Effective lymphoedema management depends on early detection and treatment, emphasizing the need for measurement instruments with strong clinimetric properties. This systematic review aimed to identify and evaluate the measurement properties of imaging modalities used to assess soft tissue oedema. A systematic search of Embase, Medline, Web of Science, and CINAHL (inception-December 2, 2025) identified studies evaluating reliability, validity, and/or responsiveness of imaging modalities capable of visualizing internal tissues that evaluated soft tissue oedema in adults. Methodological quality was assessed using the COSMIN Risk of Bias checklist, and measurement properties were rated according to established criteria. Levels of evidence were synthesized using a best-evidence approach. Fifty-nine studies were included, covering six imaging modalities: ultrasound, fluorescence lymphography, magnetic resonance imaging (MRI), lymphoscintigraphy, computed tomography, and dual-energy X-ray absorptiometry. High-quality evidence for sufficient validity was identified exclusively for MRI, particularly in quantifying fluid and soft tissue. High-quality evidence for sufficient reliability was found for MRI in quantifying soft tissue and for fluorescence lymphography when assessing the combination of lymphatic flow and anatomy. Other modalities demonstrated only low- to moderate-level or inconsistent evidence across constructs. MRI currently provides the most robust and consistent evidence for valid and reliable quantification of soft tissue oedema. Standardization of methodology, broader inclusion of patient populations and body regions, and evaluation of underexplored measurement properties are essential to advance the role of imaging in lymphoedema assessment and management. CRD42024611323.
Osteoporotic vertebral fractures (OVFs) significantly impact morbidity, mortality, function and quality of life. Stable fractures with controllable pain can be managed conservatively. Those that remain painful can be treated with minimally invasive measures of vertebral augmentation (VA). In contrast, severe and unstable OVFs risk collapse, progressive kyphosis and neurological damage. Conventional VA is considered insufficient for these unstable fractures, making surgical stabilisation necessary to restore spinal load-bearing capacity, and allow fast and painless mobilisation. However, spinal surgery in elderly osteoporotic patients is challenging due to tissue fragility and common comorbidities. In addition, poor bone quality increases the risk of surgical implant failure, new fractures, complications and re-intervention.This trial aims to assess the non-inferiority of a novel minimally invasive interventional technique, called stent-screw assisted internal fixation (SAIF), compared with multilevel surgical stabilisation in terms of quality of life and radiological outcomes such as vertebral height restoration and kyphotic correction in patients with unstable OVFs. This is a multicentre, parallel-group, randomised controlled non-inferiority trial that will be conducted in four European centres.A total of 140 patients with type Osteoporotic Fracture (OF) 3-5 OVFs (according to the AO Spine-DGOU Osteoporotic Fracture Classification System) will be randomised 1:1 to receive either SAIF intervention or surgical stabilisation.The primary outcome is to determine whether the effect of SAIF intervention is not inferior to surgical stabilisation on pain and quality of life at 1-year follow-up.The principal secondary outcomes are to evaluate whether the effect of the SAIF intervention is not inferior to surgical stabilisation in terms of radiological outcome and cost-effectiveness. Ethics approval was obtained from the Ethics Committee of the Canton Ticino, Switzerland (CE 4004). Results will be disseminated through international publications in peer-reviewed journals, in addition to international conference presentations. NCT07071870.
