ObjectivePrimary liver cancer, predominantly hepatocellular carcinoma (HCC), is a significant global health issue, ranking as the sixth most diagnosed cancer and the third leading cause of cancer-related mortality. Accurate and early diagnosis of HCC is crucial for effective treatment, as HCC and non-HCC malignancies like intrahepatic cholangiocarcinoma (ICC) exhibit different prognoses and treatment responses. Traditional diagnostic methods, including liver biopsy and contrast-enhanced ultrasound (CEUS), face limitations in applicability and objectivity. The primary objective of this study was to develop an advanced, light-weighted classification network capable of distinguishing HCC from other non-HCC malignancies by leveraging the automatic analysis of brightness changes in CEUS images. The ultimate goal was to create a user-friendly and cost-efficient computer-aided diagnostic tool that could assist radiologists in making more accurate and efficient clinical decisions. MethodsThis retrospective study encompassed a total of 161 patients, comprising 131 diagnosed with HCC and 30 with non-HCC malignancies. To achieve accurate tumor detection, the YOLOX network was employed to identify the region of interest (ROI) on both B-mode ultrasound and CEUS images. A custom-developed algorithm was then utilized to extract brightness change curves from the tumor and adjacent liver parenchyma regions within the CEUS images. These curves provided critical data for the subsequent analysis and classification process. To analyze the extracted brightness change curves and classify the malignancies, we developed and compared several models. These included one-dimensional convolutional neural networks (1D-ResNet, 1D-ConvNeXt, and 1D-CNN), as well as traditional machine-learning methods such as support vector machine (SVM), ensemble learning (EL), k-nearest neighbor (KNN), and decision tree (DT). The diagnostic performance of each method in distinguishing HCC from non-HCC malignancies was rigorously evaluated using four key metrics: area under the receiver operating characteristic (AUC), accuracy (ACC), sensitivity (SE), and specificity (SP). ResultsThe evaluation of the machine-learning methods revealed AUC values of 0.70 for SVM, 0.56 for ensemble learning, 0.63 for KNN, and 0.72 for the decision tree. These results indicated moderate to fair performance in classifying the malignancies based on the brightness change curves. In contrast, the deep learning models demonstrated significantly higher AUCs, with 1D-ResNet achieving an AUC of 0.72, 1D-ConvNeXt reaching 0.82, and 1D-CNN obtaining the highest AUC of 0.84. Moreover, under the five-fold cross-validation scheme, the 1D-CNN model outperformed other models in both accuracy and specificity. Specifically, it achieved accuracy improvements of 3.8% to 10.0% and specificity enhancements of 6.6% to 43.3% over competing approaches. The superior performance of the 1D-CNN model highlighted its potential as a powerful tool for accurate classification. ConclusionThe 1D-CNN model proved to be the most effective in differentiating HCC from non-HCC malignancies, surpassing both traditional machine-learning methods and other deep learning models. This study successfully developed a user-friendly and cost-efficient computer-aided diagnostic solution that would significantly enhances radiologists’ diagnostic capabilities. By improving the accuracy and efficiency of clinical decision-making, this tool has the potential to positively impact patient care and outcomes. Future work may focus on further refining the model and exploring its integration with multimodal ultrasound data to maximize its accuracy and applicability.

OBJECTIVE: This study aimed to explore the association between body mass index (BMI) and mortality based on the 10-year population-based multicenter prospective study. METHODS: A general population-based multicenter prospective study was conducted at four sites in rural China between 2013 and 2023. Multivariate Cox proportional hazards models and restricted cubic spline analyses were used to assess the association between BMI and mortality. Stratified analyses were performed based on the individual characteristics of the participants. RESULTS: Overall, 19,107 participants with a sum of 163,095 person-years were included and 1,910 participants died. The underweight (< 18.5 kg/m ²) presented an increase in all-cause mortality (adjusted hazards ratio [ aHR] = 2.00, 95% confidence interval [ CI]: 1.66-2.41), while overweight (≥ 24.0 to < 28.0 kg/m ²) and obesity (≥ 28.0 kg/m ²) presented a decrease with an aHR of 0.61 (95% CI: 0.52-0.73) and 0.51 (95% CI: 0.37-0.70), respectively. Overweight ( aHR = 0.76, 95% CI: 0.67-0.86) and mild obesity ( aHR = 0.72, 95% CI: 0.59-0.87) had a positive impact on mortality in people older than 60 years. All-cause mortality decreased rapidly until reaching a BMI of 25.7 kg/m ² ( aHR = 0.95, 95% CI: 0.92-0.98) and increased slightly above that value, indicating a U-shaped association. The beneficial impact of being overweight on mortality was robust in most subgroups and sensitivity analyses. CONCLUSION This study provides additional evidence that overweight and mild obesity may be inversely related to the risk of death in individuals older than 60 years. Therefore, it is essential to consider age differences when formulating health and weight management strategies.
Humans ; Body Mass Index ; China/epidemiology* ; Male ; Female ; Middle Aged ; Prospective Studies ; Rural Population/statistics & numerical data* ; Aged ; Follow-Up Studies ; Adult ; Mortality ; Cause of Death ; Obesity/mortality* ; Overweight/mortality*

