Aral Varol

Background: Metabolic syndrome (MetS) is associated with increased cardiovascular risk, which rises with the number of MetS components. Conventional carotid ultrasound markers may not fully characterize dynamic arterial wall mechanics, whereas speckle-tracking carotid strain ultrasonography (STCS-US) enables integrated assessment of carotid stiffness and deformation throughout the cardiac cycle. This study aimed to assess multidimensional carotid mechanics in individuals with MetS and their association with cumulative metabolic burden. Methods: This retrospective cross-sectional study included 100 participants:50 patients with MetS diagnosed according to the National Cholesterol Education ProgramAdult Treatment Panel III (ATP III) criteria and 50 age and sex matched controls without MetS Carotid Doppler US images were analyzed using the Arterial Analysis software program, which quantified 22 vascular parameters per participant, comprising five stiffness parameters (β-stiffness index [β-SI], arterial compliance [AC], arterial distensibility [AD], elastic modulus [EM], and pulse wave velocity [PWV]) and six strain parameters (radial strain, radial displacement [DP], radial strain rate [SR], circumferential strain, circumferential displacement, and circumferential strain rate), each obtained in both the longitudinal and transverse planes. Correlations between the number of MetS components and imaging-derived vascular parameters were analyzed. Classification and Regression Trees (CART) analysis was performed to evaluate the relative contribution of clinical and structural variables to STCS-US-derived vascular parameters. Results: Patients with MetS exhibited significantly increased stiffness parameters, including EM and PWV, and reduced compliance indices compared with controls (all p < 0.05). Strain analysis demonstrated significantly lower radial and circumferential strain percentages in the MetS group, with more pronounced alterations observed in the transverse plane (all p < 0.001). Increasing MetS component count was associated with higher stiffness indices and lower strain values. CART analysis demonstrated that MetS status remained one of the most important determinants across the reported STCS-US-derived vascular parameters. Conclusion: STCS-US revealed coordinated impairment in carotid arterial stiffness and strain parameters in MetS, particularly in the transverse plane. These findings suggest that STCS-US may provide complementary functional information alongside conventional stiffness assessment and may reflect the cumulative metabolic burden on vascular mechanics.

Background: Arteriogenic insufficiency (AI) is the most common vascular cause of erectile dysfunction (ED) and is traditionally diagnosed using semi-invasive papaverine-induced penile Doppler ultrasonography (P-PDUS). Aim: To evaluate the feasibility and diagnostic accuracy of the cavernosal artery ondulation index (CA-OI) measured during the flaccid penile phase for identifying AI in men with ED. Methods: This prospective study included 90 men with clinically confirmed ED (International Index of Erectile Function-5 ≤ 25) and suspected vascular etiology. Gray-scale and color Doppler ultrasonography were performed in the flaccid state to measure CA-OI. AI was diagnosed according to established P-PDUS criteria, and patients were classified into AI (n = 28) and non-AI (n = 62) groups. CA-OI values were compared between groups, and receiver operating characteristic curve analysis was used to evaluate diagnostic performance. Outcomes: The primary outcome was the diagnostic accuracy of flaccid-phase CA-OI for detecting AI. Results: Patients with AI were significantly older than those without AI (59.8 ± 8.8 vs. 46.4 ± 13.5 years, P < .001) and had a higher prevalence of diabetes mellitus (P < .001). Mean CA-OI values were significantly higher in the AI group (2.76 ± 0.28 vs. 2.33 ± 0.17, P < .001). Receiver operating characteristic analysis demonstrated excellent diagnostic performance, with an area under the curve of 0.906. A CA-OI cut-off value of 2.46 yielded 89.3% sensitivity and 79% specificity for identifying AI. Clinical Implications: Measurement of CA-OI in the flaccid phase offers a non-invasive and clinically practical approach for the initial screening of arteriogenic ED, potentially reducing the need for intracavernosal injection. Strengths and Limitations: The principal strength of this study is the first evaluation of CA-OI as a non-invasive diagnostic marker obtained during the flaccid penile phase; limitations include the single-center design and the use of P-PDUS rather than angiography as the reference standard. Conclusion: CA-OI measured during the flaccid penile phase is a highly accurate, non-invasive marker for AI in men with ED.

Introduction: This study aimed to evaluate the diagnostic performance of GPT-4V, a vision-enabled large language model, in classifying liver lesions on MRI according to LI-RADS v2018 criteria. Methods: Seventy contrast-enhanced liver MRI examinations were retrospectively selected, comprising 10 cases from each LI-RADS category. Each case was presented to GPT-4V as a standardised set of seven anonymised axial MRI slices, accompanied only by lesion size. The model was prompted to assign a single LI-RADS category based solely on visual input. The model’s performance was assessed using accuracy metrics, Cohen’s kappa, ROC analysis, and correlation with lesion size. Results: GPT-4V achieved an overall classification accuracy of 37.1%. While the accuracy for LR-5 was high (90%), it was notably poor in LR-3 (0%) and LR-4 (20%). More than half of the lesions were misclassified as LR-5 (54.2%). Binary classification into benign (LR-1 and LR-2) versus malignant (LR-4, LR-5, LR-M, LR-TIV) yielded an accuracy of 84.3%, with an AUC of 0.72, sensitivity of 100%, and specificity of 45%. Cohen’s kappa values were 0.27 for detailed classification and 0.54 for benign–malignant grouping. Lesion size positively correlated with classification accuracy (ρ = 0.26, p = 0.031). The model demonstrated a tendency to favour high-certainty categories, often defaulting to LR-5 when diagnostic ambiguity was present. Conclusions: GPT-4V demonstrated limited performance in detailed LI-RADS classification, with a strong bias toward LR-5 and poor accuracy in intermediate categories. While the model demonstrated relatively better performance in distinguishing benign from malignant lesions, its current form remains inadequate for precise image-based categorisation. Further development with structured visual training and clinically contextual prompting is warranted.