Active surveillance (AS) is a well-established strategy for managing prostate cancer (PCa); however, available evidence indicates that approximately 28% of patients may experience progression within the first 5 years. Magnetic resonance imaging (MRI) may help optimize patient selection. This study assessed whether the presence of focal lesions (prostate imaging reporting and data system (PI-RADS) ≥ 3) on baseline MRI is associated with an increased risk of histological progression and AS discontinuation. We conducted a retrospective study of consecutive patients enrolled in AS at a comprehensive cancer center since 2012, recorded in the Urodata platform, based on a prospectively maintained cohort and including patients meeting European Association of Urology (EAU) criteria for low- or favorable intermediate-risk PCa. Inclusion criteria were diagnostic and confirmatory transperineal biopsies performed with the institution's standard technique, available baseline MRI, and a minimum follow-up of one year. Patients with prior treatment were excluded. Histological progression was defined as an increase in Gleason grade group on follow-up biopsy compared with baseline. Progression at 5 years was evaluated using Kaplan-Meier curves and multivariable Cox proportional hazards models. Patients with PI-RADS ≥ 3 were classified as visible lesions on MRI (MRI+), and those with PI-RADS <3 as no visible lesions on MRI (MRI-). A total of 105 patients were included, with a median age of 64 years (interquartile range (IQR) 59-70) and a median prostate-specific antigen (PSA) of 5.8 ng/mL (IQR 4.7-8.9). The median interval between initial and confirmatory biopsy was 24 months, and median follow-up was 6 years. Seventy-one patients were classified as MRI+. Baseline characteristics were similar between groups. Five-year progression-free survival was 41% in MRI+ versus 77% in MRI- groups (hazard ratios (HR) 5.5; 95% confidence interval (CI) 2.3-13). Only 28% of patients with PI-RADS 5 remained free of progression at 5 years. Baseline PSA >6 ng/mL and PSA density >0.2 were independently associated with histological progression. The presence of PI-RADS ≥ 3 lesions on baseline MRI is significantly associated with an increased risk of histological progression and AS discontinuation. MRI may be a key tool for optimizing candidate selection for AS.
Visualisation of both soft-tissue and osseous structures is required for optimal cervical spine assessment in patients with cervical radiculopathy. This study investigated whether radiologists considered an MRI-based synthetic CT to be of added value to MRI and X-ray in the assessment of the cervical spine in these patients. In this exploratory single-centre, retrospective study that was approved by the local institutional review board (NWMO 190716), three radiologists randomly and independently evaluated two sets of images of 24 patients aged 50 years or older that had received an MRI, X-ray and an MRI-based synthetic CT for cervical radiculopathy, using a questionnaire. Prior to this assessment they received instructions on the questionnaire with a separate test set. Image set 1 consisted of MRI and X-ray, and set 2 of MRI and X-ray, complemented by the synthetic CT. The radiologists reported significantly improved quality of assessment in general and in assessment of cortical delineation, intervertebral joints and neural foramina at C3-4 and C6-7 and presence of posterior lipping, facet arthrosis, bony apposition in the spinal canal when the synthetic CT was presented with MRI and X-ray (median score of 3.8 [IQR 0.2] vs 3.2 [IQR 0.3], p < 0.001). The visibility of trabecular bone was not significantly different between the two sets. Subjective evaluation by radiologists demonstrated that adding the synthetic cervical spine CT to the MRI and X-ray in patients with cervical radiculopathy could improve the quality of image evaluation and diagnostic confidence, warranting confirmation in larger studies.
Accurate subtype classification of inner ear malformations (IEMs) on temporal bone high-resolution computed tomography (HRCT) is important for preoperative risk assessment and surgical planning, but depends heavily on subspecialty expertise. This study aimed to develop and externally validate a deep learning model for multiclass IEM diagnosis on temporal bone HRCT. We developed and externally validated a weakly supervised deep learning framework for one-stage classification of automatically extracted temporal bone regions of interest from HRCT in a multicenter cohort of 9,182 ears (3,161 IEMs and 6,021 normal ears). The framework was based on an improved Transformer-based multiple instance learning architecture. Performance was evaluated on internal and independent external datasets. An independent reader study compared the model with five senior and five junior otolaryngologists. Explainability was assessed using attention-based slice ranking and Grad-CAM. The study was approved by an institutional review board, and informed consent was waived. On the independent external test set, the proposed model demonstrated strong multiclass performance, with an accuracy of 93.9 % (95 % CI, 92.4 %-95.3 %), a macro-F1 score of 88.2 % (95 % CI, 84.3 %-91.4 %), and a multiclass AUC of 0.990 (95 % CI, 0.985-0.995). Performance was consistent on the internal validation set, with an accuracy of 96.5 % (95 % CI, 95.6 %-97.4 %), a macro-F1 score of 87.5 % (95 % CI, 83.0 %-91.0 %), and a multiclass AUC of 0.994 (95 % CI, 0.989-0.998). In the reader study, the model obtained an accuracy of 94.0 % and a macro-F1 score of 89.9 %, outperforming both senior otolaryngologists (mean macro-F1, 75.7 %, Holm-adjusted p = 0.0046) and junior otolaryngologists (mean macro-F1, 62.9 %, Holm-adjusted p < 0.001). The model also provided interpretable outputs by identifying the top three highest-contributing slices and generating Grad-CAM heatmaps. This weakly supervised framework enabled accurate and generalizable subtype-level diagnosis of IEMs on temporal bone HRCT in a large multicenter cohort and may serve as a practical decision-support tool for standardized preoperative assessment, particularly in settings with limited subspecialty expertise.
