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 aimed to evaluate whether the onset of the plateau phase of slow hepatitis B surface antigen decline in patients with chronic hepatitis B treated with intermittent interferon therapy is related to the frequency of dendritic cell subsets and expression of the costimulatory molecules CD40,CD80,CD83,and CD86. Method This was a cross-sectional study in which patients were divided into a natural history group(namely NH group),a long-term oral nucleoside analogs treatment group(namely NA group),and a plateau-arriving group(namely P group).The percentage of plasmacytoid dendritic cell and myeloid dendritic cell subsets in peripheral blood lymphocytes and monocytes and the mean fluorescence intensity of their surface costimulatory molecules were detected using a flow cytometer. Results In total,143 patients were enrolled(NH group,n = 49;NA group,n = 47;P group,n = 47).The results demonstrated that CD141/CD1c double negative myeloid dendritic cell(DNmDC)/lymphocytes and monocytes(%)in P group(0.041[0.024,0.069])was significantly lower than that in NH group(0.270[0.135,0.407])and NA group(0.273[0.150,0.443]),and CD86 mean fluorescence intensity of DNmDCs in P group(1832.0[1484.0,2793.0])was significantly lower than that in NH group(4316.0[2958.0,5169.0])and NA group(3299.0[2534.0,4371.0]),Adjusted P all<0.001. Conclusion Reduced DNmDCs and impaired maturation may be associated with the onset of the plateau phase during intermittent interferon therapy in patients with chronic hepatitis B.

Objective To explore characteristics of clinical parameters and cytokines in patients with drug-induced liver injury(DILI)caused by different drugs and their correlation with clinical indicators. Method The study was conducted on patients who were up to Review of Uncertainties in Confidence Assessment for Medical Tests(RUCAM)scoring criteria and clinically diagnosed with DILI.Based on Chinese herbal medicine,cardiovascular drugs,non-steroidal anti-inflammatory drugs(NSAIDs),anti-infective drugs,and other drugs,patients were divided into five groups.Cytokines were measured by Luminex technology.Baseline characteristics of clinical biochemical indicators and cytokines in DILI patients and their correlation were analyzed. Results 73 patients were enrolled.Age among five groups was statistically different(P=0.032).Alanine aminotransferase(ALT)(P=0.033)and aspartate aminotransferase(AST)(P=0.007)in NSAIDs group were higher than those in chinese herbal medicine group.Interleukin-6(IL-6)and tumor necrosis factor alpha(TNF-α)in patients with Chinese herbal medicine(IL-6:P<0.001;TNF-α:P<0.001)and cardiovascular medicine(IL-6:P=0.020;TNF-α:P=0.001)were lower than those in NSAIDs group.There was a positive correlation between ALT(r=0.697,P=0.025),AST(r=0.721,P=0.019),and IL-6 in NSAIDs group. Conclusion Older age may be more prone to DILI.Patients with NSAIDs have more severe liver damage in early stages of DILI,TNF-α and IL-6 may partake the inflammatory process of DILI.

Objective To evaluate the in vitro effect of combinations of 5 β-lactam antibiotics with different β-lac-tamase inhibitors on the activity of multidrug-resistant Mycobacterium tuberculosis(MDR-TB),and identify the most effective combination of β-lactam antibiotics and β-lactamase inhibitors against MDR-TB.Methods MDR-TB strains collected in Henan Province Antimicrobial Resistance Surveillance Project in 2021 were selected.The mini-mum inhibitory concentrations(MIC)of 5 β-lactam antibiotics or combinations with different β-lactamase inhibitors on clinically isolated MDR-TB strains were measured by MIC detection method,and the blaC mutation of the strains was analyzed by polymerase chain reaction(PCR)and DNA sequencing.Results A total of 105 strains of MDR-TB were included in the analysis.MIC detection results showed that doripenem had the highest antibacterial activity against MDR-TB,with a MIC50 of 16 μg/mL.MIC values of most β-lactam antibiotics decreased significantly after combined with β-lactamase inhibitors.A total of 13.33％(n=14)strains had mutations in blaC gene,mainly 3 nu-cleotide substitution mutations,namely AGT333AGG,AAC638ACC and ATC786ATT.BlaC proteins Ser111 Arg and Asn213Thr enhanced the synergistic effect of clavulanic acid/sulbactam and meropenem on MDR-TB compared with synonymous single-nucleotide mutation.Conclusion The combination of doripenem and sulbactam has the strongest antibacterial activity against MDR-TB.Substitution mutations of BlaC protein Ser111 Arg and Asn213Thr enhances the sensitivity of MDR-TB to meropenem through the synergy with clavulanic acid/sulbactam.