Objective: To establish a machine learning-based precision blood donor recruitment model at the county level and assess its generalizability and applicability. Methods: A retrospective study was conducted using blood donation and SMS recruitment data from the Taicang Branch of the Suzhou Blood Center between 2019 and 2024. Multiple machine learning algorithms were employed, including extreme gradient boosting, support vector machine, k-nearest neighbor, logistic regression, decision tree, random forest, and multilayer perceptron. These were combined with techniques such as synthetic minority oversampling, undersampling, and cost-sensitive learning (using MFE and MSFE loss functions). Model parameters were optimized through grid search to identify the best-performing model. Results: In a prospective comparative study against conventional methods, the machine learning models increased the recruitment success rate among high-willingness donors by an average of 129.15%, and the recruitment efficiency per SMS improved by 125.02% compared with the traditional method. Under full-scale SMS sending, the recruitment rate per SMS increased by 42.61%, and SMS sending efficiency improved by 31.77%, significantly enhancing recruitment performance. Conclusion: This study represents the first application of a machine learning-based precision donor recruitment model at the county-level in China. The precise recruitment framework not only improves recruitment efficiency and reduces recruitment costs but also demonstrates strong scalability and generalizability. It provides a scientific and feasible intelligent pathway to ensure the safety and sustainability of the blood supply.

Alzheimer's disease (AD) is a progressive neurodegenerative disease associated with dysfunctions related to thinking, learning, and memory of the brain. AD has multiple pathological characteristics with complicated causes, constructing a suitable pathological model is crucial for the research of AD. Microfluidic chip technology integrates multiple functional units on a chip, which can realize microenvironmental control similar to the physiological environment. It is well applied in the construction of pathological model, early diagnosis as well as drug screening of AD. This paper focuses on the construction of AD microfluidic chips model from the perspective of cell type, culture formats and the chips structure as well as the research progress of microfluidic chips in AD application based on the pathological characteristics of AD, which will provide a reference for further elucidation of AD mechanism and drug development.

Biosensor analysis technology is a kind of technology with high specificity that can convert biological reactions into optical and electrical signals. In the development of drugs for Alzheimer's disease (AD), according to different disease hypotheses and targets, this technology plays an important role in confirming targets and screening active compounds. This paper briefly describes the pathogenesis of AD and the current situation of therapeutic drugs, introduces three biosensor analysis techniques commonly used in the discovery of AD drugs, such as surface plasmon resonance (SPR), biolayer interferometry (BLI) and fluorescence analysis technology, explains its basic principle and application progress, and summarizes their advantages and limitations respectively.

Objective:To compare the morphological differences of psoas major muscles between patients with lumbar disc herniation (LDH) of lower limb pain and lumbocrural pain based on CT imaging data.Methods:Sixty patients with LDH admitted to Department of Neurosurgery, 900 th Hospital of PLA Joint Logistic Team from January 2012 to February 2023 were included. According to clinical symptoms, they were divided into lower limb pain group and lumbocrural pain group ( n=30). 3D CT images of the psoas major muscles in the 2 groups were reconstructed; the longest transverse axis perpendicular to the longitudinal axis of the psoas major muscle was chosen as the cross-sectional area, and the maximum psoas major muscle cross-sectional area was calculated; maximum psoas major muscle cross-sectional area index (PI max) was defined as ratio of maximum psoas major muscle cross-sectional area and L 5 vertebral cross-sectional area. PI max difference between lower limb pain group and lumbocrural pain group was compared; PI max difference among patients with different pain degrees (visual analog scale [VAS] scores) or pain courses was further compared in both lower limb pain group and lumbocrural pain group. Pearson correlation was used to analyze the correlations of PI max with pain degree and pain course in the 2 groups. Results:PI max in lower limb pain group was significantly larger than that in lumbocrural pain group (0.62±0.05 vs. 0.54±0.04, t=7.320, P<0.001). PI max in patients with severe pain from both lower limb pain group and lumbocrural pain group was significantly smaller than that in patients with moderate pain (0.61±0.05 vs. 0.65±0.04, t=2.422, P=0.022; 0.53±0.03 vs. 0.58±0.04, t=3.502, P=0.002). PI max in patients with short pain course from both lower limb pain group and lumbocrural pain group was significantly larger than that in patients with long pain course (0.64±0.05 vs. 0.59±0.04, t=2.570, P=0.016; 0.57±0.04 vs. 0.53±0.03, t=2.941, P=0.007). Pearson correlation showed that PI max was negatively correlated with pain degree and pain course in LDH patients from both groups ( P<0.05). Conclusion:Atrophy of psoas major muscles in LDH patients is aggravated with increased pain degree and pain course.