Body composition (skeletal muscle, adipose tissue, and bone) is a general measure of health that can be determined from abdominal computed tomography (CT) using deep learning (DL) algorithms. Use of a volumetric approach, rather than the more commonly used single axial slice, likely improves the reliability and validity of body composition biomarkers, but a standardized approach is needed. A study was designed to test the reliability of volumetric body composition assessment in serial CTs. Residents of Southern Minnesota and Western Wisconsin who had two consecutive abdominal CT exams 1 to 14 days apart from 2010 to 2021 were identified. Previously-validated DL models were applied to both CT exams, and three methods of determining body composition biomarkers were compared: two-dimensional (2D) model on a slice at the midpoint of L3, three-dimensional (3D) model on a slice at the midpoint of L3, and 3D model on slices extending from L2 to L4. For each method, reliability was assessed by the intraclass correlation coefficient (ICC) separately in paired noncontrast and paired contrast-enhanced CT exams and validity was assessed by the strength of association of each body composition measure with age. There were 3,376 persons with paired noncontrast and 9,103 persons with paired contrast-enhanced CT exams. The 3D body composition biomarkers from L2-L4 showed generally better reliability than the 2D or 3D biomarkers at the midpoint of L3 with higher ICCs observed with or without use of contrast. The 3D body composition more strongly correlated with age than 2D body composition for most biomarkers. The 3D bone area from L2-L4 more strongly correlated with age than 3D bone area at the midpoint of L3. Volumetric L2-L4 body composition analysis shows better reliability and validity than single slice L3 analysis, and this may result in more accurate clinical associations with body composition.
To map the available evidence on the educational use of social media in radiology education, across three domains: educational use and adoption; participant perceptions, preferences, and satisfaction; and knowledge or performance outcomes, and to identify methodological priorities for future evaluation. A scoping review was conducted according to PRISMA-ScR principles. PubMed/MEDLINE, Scopus, Web of Science, Google Scholar, and EPOS were searched for English-language records published between 2015 and 2025. Eligible records addressed learner-oriented radiology education contexts and reported original quantitative participant-level data on educational use, perceptions, satisfaction, knowledge, retention, or performance. Two reviewers independently screened records and assessed full texts. Findings were synthesized narratively because of marked methodological heterogeneity. The search identified 839 records. After 183 duplicates were removed, 656 records were screened; 38 reports were assessed in full text and 10 were included. The evidence comprised four survey or questionnaire studies, two educational initiatives with participant evaluation, and four small comparative or pre/post studies. Most findings concerned educational use, adoption, preferences, and perceived usefulness. Knowledge or performance outcomes were assessed in four heterogeneous studies, with findings ranging from no significant knowledge benefit or comparable retention to favorable results in small, context-specific settings. No platform-specific educational advantage was established. The available literature is better suited to characterize how social media are being used in radiology education than to establish educational effectiveness. Participant-level evidence remains limited, predominantly self-reported, and context-specific, while knowledge and performance outcome evaluation is small-scale and heterogeneous.
