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.

ObjectiveTo explore the current status and issues regarding the application of ancient books in clinical practice guidelines and expert consensus of traditional Chinese medicine （TCM） published in China， and to provide methodological recommendations for the incorporation of ancient books in the development of TCM guidelines. MethodsWe searched China National Knowledge Infrastructure （CNKI）， WanFang Data， VIP， SinoMed， PubMed， Embase， as well as six industry websites including China Association of Chinese Medicine， National Group Standards Information Platform， and Chinese Association of the Integration of Traditional and Western Medicine，etc. TCM clinical practice guidelines or expert consensus issued during January 1st， 2017， to November 26th， 2022 were searched. Clinical practice guidelines or expert consensus that explicitly referred to ancient books were included， and the content regarding the searching for ancient books， sources of access to ancient books， methods of evaluating the level of evidence， methods of evaluating the level of recommendation， and methods of evaluating the evidence for the ancient books were analysed. ResultsA total of 1,215 TCM clinical practice guidelines or expert consensus were retrieved， with 442 articles explicitly mentioning the application of ancient books， including 300 （67.87%） clinical practice guidelines and 142 （32.13%） expert consensus. Sixty of the 442 publications explicitly reported that ancient books searching had been conducted （13.57%）； among these 60 publications 27 （45.00%） explicitly reported ancient books searching strategies， and the most frequent method was manual searching with a total of 24 articles （40.00%）. The most popular search source was Chinese Medical Dictionary， a TCM classics database， with a total of 18 articles. 197 articles （44.57%） explicitly reported the evaluation criteria for the level of evidence， of which 141 articles （71.57%） involved the evaluation criteria for the ancient books； 413 articles （93.44%） mentioned ancient books in the recommendations， and only the source of formula name was mentioned in 409 （99.03%） of the publications. ConclusionThe current application of ancient books in TCM clinical practice guidelines and expert consensus is limited， with issues of non-standard searching and evaluation methods. Standar-dization and uniformity are needed in evidence grading and recommendation standards. Future research should clarify the scope and methods of applying ancient book， emphasize their integration with modern research evidence， and enhance their value and quality in the development of TCM clinical practice guidelines.

Objective To construct machine learning models that can be used to predict AUC fold change(FC)using a database of existing pharmacokinetic(PK)and drug-drug interaction(DDI)information,which can be used to explore the possibility of predicting existing drug interactions and to provide certain rational recommendations for clinical drug use.Methods PK data of DDIs and AUC fold change data were extracted from FDA-approved drug labels.Peptide and pharmacodynamic(PD)information related to drug interactions were retrieved through DrugBank,and PPDT identification of relevant peptide IDs was performed using Protein Resource(UniProt),and a matrix normalization code was used to generate multidimensional vector data that were easy to analysis.The effect of PPDT on the AUC,and the resulting multiplicity change was used as the dependent variable for machine learning model construction.The model with the smallest root mean square error(RMES)value was used for model construction to train a bagged decision tree(Bagged)prediction model.The models were tested using the trained models for some of the drug tests.The models were evaluated by reviewing the available literature findings on detection of drug interaction pairs and analyzing and comparing the predicted values.Results A total of 16 pairs of model drug pairs were tested for the effects of 16 drugs on tacrolimus,and it was found that the accuracy of the prediction of the presence or absence of drug interactions was 81.25％;the prediction results were classified according to the FDA standard classification of the strong and weak for the strength of drug interactions,and the results showed that the prediction of the strength of drug interactions,with a large deviation from the larger prediction was less.Conclusion The evaluation of the model to predict the presence or absence of drug interactions was general;however,after classifying the strengths and weaknesses of drug interactions,the prediction of drug interactions was better,and the prediction results indicated that the model prediction performance has a certain reference value for potential DDI assessment before clinical trials.

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 was aimed at investigating the carrier rate of, and molecular variation in, α- and β-globin gene mutations in Hunan Province. METHODS: We recruited 25,946 individuals attending premarital screening from 42 districts and counties in all 14 cities of Hunan Province. Hematological screening was performed, and molecular parameters were assessed. RESULTS: The overall carrier rate of thalassemia was 7.1%, including 4.83% for α-thalassemia, 2.15% for β-thalassemia, and 0.12% for both α- and β-thalassemia. The highest carrier rate of thalassemia was in Yongzhou (14.57%). The most abundant genotype of α-thalassemia and β-thalassemia was -α ^3.7/αα (50.23%) and β ^IVS-II-654/β ^N (28.23%), respectively. Four α-globin mutations [CD108 (ACC>AAC), CAP +29 (G>C), Hb Agrinio and Hb Cervantes] and six β-globin mutations [CAP +8 (C>T), IVS-II-848 (C>T), -56 (G>C), beta nt-77 (G>C), codon 20/21 (-TGGA) and Hb Knossos] had not previously been identified in China. Furthermore, this study provides the first report of the carrier rates of abnormal hemoglobin variants and α-globin triplication in Hunan Province, which were 0.49% and 1.99%, respectively. CONCLUSION Our study demonstrates the high complexity and diversity of thalassemia gene mutations in the Hunan population. The results should facilitate genetic counselling and the prevention of severe thalassemia in this region.
Humans ; beta-Thalassemia/genetics* ; alpha-Thalassemia/genetics* ; Hemoglobinopathies/genetics* ; China/epidemiology* ; High-Throughput Nucleotide Sequencing

