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.

BACKGROUND/OBJECTIVES: Heavy alcohol consumption causes the development of alcoholic liver disease (ALD), a neglected but important public health problem. Many studies have pointed out that probiotics could improve gut health, which is also considered to be a cause of ALD. Therefore, this study screened the probiotics, Lactobacillus casei GKC1 (GKC1), L. fermentum GKF3 (GKF3), Bifidobacterium lactis GKK2 (GKK2), L. rhamnosus GKLC1 (GKLC1), L. paracasei GKS6 (GKS6), and L. plantarum GKM3 (GKM3), for their potential benefits in alleviating ALD for applications to disease prevention. SUBJECTS/METHODS: C57BL/6N mice were divided into 8 groups (n = 6 in each): normal control, positive control (alcohol-diet fed), and treatments of feeding probiotics GKC1, GKF3, GKK2, GKLC1, GKS6, and GKM3 under an oral dose 0.82 g/kg B.W. per day by oral gavage. The experiment was conducted for 8 weeks, and the concentrations of alanine aminotransferase (ALT), aspartate aminotransferase, triglyceride (TG), and total cholesterol (TC) in mice were measured. The glutathione (GSH), catalase (CAT), and histology were analyzed after sacrifice. RESULTS: The results showed a decrease in the serum ALT, liver TG, and liver TC levels in the GKS6, GKM3, and GKLC1 groups compared to the positive control. In addition, the decreasing GSH and CAT levels were inhibited in the GKS6 and GKM3 groups. The histopathological results showed that all probiotics could reduce the accumulation of liver fat. Furthermore, there was a significant difference in GKLC1 with lower stomach damage compared to the alcohol-fed mice without any addition of probiotics. CONCLUSIONS GKLC1, GKS6, and GKM3 can be used as supplements for alleviating the development of ALD.

Objective: Patients with opioid use disorder (OUD) have impaired attention, inhibition control, and memory function. The aldehyde dehydrogenase 2 (ALDH2 ) gene has been associated with OUD and ALDH2 gene polymorphisms may affect aldehyde metabolism and cognitive function in other substance use disorder. Therefore, we aimed to investigate whether ALDH2 genotypes have significant effects on neuropsychological functions in OUD patients undergoing methadone maintenance therapy (MMT). Methods: OUD patients undergoing MMT were investigated and followed-up for 12 weeks. ALDH2 gene polymorphisms were genotyped. Connors’ Continuous Performance Test (CPT) and the Wechsler Memory Scale-Revised (WMS-R) were administered at baseline and after 12 weeks of MMT. Multivariate linear regressions and generalized estimating equations (GEEs) were used to examine the correlation between the ALDH2 genotypes and performance on the CPTs and WMS-R. Results: We enrolled 86 patients at baseline; 61 patients completed the end-of-study assessments. The GEE analysis showed that, after the 12 weeks of MMT, OUD patients with the ALDH2 *1/*2＋*2/*2 (ALDH2 inactive) genotypes had significantly higher commission error T-scores (p = 0.03), significantly lower hit reaction time T-scores (p = 0.04), and significantly lower WMS-R visual memory index scores (p = 0.03) than did patients with the ALDH2 1 */*1 (ALDH2 active) genotype. Conclusion OUD patients with the ALDH2 inactive genotypes performed worse in cognitive domains of attention, impulse control, and memory than did those with the ALDH2 active genotype. We conclude that the ALDH2 gene is important in OUD and is associated with neuropsychological performance after MMT.