To evaluate the diagnostic performance of the Liver Imaging Reporting and Data System (LI-RADS) nonradiation Treatment Response Assessment (TRA) version 2024 (v2024) in assessing treatment response of hepatocellular carcinoma (HCC) following downstaging locoregional therapy (LRT) plus immune-targeted therapy; to compare its diagnostic performance with LI-RADS TRA version 2018 (v2018); and to explore the added diagnostic value of its ancillary features (AFs). This retrospective study included patients with HCC who underwent LRT plus immune-targeted therapy followed by hepatectomy between January 2021 and December 2025. Two radiologists independently assigned LR-TR classifications to lesions on MRI according to the LI-RADS Nonradiation TRA v2024 and v2018. With pathological findings from hepatectomy as the reference standard, the sensitivity, specificity and accuracy of the two versions were calculated. Diagnostic performance for identifying residual tumor viability was compared using the McNemar test. A total of 132 patients (19 women, 113 men; mean age, 55.9 ± 9.5 years [standard deviation]) with 143 treated HCC lesions were enrolled in this study. The LR-TR Viable category of v2024 yielded a sensitivity of 66.3% (95% confidence interval [CI]: 57.1%-75.5%) and a specificity of 95.2% (95% CI: 84.2%-98.7%) for predicting pathologic viability. For identifying complete pathologic necrosis, the LR-TR Nonviable category showed a sensitivity of 76.2% (95% CI: 63.3%-89.1%) and a specificity of 95.1% (95% CI: 90.8%-99.3%). The incorporation of AFs significantly improved the sensitivity for detecting pathologic viability (87.1% vs. 66.3%, P < 0.001), with no significant difference in specificity (90.5% vs. 95.2%, P = 0.500). Overall, LI-RADS nonradiation TRA v2024 integrated with AFs exhibited superior diagnostic performance relative to LI-RADS TRA v2018. The LI-RADS nonradiation TRA v2024 incorporating AFs showed excellent diagnostic performance and outperformed LI-RADS TRA v2018 for assessing tumor viability in HCC patients treated with LRT plus immune-targeted therapy.
The purpose of this study is to propose a new classification of osteonecrosis of the femoral head (ONFH) based on the relationship between the epiphyseal line and osteonecrotic boundary, to offer a reference for prognostic assessment of ONFH and provide strategic guidance for hip surgical interventions. This retrospective, single-center study commenced by gathering data on patients diagnosed with ONFH between June 2018 and May 2023. Patients were classified into five types based on the relationship between the epiphyseal line and the osteonecrotic boundary on magnetic resonance imaging coronal midsections. Various imaging findings, including bone marrow edema, cystic lesions, microfractures, collapse, and crescent signs, were assessed. Statistical analyses were performed to determine the association between this relationship and disease progression. We researched 289 ONFH patients (42.53 years ± 13.66, 193 men) (405 hips) who met our selection criteria. Type A comprised 35.31 %, Type B 56.54 %, and Type C 8.15 %. Significant differences were observed among Types A, B, and C concerning the presence of cystic lesions, microfractures, collapse, and crescent signs (p < 0.001). Univariate logistic regression analysis demonstrates a progressive increase in structural impairment risks across classifications. Additionally, significant differences were also found in the subtypes of Type B (B1, B2, B3). The relationship between the epiphyseal line and the osteonecrotic boundary is closely associated with the progression of ONFH, particularly in the development of femoral head collapse. This study introduced a novel imaging classification system for ONFH, providing valuable insights for clinical management.
To evaluate the clinical feasibility of using the non-zero-calcification-risk Multi-Ethnic Study of Atherosclerosis (nzcr-MESA) score, a simplified risk calculator that includes only age, sex, and ethnicity, for predicting the extent of coronary artery calcification (CAC) as measured by computed tomography (CT). In this retrospective study, 241 Caucasian patients underwent coronary CT angiography (CCTA) for the assessment of coronary artery disease. The objective extent of coronary calcification was quantified using Agatston scores. The nzcr-MESA score was calculated based on patient demographics and compared to the extent of calcification. Statistical analysis involved Spearman correlation coefficients and regression models to determine the relationship between nzcr-MESA and CAC. Significance was set at P < 0.05. A positive correlation was found between nzcr-MESA scores and the degree of CAC (P < 0.0001, females R = 0.56, males R = 0.39) as well as to disease severity according to the coronary artery disease-reporting and data system (CAD-RADS) scores (P < 0.0001, females R = 0.45, males R = 0.42), suggesting that the simplified score is a predictor of calcification and coronary disease. The optimal nzcr-MESA cutoff values for preselecting patients with severe coronary artery calcification (Agatston Score ≥ 400) were 90 % for females and 95 % for males. The nzcr-MESA score offers a clinically feasible, simplified method for estimating CAC extent, with significant implications for CT coronary imaging. Given its ease of use and its capability to correctly predict severe CAC, nzcr-MESA can be useful to preselect suitable diagnostic protocols. Integrating nzcr-MESA into clinical routine may help to streamline diagnostic workflows.