Humans ; Aged ; Lung Diseases ; Pneumonia/drug therapy* ; Adrenal Cortex Hormones/therapeutic use* ; Pulmonary Disease, Chronic Obstructive/drug therapy* ; Administration, Inhalation ; Community-Acquired Infections/drug therapy*

Objective To compare the effects of 20 mg qd and 10 mg bidadministration of iprrazole enteric-coated tablets on the control of gastric acid in healthy subjects.Methods A randomized,single-center,parallel controlled trial was designed to include 8 healthy subjects.Randomly divided into 2 groups,20 mg qd administration group:20 mg enteric-coated tablets of iprrazole in the morning;10 mg bid administration group:10 mg enteric-coated tablets of iprrazole in the morning and 10 mg in the evening.The pH values in the stomach of the subjects before and 24 h after administration were monitored by pH meter.The plasma concentration of iprazole after administration was determined by HPLC-MS/MS.The main pharmacokinetic parameters were calculated by Phoenix WinNonlin(V8.0)software.Results The PK parameters of iprrazole enteric-coated tablets and reference preparations in fasting group were as follows:The Cmax of 20 mg qd group and 10 mg bid group were(595.75±131.15)and(283.50±96.98)ng·mL-1;AUC0-t were(5 531.94±784.35)and(4 686.67±898.23)h·ng·mL-1;AUC0-∞ were(6 003.19±538.59)and(7 361.48±1 816.77)h·ng·mL-1,respectively.The mean time percentage of gastric pH＞3 after 20 mg qd and 10 mg bid were 82.64％and 61.92％,and the median gastric pH within 24 h were 6.25±1.49 and 3.53±2.05,respectively.The mean gastric pH values within 24 h were 5.71±1.36 and 4.23±1.45,respectively.The correlation analysis of pharmacokinetic/pharmacodynamics showed that there was no significant correlation between the peak concentration of drug in plasma and the inhibitory effect of acid.Conclusion Compared with the 20 mg qd group and the 10 mg bid group,the acid inhibition effect is better,the administration times are less,and the safety of the two administration regimes is good.

Objective:To evaluate the diagnostic efficacy and practicality of the Jaundice color card (JCard) as a screening tool for neonatal jaundice.Methods:Following the standards for reporting of diagnostic accuracy studies (STARD) statement, a multicenter prospective study was conducted in 9 hospitals in China from October 2019 to September 2021. A total of 845 newborns who were admitted to the hospital or outpatient department for liver function testing due to their own diseases. The inclusion criteria were a gestational age of ≥35 weeks, a birth weight of ≥2 000 g, and an age of ≤28 days. The neonate′s parents used the JCard to measure jaundice at the neonate′s cheek. Within 2 hours of the JCard measurement, transcutaneous bilirubin (TcB) was measured with a JH20-1B device and total serum bilirubin (TSB) was detected. The Pearson′s correlation analysis, Bland-Altman plots and the receiver operating characteristic (ROC) curve were used for statistic analysis.Results:Out of the 854 newborns, 445 were male and 409 were female; 46 were born at 35-36 weeks of gestational age and 808 were born at ≥37 weeks of gestational age. Additionally, 432 cases were aged 0-3 days, 236 cases were aged 4-7 days, and 186 cases were aged 8-28 days. The TSB level was (227.4±89.6) μmol/L, with a range of 23.7-717.0 μmol/L. The JCard level was (221.4±77.0) μmol/L and the TcB level was (252.5±76.0) μmol/L. Both the JCard and TcB values showed good correlation ( r=0.77 and 0.80, respectively) and agreements (96.0% (820/854) and 95.2% (813/854) of samples fell within the 95% limits of agreement, respectively) with TSB. The JCard value of 12 had a sensitivity of 0.93 and specificity of 0.75 for identifying a TSB ≥205.2?μmol/L, and a sensitivity of 1.00 and specificity of 0.35 for identifying a TSB ≥342.0?μmol/L. The TcB value of 205.2?μmol/L had a sensitivity of 0.97 and specificity of 0.60 for identifying TSB levels of 205.2 μmol/L, and a sensitivity of 1.00 and specificity of 0.26 for identifying TSB levels of 342.0 μmol/L. The areas under the ROC curve (AUC) of JCard for identifying TSB levels of 153.9, 205.2, 256.5, and 342.0 μmol/L were 0.96, 0.92, 0.83, and 0.83, respectively. The AUC of TcB were 0.94, 0.91, 0.86, and 0.87, respectively. There were both no significant differences between the AUC of JCard and TcB in identifying TSB levels of 153.9 and 205.2 μmol/L (both P>0.05). However, the AUC of JCard were both lower than those of TcB in identifying TSB levels of 256.5 and 342.0 μmol/L (both P<0.05). Conclusions:JCard can be used to classify different levels of bilirubin, but its diagnostic efficacy decreases with increasing bilirubin levels. When TSB level are ≤205.2 μmol/L, its diagnostic efficacy is equivalent to that of the JH20-1B. To prevent the misdiagnosis of severe jaundice, it is recommended that parents use a low JCard score, such as 12, to identify severe hyperbilirubinemia (TSB ≥342.0 μmol/L).