumour behaviour, and interventions aimed at minimizing muscle loss, such as enhanced nutritional support, may improve the long-term patient prognosis. Future studies should explore interventional strategies to mitigate sarcopenia in this population.

Repeated whole-spine imaging in children and adolescents with scoliosis causes significant cumulative radiation exposure. Low-dose slot-scanning imaging can reduce radiation exposure. The aim of this study was to compare whole-spine imaging using the slot-scanning technique (SST) with conventional stitched radiography (SR) with regard to image quality and radiation dose. We recruited 30 patients with a clinical indication for whole-spine imaging by SST, who had prior whole-spine imaging by SR available. 41 images acquired using both techniques were analyzed. The dose area product (DAP) and patient age, weight, height, and body-mass index were recorded. Image quality was assessed for 28 anatomical landmarks on a 4-point Likert scale by 4 raters. Group-wise comparisons were performed using paired t-tests. Interrater rating reliability was assessed using the interrater correlation coefficient. The mean DAP was significantly lower for SST than for SR (SST 8.0 ± 5.7 cGy · cm², SR 32.5 ± 34.5 cGy · cm², p < 0.001). The image quality was rated good for both modalities (means: SST 2.8 ± 0.6, SR 2.7 ± 0.5, p = 0.38). There was no significant difference in image quality ratings between both modalities. The inter-rater reliability was excellent with an inter-rater correlation coefficient of 0.86. Orthopedic surgeons rated the image quality of SST significantly better than that of SR. This study shows a significant reduction in radiation exposure for slot-scanning whole-spine imaging compared to stitched radiography with equal image quality, which facilitates the reduction of radiation exposure for repeated whole-spine imaging in children and adolescents to 25%.