Objective: This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans. Methods: Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net’s segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness. Results: The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements. Conclusion Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Objective: This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans. Methods: Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net’s segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness. Results: The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements. Conclusion Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Objective: This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans. Methods: Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net’s segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness. Results: The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements. Conclusion Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Objective: This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans. Methods: Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net’s segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness. Results: The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements. Conclusion Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Objective: This study aims to overcome challenges in lumbar spine imaging, particularly lumbar spinal stenosis, by developing an automated segmentation model using advanced techniques. Traditional manual measurement and lesion detection methods are limited by subjectivity and inefficiency. The objective is to create an accurate and automated segmentation model that identifies anatomical structures in lumbar spine magnetic resonance imaging scans. Methods: Leveraging a dataset of 539 lumbar spinal stenosis patients, the study utilizes the residual U-Net for semantic segmentation in sagittal and axial lumbar spine magnetic resonance images. The model, trained to recognize specific tissue categories, employs a geometry algorithm for anatomical structure quantification. Validation metrics, like Intersection over Union (IOU) and Dice coefficients, validate the residual U-Net’s segmentation accuracy. A novel rotation matrix approach is introduced for detecting bulging discs, assessing dural sac compression, and measuring yellow ligament thickness. Results: The residual U-Net achieves high precision in segmenting lumbar spine structures, with mean IOU values ranging from 0.82 to 0.93 across various tissue categories and views. The automated quantification system provides measurements for intervertebral disc dimensions, dural sac diameter, yellow ligament thickness, and disc hydration. Consistency between training and testing datasets assures the robustness of automated measurements. Conclusion Automated lumbar spine segmentation with residual U-Net and deep learning exhibits high precision in identifying anatomical structures, facilitating efficient quantification in lumbar spinal stenosis cases. The introduction of a rotation matrix enhances lesion detection, promising improved diagnostic accuracy, and supporting treatment decisions for lumbar spinal stenosis patients.

Objective: aaWilson’s disease (WD) is a rare genetic disorder of copper metabolism, and longitudinal follow-up studies are limited. We performed a retrospective analysis to determine the clinical characteristics and long-term outcomes in a large WD cohort. Methods: aaMedical records of WD patients diagnosed from 2006–2021 at National Taiwan University Hospital were retrospectively evaluated for clinical presentations, neuroimages, genetic information, and follow-up outcomes. Results: aaThe present study enrolled 123 WD patients (mean follow-up: 11.12 ± 7.41 years), including 74 patients (60.2%) with hepatic features and 49 patients (39.8%) with predominantly neuropsychiatric symptoms. Compared to the hepatic group, the neuropsychiatric group exhibited more Kayser-Fleischer rings (77.6% vs. 41.9%, p < 0.01), lower serum ceruloplasmin levels (4.9 ± 3.9 vs. 6.3 ± 3.9 mg/dL, p < 0.01), smaller total brain and subcortical gray matter volumes (p < 0.0001), and worse functional outcomes during follow-up (p = 0.0003). Among patients with available DNA samples (n = 59), the most common mutations were p.R778L (allelic frequency of 22.03%) followed by p.P992L (11.86%) and p.T935M (9.32%). Patients with at least one allele of p.R778L had a younger onset age (p = 0.04), lower ceruloplasmin levels (p < 0.01), lower serum copper levels (p = 0.03), higher percentage of the hepatic form (p = 0.03), and a better functional outcome during follow-up (p = 0.0012) compared to patients with other genetic variations. Conclusion aaThe distinct clinical characteristics and long-term outcomes of patients in our cohort support the ethnic differences regarding the mutational spectrum and clinical presentations in WD.

OBJECTIVE: Altered event-related potential (ERP) performances have been noted in attention deficit hyperactivity disorder (ADHD) patients and reflect neurocognitive dysfunction. Whether these ERP alterations and correlated dysfunctions exist in healthy parents with ADHD offspring is worth exploring. METHODS: Thirteen healthy parents with ADHD offspring and thirteen healthy controls matched for age, sex and years of education were recruited. The auditory oddball paradigm was used to evaluate the P300 wave complex of the ERP, and the Wechsler Adult Intelligence Scale-Revised, Wisconsin Card Sorting Test, and continuous performance test were used to measure neurocognitive performance. RESULTS: Healthy parents with ADHD offspring had significantly longer auditory P300 latency at Fz than control group. However, no significant differences were found in cognitive performance. CONCLUSION: The presence of a subtle alteration in electro-neurophysiological activity without explicit neurocognitive dysfunction suggests potential candidate of biological marker for parents with ADHD offspring.
Adult ; Attention Deficit Disorder with Hyperactivity ; Biomarkers ; Cognition ; Education ; Evoked Potentials ; Humans ; Intelligence ; Parents ; Wisconsin

Arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVD/C) is predominantly an inherited cardiomyopathy with typical histopathological characteristics of fibro-fatty infiltration mainly involving the right ventricular (RV) inflow tract, RV outflow tract, and RV apex in the majority of patients. The above pathologic evolution frequently brings patients with ARVD/C to medical attention owing to the manifestation of syncope, sudden cardiac death (SCD), ventricular arrhythmogenesis, or heart failure. To prevent future or recurrent SCD, an implantable cardiac defibrillator (ICD) is highly desirable in patients with ARVD/C who had experienced unexplained syncope, hemodynamically intolerable ventricular tachycardia (VT), ventricular fibrillation, and/or aborted SCD. Notably, the management of frequent ventricular tachyarrhythmias in ARVD/C is challenging, and the use of antiarrhythmic drugs could be unsatisfactory or limited by the unfavorable side effects. Therefore, radiofrequency catheter ablation (RFCA) has been implemented to treat the drug-refractory VT in ARVD/C for decades. However, the initial understanding of the link between fibro-fatty pathogenesis and ventricular arrhythmogenesis in ARVD/C is scarce, the efficacy and prognosis of endocardial RFCA alone were limited and disappointing. The electrophysiologists had broken through this frontier after better illustration of epicardial substrates and broadly application of epicardial approaches in ARVD/C. In recent works of literature, the application of epicardial ablation also successfully results in higher procedural success and decreases VT recurrences in patients with ARVD/C who are refractory to the endocardial approach during long-term follow-up. In this article, we review the important evolution on the delineation of arrhythmogenic substrates, ablation strategies, and ablation outcome of VT in patients with ARVD/C.
Anti-Arrhythmia Agents ; Arrhythmogenic Right Ventricular Dysplasia ; Cardiomyopathies ; Catheter Ablation ; Catheters ; Death, Sudden, Cardiac ; Defibrillators ; Epicardial Mapping ; Follow-Up Studies ; Heart Failure ; Humans ; Prognosis ; Recurrence ; Syncope ; Tachycardia ; Tachycardia, Ventricular ; Ventricular Fibrillation

Arrhythmogenic right ventricular dysplasia/cardiomyopathy (ARVD/C) is predominantly an inherited cardiomyopathy with typical histopathological characteristics of fibro-fatty infiltration mainly involving the right ventricular (RV) inflow tract, RV outflow tract, and RV apex in the majority of patients. The above pathologic evolution frequently brings patients with ARVD/C to medical attention owing to the manifestation of syncope, sudden cardiac death (SCD), ventricular arrhythmogenesis, or heart failure. To prevent future or recurrent SCD, an implantable cardiac defibrillator (ICD) is highly desirable in patients with ARVD/C who had experienced unexplained syncope, hemodynamically intolerable ventricular tachycardia (VT), ventricular fibrillation, and/or aborted SCD. Notably, the management of frequent ventricular tachyarrhythmias in ARVD/C is challenging, and the use of antiarrhythmic drugs could be unsatisfactory or limited by the unfavorable side effects. Therefore, radiofrequency catheter ablation (RFCA) has been implemented to treat the drug-refractory VT in ARVD/C for decades. However, the initial understanding of the link between fibro-fatty pathogenesis and ventricular arrhythmogenesis in ARVD/C is scarce, the efficacy and prognosis of endocardial RFCA alone were limited and disappointing. The electrophysiologists had broken through this frontier after better illustration of epicardial substrates and broadly application of epicardial approaches in ARVD/C. In recent works of literature, the application of epicardial ablation also successfully results in higher procedural success and decreases VT recurrences in patients with ARVD/C who are refractory to the endocardial approach during long-term follow-up. In this article, we review the important evolution on the delineation of arrhythmogenic substrates, ablation strategies, and ablation outcome of VT in patients with ARVD/C.

Interleukin (IL)-20, a proinflammatory cytokine of the IL-10 family, is involved in acute and chronic renal failure. The aim of this study was to elucidate the role of IL-20 during diabetic nephropathy development. We found that IL-20 and its receptor IL-20R1 were upregulated in the kidneys of mice and rats with STZ-induced diabetes. In vitro, IL-20 induced MMP-9, MCP-1, TGF-β1 and VEGF expression in podocytes. IL-20 was upregulated by hydrogen peroxide, high-dose glucose and TGF-β1. In addition, IL-20 induced apoptosis in podocytes by activating caspase-8. In STZ-induced early diabetic nephropathy, IL-20R1-deficient mice had lower blood glucose and serum BUN levels and a smaller glomerular area than did wild-type controls. Anti-IL-20 monoclonal antibody (7E) treatment reduced blood glucose and the glomerular area and improved renal functions in mice in the early stage of STZ-induced diabetic nephropathy. ELISA showed that the serum IL-20 level was higher in patients with diabetes mellitus than in healthy controls. The findings of this study suggest that IL-20 induces cell apoptosis of podocytes and plays a role in the pathogenesis of early diabetic nephropathy.
Animals ; Apoptosis ; Blood Glucose ; Caspase 8 ; Diabetes Mellitus ; Diabetic Nephropathies* ; Enzyme-Linked Immunosorbent Assay ; Glucose ; Humans ; Hydrogen Peroxide ; In Vitro Techniques ; Interleukin-10 ; Interleukins ; Kidney ; Kidney Failure, Chronic ; Mice ; Podocytes* ; Rats ; Vascular Endothelial Growth Factor A