PIK3CA-related overgrowth spectrum (PROS) comprises a heterogeneous group of disorders caused by postzygotic activating variants in PIK3CA, leading to mosaic activation of the PI3K pathway. The natural history of PROS is highly variable and depends on the timing and distribution of the somatic mutation. To better characterize its radiological progression, we analyzed a cohort of patients with PROS. This multicenter study was conducted at Hôpital Femme Mère Enfant (Lyon) and Hôpital Necker-Enfants Malades (Paris). Pediatric and adult patients with a confirmed PIK3CA variant and at least two MRI examinations performed ≥ 2 months apart were included. Patients who underwent interventional surgical or radiological procedures, or who received targeted therapies, were excluded. Target lesions were identified on baseline MRI and evaluated on follow-up scans according to RECIST criteria. Among 67 PROS patients screened between 2008 and 2021, 30 met the inclusion criteria, including 43.3 % children (13/30) and 43.3 % females (13/30). The median age at first MRI was 19 years (interquartile range 5-34). Lesions involved the head and neck region (10/30), trunk and upper limbs (10/30), and lower limbs (10/30), with some patients presenting multisite involvement. Over a median follow-up of 75.7 months (range 2.1-160.3), 83.3 % (n = 25) of patients had progressive disease, with a median increase of 36.6 % in lesion size (range 2.6-230.7 %) and a median annualized growth rate of 7.2 %/year (range 0.6-40.1 %). This study provides the first radiological description of PROS natural history, showing that tissue malformations generally enlarge over time and often continue progressing into adulthood.
To characterize different hip morphology phenotypes in individuals with femoroacetabular impingement (FAI) syndrome with cam morphology. Supine pelvic anterior-posterior radiographs from 119 hips from 119 individuals diagnosed with FAI syndrome with cam morphology from a RCT were analysed. The lateral center edge angle (LCEA), Wiberg center edge angle (W-CEA), global retroversion, protrusio acetabuli, and femoral neck shaft angle (FNSA) measures were assessed independently by two raters. Criteria from the Lisbon Agreement on FAI imaging were used to characterize different hip morphology phenotypes based on overcoverage, undercoverage, retroversion and femoral orientation. Radiographic measure's reliability was calculated using intraclass correlation coefficients and kappa statistic. Reliability for all measurements ranged from good to almost perfect agreement. Global retroversion of the acetabulum was percent in 39% of the participants. Based on the W-CEA, 9% of participants had acetabular undercoverage and 9% had overcoverage, while the LCEA identified 5% with undercoverage and 17% with overcoverage. Coxa valga was present in 27% of participants, with only one participant having coxa vara. No participants had protrusio acetabuli. Women had a slightly higher prevalence of coxa valga than men. Close to 40% of patients with FAI syndrome with cam morphology also displayed pincer morphology by acetabular retroversion. The prevalence of pincer morphology, based on the W-CEA and LCEA measures was low, indicating that isolated use of these parameters may lead to underdiagnosis of pincer morphology. The prevalence of hip dysplasia determined by the W-CEA and LCEA measures was low, ranging from 2 to 5%.
This prospective, single-center clinical trial evaluated the technical feasibility and clinical success of the Automatic Needle Targeting (ANT-C) patient-mounted robotic needle-guidance platform for CT-guided trans-thoracic lung biopsy under local anesthesia, incorporating an exploratory post-hoc geometric analysis of parallel artificial intelligence path planning via the NDAnalyzer software. Out of thirty-one recruited patients, one was excluded due to on-table lesion resolution, leaving a final cohort of thirty evaluable subjects with a median nodule size of 24 mm. While interventional radiologists executed target acquisition using independently planned trajectories, the automated software generated alternative paths in a parallel, blinded fashion to ensure zero clinical influence on the live procedure. The platform demonstrated a high technical success rate of 93.3% and a 100% clinical success rate, successfully harvesting adequate tissue cores for full histopathological and molecular profiling in all cases. Operational efficiency improved significantly as operators gained familiarity, with median procedural durations decreasing from 61.4 min in the first ten cases to 30.5 min in the final ten cases. Concurrently, median total radiation exposure significantly decreased from 427.3 mGy·cm to 303.7 mGy·cm between the early and late deciles. No major complications occurred, and minor, self-limiting Grade A adverse events were limited to asymptomatic perilesional hemorrhage and trace pneumothoraces. Furthermore, exploratory modeling revealed a median angular deviation of only 3.5° between the operator's selected path and the closest automated alternative. The robotic system is clinically viable and technically feasible for awake patients, though the AI's automated path-ranking cost-function requires further refinement prior to live workflow integration.
