Contrast-induced nephropathy (CIN) represents a significant complication associated with the use of iodinated contrast media (ICM), especially in individuals with preexisting renal impairment. The pathophysiology of CIN encompasses oxidative stress, inflammation, endothelial dysfunction, and hemodynamic disturbances, resulting in acute kidney injury (AKI). Early detection is essential for effective management; however, conventional markers like serum creatinine (sCr) and estimated glomerular filtration rate (eGFR) exhibit limitations in sensitivity and timeliness. This review emphasizes the increasing significance of novel biomarkers in enhancing early detection and risk stratification of contrast-induced nephropathy (CIN). Recent advancements in artificial intelligence and computational analytics have improved the predictive capabilities of these biomarkers, enabling personalized risk assessment and precision medicine strategies. Additionally, we discuss mitigation strategies, including hydration protocols, pharmacological interventions, and procedural modifications, aimed at reducing CIN incidence. Incorporating biomarker-driven assessments into clinical decision-making can enhance patient management and outcomes. Future research must prioritize the standardization of biomarker assays, the validation of predictive models across diverse patient populations, and the exploration of novel therapeutic targets. Utilizing advancements in biomarkers and risk mitigation strategies allows clinicians to improve the safety of contrast-enhanced imaging and reduce the likelihood of renal injury.

The association between vesicoureteral reflux (VUR) and chronic kidney disease (CKD) risk remains unestablished. We investigated the incidence of CKD in children with VUR in Taiwan and evaluated whether they had a higher risk of CKD than the general population. A nationwide population-based cohort study was conducted among children with VUR identified using Taiwan’s National Health Insurance Research Database from 2000 to 2013. VUR was defined according to the International Classification of Diseases, Ninth Revision, Clinical Modification codes. We identified the children with VUR and randomly selected comparison children according to a 1:1 ratio, matching them by age, gender, index year and comorbidity using data from the National Health Insurance Research Database. In total, 8648 children with VUR and 8648 comparison children were included. All children were followed from the study date until a diagnosis of CKD, termination of insurance, or the end of 2013. Cox proportional hazards regressions were performed to compare the hazard ratios for CKD between the 2 cohorts. Incident cases of CKD were identified. After adjustment for potential confounders, the study cohort was independently associated with a higher risk of CKD (adjusted hazard ratio, 3.78; 95% confidence interval, 2.10–7.18). This population-based cohort study indicated that children with VUR have a higher risk of CKD than those without VUR.

Hemichorea induced by nonketotic hyperglycemia is a rare complication of diabetes mellitus. Here we present a case of 80-year-old female patient. She had a history of involuntary movement of the left body. Imaging examination showed abnormal signs of bilateral basal ganglia. After relevant treatment was conducted, the patient's hemichorea symptoms were significantly improved. Imaging features of nonketotic hyperglycemia Hemichorea plays a very important role in diagnosing and treatment of this disease.

Background: The use of three-dimensional(3D) printing is broadly across many medical specialties. It is an innovative, and rapidly growing technology to produce custom anatomical models and medical conditions models for medical teaching, surgical planning, and patient education. This study aimed to evaluate the accuracy and feasibility of 3D printing in creating a superficial femoral artery pseudoaneurysm model based on CT scans for endovascular training. Methods: A case of a left superficial femoral artery pseudoaneurysm was selected, and the 3D model was created using DICOM files imported into Materialise Mimics 22.0 and Materialise 3-Matic software, then printed using vat polymerization technology. Two 3D-printed models were created, and a series of comparisons were conducted between the 3D segmented images from CT scans and these two 3D-printed models. Ten comparisons involving internal diameters and angles of the specific anatomical location were measured. Results: The study found that the absolute mean difference in diameter between the 3D segmented images and the 3D printed models was 0.179±0.145 mm and 0.216±0.143mm, respectively, with no significant difference between the two sets of models. Additionally, the absolute mean difference in angle was 0.99±0.65° and 1.00±0.91°, respectively, and the absolute mean difference in angle between the two sets of data was not significant. Bland-Altman analysis confirmed a high correlation in dimension measurements between the 3D-printed models and segmented images. Furthermore, the accuracy of a 3D-printed femoral pseudoaneurysm model was further tested through the simulation of a superficial femoral artery pseudoaneurysm coiling procedure using the Philips Azurion7 in the angiography room. Conclusions: 3D printing is a reliable technique for producing a high accuracy 3D anatomical model that closely resemble a patient’s anatomy based on CT images. Additionally, 3D printing is a feasible and viable option for use in endovascular training and medical education. In general, 3D printing is an encouraging technology with diverse possibilities in medicine, including surgical planning, medical education, and medical device advancement.

This study aimed to develop and evaluate an AI-driven platform, the Adaptive RAG Assistant MRI Platform (ARAMP), for assisting in the diagnosis and reporting of brain metastases using post-contrast axial T1-weighted (AX_T1+C) MRI. In this retrospective study, 2447 cancer patients who underwent MRI between 2010 and 2022 were screened. A subset of 100 randomized patients with confirmed brain metastases and 100 matched non-cancer controls were selected for evaluation. ARAMP integrates quantitative radiomic feature extraction with an adaptive Retrieval-Augmented Generation (RAG) framework based on a large language model (LLM, GPT-4o), incorporating five authoritative medical references. Three board-certified neuroradiologists and an independent LLM (Gemini 2.0 Pro) assessed ARAMP performance. Metrics of the assessment included Pre-/Post-Trained Inference Difference, Inter-Inference Agreement, and Sensitivity. Post-training, ARAMP achieved a mean Inference Similarity score of 67.45%. Inter-Inference Agreement among radiologists averaged 30.20% (p = 0.01). Sensitivity for brain metastasis detection improved from 0.84 (pre-training) to 0.98 (post-training). ARAMP also showed improved reliability in identifying brain metastases as the primary diagnosis post-RAG integration. This adaptive RAG-based framework may improve diagnostic efficiency and standardization in radiological workflows.