Objective:To investigate the application value of single-sperm sequencing in resolving the carrier status of preim-plantation genetic testing(PGT)for chromosomal structural rearrangements in Robertsonian translocations.Methods:Haplotypes were constructed by single-sperm isolation combined with single-sperm sequencing for a patient with 45,XY,der(13;14)(q10;q10).Twenty single-sperm samples were isolated by mechanical braking and subjected to whole-genome amplification(WGA),and then the Asian Screening Array(ASA)gene chip was used to detect the 183 708 single nucleotide polymorphisms(SNP)of the WGA products.The single sperm associated with the translocation that could be used as haplotype inference was detected by copy number variation(CNV)sequencing,and the chromosomal haplotypes with normal and Robertsonian translocations were inferred.Three biopsy samples of embryonic trophoblast cells were used as the objects.After whole-genome amplification,high-throughput sequencing was employed to determine the status of the translocation chromosome carried by the embryos.The available blastocysts were selected for transfer,and the amniotic fluid samples were taken at 18 weeks of gestation to confirm whether the fetus carried the pathogenic muta-tion.Results:A total of 6 037 SNP sites were screened by single-sperm sequencing,and 30 sites selected to distinguish normal and translocation haplotypes.Preimplantation haplotype analysis showed that all the three embryos were euploids without Robertsonian translocation chromosome.Genetic testing of amniotic fluid in the second trimester confirmed that the karyotype of the fetus was 46,XN,carrying no Robertsonian translocation chromosome.Conclusion:For male carriers of Robertsonian translocation,single sperm sequencing can be used to screen SNP sites to construct haplotypes for distinguishing normal and Robertsonian translocation em-bryos,and to provide a basis for embryo selection by preimplantation chromosomal structural genetic testing.

AIM To investigate the effects of total Astragalus saponins on the improvement of sarcopenia in a rat model of type 2 diabetes mellitus(T2DM).METHODS The rats were divided into the normal group for a normal feeding and the model group for the feeding of high-sugar and high-fat diet combined with intraperitoneal injection of STZ to establish a T2DM model.The successful model rats were randomly divided into the model group,the metformin group(0.2 g/kg)and the total Astragalus saponins group(80 mg/kg),and given corresponding doses of drugs by gavage.After 12 weeks administration,the rats had their FBG,postprandial blood glucose(PG2h)and wet weight of skeletal muscle measured;their serum levels of INS,C-peptide(C-P),IGF-1,TNF-α and IL-1β detected by ELISA;their morphological changes of skeletal muscle observed by HE staining;their protein expressions of PI3K,p-Akt,mTOR,S6K1,FoxO1 and Murf1 in skeletal muscle detected by Western blot;and their mRNA expressions of Pi3k,Akt and mtor in skeletal muscle detected by RT-qPCR method.RESULTS Compared with the model group,the total Astragalus saponins group displayed decreased levels of FBG,PG2h,OGTT-AUC,HOMA-IR,TNF-α and IL-1β(P＜0.01);increased levels of INS,C-P,IGF-1 and wet weight of skeletal muscle(P＜0.05,P＜0.01);improved skeletal muscle atrophy and increased protein expressions of PI3K,p-Akt,mTOR and S6K1 in skeletal muscle(P＜0.05,P＜0.01);decreased protein expressions of FoxO1 and Murf1(P＜0.05,P＜0.01);and increased mRNA expressions of Pi3k,Akt and mtor(P＜0.01).CONCLUSION The improvement effects of total Astragalus saponins on sarcopenia in T2DM rats may be associated with the regulation of PI3K/Akt/mTOR and PI3K/Akt/FoxO1 pathways.

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.