Video-assisted thoracoscopic surgery (VATS) is the standard treatment for small pulmonary nodules (SPNs), yet accurate localization of non-palpable lesions remains challenging. This study evaluated the efficacy, feasibility, and safety of a novel computed tomography (CT)-guided microcoil localization system for preoperative SPN management. We developed a novel CT-guided microcoil system featuring a coaxial needle and a retractable coil with a long, visible distal tail to improve stability and intraoperative identification. In this prospective single-arm study, 178 patients with 231 SPNs (≤10 mm, cN0) underwent preoperative localization followed by VATS. The primary endpoint was technical success, defined as deployment within 10 mm of the lesion without fracture, displacement, or dislodgement. Technical success was achieved in 224 of 231 nodules, yielding a rate of 97.0% (95% CI: 93.7-98.6%). The mean localization time was 4.4 min, and the median coil-to-nodule distance was 4.9 mm. All nodules were completely resected with negative margins on the first attempt and were successfully identified for definitive pathological diagnosis. Complications were infrequent and mild: asymptomatic pneumothorax occurred in 12 cases (5.2%) and minor hemorrhage in 11 (4.8%), none of which required treatment; these rates compare favorably to those of conventional hook-wire systems. Patient pain was negligible, and 97.4% of coils were retrieved with the specimen. Failures occurred in 3.0% of nodules due to wire fracture, dislodgement, or displacement. This novel microcoil system demonstrated high efficacy, procedural efficiency, and excellent safety, addressing key limitations of existing localization methods.
To evaluate whether an extended multiphasic computed tomography (CT) protocol including an additional intermediate phase improves detection of the primary tumor in patients with neuroendocrine neoplasms of unknown primary site (CUP-NEN). This retrospective single-center study included 194 patients with histologically confirmed NEN of unknown origin who underwent contrast-enhanced CT between 2010 and 2023. All examinations were performed using an extended protocol comprising unenhanced, arterial, intermediate, and portal venous phases. The primary endpoint was detection of the primary tumor, confirmed by histopathology. Secondary analyses included metastatic patterns and association with tumor grade. The primary tumor was identified in 79 of 194 patients (40.7%). The gastrointestinal tract was the most frequent location (18.0%), particularly the small intestine (13.4%), followed by the pancreas (15.0%). The mean tumor size was 14 mm (range 6-38 mm). Most patients presented with distant metastases, predominantly in the liver (76.3%), and lymph node metastases were observed in 50.5% of cases. No significant association between tumor grade and detection rate was found (p = 0.51). In a phase-based analysis, the intermediate phase provided exclusive or optimal visualization of primary tumors in 84.8% of detected cases. An extended multiphasic CT protocol including an intermediate phase may improve detection of small hypervascular primary tumors in patients with CUP-NEN. Optimization of CT acquisition protocols represents a practical and widely accessible strategy to enhance diagnostic performance in this patient population.