Objective To explore the effect of microRNA-22-3p (miR-22-3p) regulating the expression of Kruppel-like factor 6 (KLF6) on the cardiomyocyte-like differentiation of bone marrow mesenchymal stem cell (BMSC). Methods Rat BMSC was isolated and cultured,and the third-generation BMSC was divided into a control group,a 5-azacytidine(5-AZA)group,a mimics-NC group,a miR-22-3p mimics group,a miR-22-3p mimics+pcDNA group,and a miR-22-3p mimics+pcDNA-KLF6 group.Real-time fluorescent quantitative PCR (qRT-PCR) was carried out to determine the expression of miR-22-3p and KLF6 in cells.Immunofluorescence staining was employed to detect the expression of Desmin,cardiac troponin T (cTnT),and connexin 43 (Cx43).Western blotting was employed to determine the protein levels of cTnT,Cx43,Desmin,and KLF6,and flow cytometry to detect the apoptosis of BMSC.The targeting relationship between miR-22-3p and KLF6 was analyzed by dual luciferase reporter gene assay. Results Compared with the control group,5-AZA up-regulated the expression of miR-22-3p (q=7.971,P<0.001),Desmin (q=7.876,P<0.001),cTnT (q=10.272,P<0.001),and Cx43 (q=6.256,P<0.001),increased the apoptosis rate of BMSC (q=12.708,P<0.001),and down-regulated the mRNA (q=20.850,P<0.001) and protein (q=11.080,P<0.001) levels of KLF6.Compared with the 5-AZA group and the mimics-NC group,miR-22-3p mimics up-regulated the expression of miR-22-3p (q=3.591,P<0.001;q=11.650,P<0.001),Desmin (q=5.975,P<0.001;q=13.579,P<0.001),cTnT (q=7.133,P<0.001;q=17.548,P<0.001),and Cx43 (q=4.571,P=0.037;q=11.068,P<0.001),and down-regulated the mRNA (q=7.384,P<0.001;q=28.234,P<0.001) and protein (q=4.594,P=0.036;q=15.945,P<0.001) levels of KLF6.The apoptosis rate of miR-22-3p mimics group was lower than that of 5-AZA group (q=8.216,P<0.001).Compared with the miR-22-3p mimics+pcDNA group,miR-22-3p mimics+pcDNA-KLF6 up-regulated the mRNA(q=23.891,P<0.001) and protein(q=13.378,P<0.001)levels of KLF6,down-regulated the expression of Desmin (q=9.505,P<0.001),cTnT (q=10.985,P<0.001),and Cx43 (q=8.301,P<0.001),and increased the apoptosis rate (q=4.713,P=0.029).The dual luciferase reporter gene experiment demonstrated that KLF6 was a potential target gene of miR-22-3p. Conclusion MiR-22-3p promotes cardiomyocyte-like differentiation of BMSC by inhibiting the expression of KLF6.
Animals ; Rats ; Myocytes, Cardiac ; Kruppel-Like Factor 6 ; Connexin 43 ; Desmin ; Cell Differentiation ; Azacitidine/pharmacology* ; Mesenchymal Stem Cells ; RNA, Messenger ; MicroRNAs

Objective:To analyze the correlation between body mass index (BMI) and the graded expression, subcutaneous depth, and diameter of elbow veins (cephalic vein, median vein, basilic vein) in patients with metabolic syndrome, to provide a method and theoretical basis for precise puncture of peripheral veins in obese patients.Methods:From January to October 2021, a total of 767 patients with metabolic syndrome with gastric volume reduction were selected as the study subjects by retrospective cohort study from the first Affiliated Hospital of Nanjing Medical University. According to the quartile Q1, M and Q3 of BMI level, they were divided into four groups: group A, group B, group C and group D. The subcutaneous depth and diameter of the cephalic vein, median vein and basilic vein were measured by B-ultrasound, and the three veins were evaluated and graded according to the grading criteria of superficial veins.The correlation between BMI and the subcutaneous depth and diameter of the three elbow veins was analyzed, and collected data such as puncture method and puncture times. Results:There was no significant correlation between BMI and subcutaneous depth and vessel diameter of the basilic vein ( r = 0.041 and 0.046, both P>0.05), the level of BMI was positively correlated with the subcutaneous depth and diameter of cephalic vein ( r = 0.275 and 0.117, both P<0.05) and median vein ( r = 0.236 and 0.148, both P<0.05), and a linear regression relationship was found ( OR values were 1.013-1.031, all P<0.05). 187 cases had direct puncture conditions under direct vision, and the success rate of one puncture was 86.63%(162/187).Venipuncture was completed under the guidance of B-mode ultrasound for 25 cases with failure under direct vision and 580 cases without direct puncture conditions under direct vision, the success rate of one puncture was 98.51% (596/605). Conclusions:With the increase of BMI level in patients with metabolic syndrome, the depth and diameter of both cephalic vein and median vein increase, venous exposure is difficult to express. The visual vein puncture guided by B-ultrasound is more accurate and catheterization is more reliable.