To develop and temporally validate LN-104-RADS, an ultrasound-based risk stratification system for station 104 supraclavicular lymph nodes (SCLNs) in esophageal squamous cell carcinoma (ESCC). This retrospective diagnostic study included consecutive patients with histopathologically confirmed ESCC who underwent pre-treatment supraclavicular ultrasound. Patients were assigned to a development cohort (January 2023 to June 2025) or a temporal validation cohort (July 2025 to January 2026). LN-104-RADS was developed by adapting the Node-RADS size-configuration framework to ultrasound assessment of station 104 lymph nodes. Metastatic nodes were defined by histopathological or cytological confirmation, while non-metastatic nodes were defined by negative pathology/cytology without subsequent progression or by radiologic stability, regression, or disappearance during at least 6 months of follow-up. The prespecified cutoff of category ≥ 4 was applied unchanged in the validation cohort. Diagnostic performance, calibration, clinical utility, and interobserver agreement were assessed using ROC analysis, calibration curves, decision curve analysis, and Cohen's kappa. The development cohort included 1806 patients, contributing 1832 station 104 lymph node observations, including 263 metastatic and 1569 non-metastatic nodes. Using LN-104-RADS category ≥ 4, the AUC, sensitivity, specificity, and accuracy were 0.968, 95.8%, 97.8%, and 97.5%, respectively. The metastatic rate increased stepwise from 0% in categories 1 and 2 to 4.8% in category 3, 81.6% in category 4, and 100% in category 5. In the temporal validation cohort of 500 patients, LN-104-RADS retained favorable performance, with an AUC of 0.920, sensitivity of 90.9%, specificity of 93.0%, and accuracy of 92.6%. Calibration analysis showed acceptable agreement in the development cohort, although some deviation was observed in the temporal validation cohort. Decision curve analysis demonstrated greater net clinical benefit for LN-104-RADS than the 'treat-all' or 'treat-none' strategies across clinically relevant threshold probabilities. Interobserver agreement for LN-104-RADS categorization was excellent (κ = 0.841). LN-104-RADS showed high diagnostic performance, reproducibility, and temporal robustness for ultrasound-based station 104 nodal assessment in ESCC. This framework may support standardized pre-treatment risk stratification and help identify patients who may warrant further diagnostic evaluation for supraclavicular nodal staging.
To compare hepatic and splenic transverse relaxivity (R2*) derived from two-dimensional (2D) and three-dimensional (3D) multi-echo gradient-echo (ME-GRE) acquisitions before and after superparamagnetic iron oxide (SPIO) administration and to evaluate the effects of SPIO on proton density fat fraction (PDFF), MR spectroscopy (MRS)-derived PDFF, native T1 mapping, and magnetic resonance elastography (MRE)-derived liver stiffness. In this retrospective study, 48 adults underwent liver MRI at 3 T before and after SPIO administration. Hepatic and splenic R2* and PDFF were obtained from 2D and 3D ME-GRE acquisitions processed using MRQuantif and LiverLab. MRS-derived PDFF, native variable flip angle T1 mapping, and MRE-derived liver stiffness were assessed before and after SPIO administration. Agreement between methods was evaluated using Pearson correlation coefficients, intraclass correlation coefficients (ICC), and Bland-Altman analysis. Predictors of SPIO-induced changes were assessed using multivariable linear regression. SPIO administration resulted in a marked increase in hepatic R2* (approximately 3.5-fold), exceeding the increase observed in the spleen (approximately 2-fold), consistent with predominant hepatic uptake. Correlation between LiverLab and MRQuantif R2* measurements remained strong; however, inter-method agreement deteriorated at higher R2* values after SPIO administration. ME-GRE-based PDFF measurements remained stable with preserved inter-method agreement. In contrast, MRS-derived PDFF demonstrated reduced reliability after SPIO administration, with apparent increases in fat fraction and weakened correlation with ME-GRE PDFF measurements. Native hepatic T1 values decreased uniformly by approximately 500 ms following SPIO administration. MRE-derived liver stiffness showed no significant change after SPIO administration. Using SPIO as a controlled in vivo susceptibility model, this study provides an integrated evaluation of the susceptibility dependence of commonly used quantitative liver MRI biomarkers. SPIO-induced susceptibility effects substantially increase hepatic R2*, impair the reliability of MRS-based fat quantification, and markedly shorten native variable flip angle T1 measurements, while preserving the relative stability of ME-GRE-based PDFF and MRE-derived liver stiffness. These findings improve understanding of the susceptibility-related performance of quantitative liver MRI techniques and may help inform the interpretation and standardization of iron-sensitive MRI biomarkers.
Plasma phosphorylated tau (p-tau) biomarkers have improved the diagnosis of Alzheimer's disease (AD), but evidence in early-onset populations remains limited. We evaluated the diagnostic performance of plasma p-tau217 and p-tau181 levels in patients with early-onset AD (EOAD) and early-onset frontotemporal dementia (EOFTD). We analyzed 185 patients (EOAD = 150, EOFTD = 35) aged ≤ 65 years from the LEAF study (2021-2023). Plasma p-tau217, p-tau181, neurofilament light (NfL), and glial fibrillary acidic protein (GFAP) levels were measured by immunoassays. Both plasma p-tau217 (AUC = 0.831) and p-tau181 (AUC = 0.862) levels demonstrated high discriminative performance, with no significant difference between the two p-tau isoforms. P-tau levels were higher in patients with EOAD, whereas NfL levels were higher in EOFTD and were elevated in those with EOAD participants with severe hippocampal atrophy. Adding NfL, GFAP, and APOE ε4 status further improved the discriminative accuracy for differentiating EOAD from EOFTD. Plasma p-tau217 and p-tau181 are effective biomarkers for distinguishing biologically defined EOAD from EOFTD. Incorporating NfL, GFAP, and APOE ε4 status further enhances diagnostic accuracy.
Pancreatic ductal adenocarcinoma (PDA) is a leading cause of cancer-related deaths, with early diagnosis hampered by nonspecific symptoms and limitations of existing imaging techniques. This study aimed to develop a deep learning (DL) algorithm to automatically classify pancreas lesions on contrast-enhanced CT scans as normal, benign, or malignant, to assist radiologists in detecting early-stage pancreatic cancer. A dataset of 1,037 portal-phase CT scans was compiled from 18 institutions. The dataset was divided into a training set (N = 732) and a test set (N = 305), which was further divided into an internal validation test set (N = 139) and an external validation test set (N = 166). After segmentation using the TotalSegmentator algorithm, the pancreas was isolated from each CT scan. A DL model combining TotalSegmentator's pre-trained encoder and nnUNet decoder layers was developed to classify pancreas lesions. Ten-fold cross-validation was applied, and model performance was assessed using precision, recall, area under the curve (AUC) and a final score (FS) representing a weighted average of the previous three metrics. The final prediction combined the outputs of ten models. Across all validation datasets (139 and 166 patients, respectively, in the internal and external dataset), precision and recall were 0.57 and 0.63, respectively, while AUC was 0.84. In the external validation dataset, malignant lesions were detected with an AUC of 0.97. The model achieved an FS of 0.72 in both internal and external validation datasets, indicating consistent performance across datasets. This study demonstrated the feasibility of using a DL algorithm for automated pancreas lesion classification in CT scans. The model showed strong performance, particularly in detecting malignant lesions. Further research is needed to assess the model's clinical applicability and performance in real-world settings.
The association of physical inactivity with subclinical alteration of cardiac structure and function is not well established. In the present study, we investigated this association with left-ventricular (LV) function assessed by advanced MRI strain analysis in a population-based cohort free of known cardiovascular disease (CVD). We used data of an MRI sub-study of the community-based KORA (Cooperative Health Research in the Region of Augsburg) cohort with 360 participants (156 women; aged 39 to 73 years). Physical activity was assessed at three examination cycles (Exam 1 [baseline], at Exam 2 [7-years follow-up] and at Exam 3 [14-years follow-up]) of the KORA S4 cohort. The MRI-derived myocardial 2D feature-tracking strain examination was conducted at Exam 3 and included longitudinal, radial and circumferential global strain parameters. Associations between physical inactivity and strain values were investigated in cross-sectional and longitudinal analyses. Current physical inactivity (Exam 3) was associated with reduced longitudinal (β = -1.09%; 95%CI -2.02; -0.15. p = 0.023), radial (β = -2.64%; 95%CI -4.79; -0.49. p = 0.016), and circumferential strain (β = -0.90%; 95%CI -1.60; -0.20. p = 0.012). Similarly, past physical inactivity reported at baseline (Exam 1) was associated with reduced strain parameters. The more often physical inactivity was reported across the three examinations (two or three times) the more reduced were radial (β = -2.09%; 95%CI -3.71; -0.47. p = 0.012) and circumferential strain (β = -0.69%; 95%CI -1.22; -0.16. p = 0.011). These associations were stronger in men than in women. In a population-based cohort without overt CVD, repeated self-reported leisure-time physical inactivity was consistently associated with lower MRI-derived LV strain parameters at Exam 3. These results indicate that sustained inactivity may be linked to subtle impairments in myocardial deformation detectable before clinical disease onset.