Objective: To investigate biological markers associated with psychotic symptoms in patients with bipolar disorder (BD) based on electronic medical records of patients, and to develop an interpretable risk prediction model that supports the identification of high-risk individuals and that facilitates decision-making for providing clinical intervention in a timely manner.
Methods: A total of 2352 patients diagnosed with BD and admitted to West China Hospital, Sichuan University were enrolled using the electronic medical records system of the hospital. The participants were divided into two subgroups, the bipolar disorder depression (BDD) group and the bipolar disorder mania (BDM) group. The logistic regression algorithm was used to train and validate the prediction model, and interpretability methods were used to analyze the contribution of each feature to individuals and the effect of the features on specific target prediction decisions.
Results: The logistic regression model demonstrated robust predictive performance across the BD, BDD, and BDM cohorts, with areas under the curve (AUC) of the receiver operating characteristic curves always exceeding 81.6%. The core predictive features included platelet distribution width (PDW), fibrinogen (FIB), platelet large cell ratio (P-LCR), activated partial thromboplastin time (APTT), prothrombin time (PT), and triglyceride (TG). The logistic regression model exhibited strong interpretability and was combined with nomograms for intuitive risk quantification and individualized prediction.
Conclusion: The logistic regression model enables rapid and simple screening of BD patients with psychotic symptoms. Distinct patterns of changes observed in blood biomarkers of BDD and BDM subgroups enrich the understanding of the underlying pathophysiological mechanisms and highlight the importance of considering subtypes in the intervention and management of patients.
Objective: To establish a method for simultaneous determination of trace levels of microcystins, cylindrospermopsin, anatoxin, and nodularin in lake water based on liquid chromatography-tandem mass spectrometry (LC-MS/MS).
Methods: After being adjusted to alkaline conditions and mixed with six internal standards, the water samples were enriched using dual HLB and ENVI-Carb cartridges. The eluates were then evaporated under nitrogen, reconstituted, and subjected to instrumental analysis. Both water and acetonitrile containing 0.1% formic acid were used as mobile phases. An ACQUITY UPLC® BEH C18 column (150 mm × 2.1 mm, 1.7 μm) was selected to separate the target cyanotoxins. Multiple reaction monitoring was applied for data acquisition, and quantification was accomplished using internal standard methods.
Results: Within certain concentration ranges, all 14 cyanotoxins examined in the study showed good linearity, with all correlation coefficients greater than 0.998. When the water volume was 100 mL, the limits of detection and quantification for the 14 cyanotoxins were 0.1-0.9 ng/L and 0.3-2.9 ng/L, respectively, and spiked recoveries and relative standard deviations were 81.7%-132.9% and 1.2%-14.9%, respectively. In the 10 lake water samples analyzed, cylindrospermopsin, anatoxin-α, and multiple microcystins were detected.
Conclusion: The method developed in the study has high-throughput capacity, as well as high sensitivity, accuracy, and reliability. The method can be applied in the simultaneous detection of microcystins, cylindrospermopsin, anatoxin, and nodularin in lake water.
Objective: To optimize the real-time quantitative polymerase chain reaction (RT-qPCR) data analysis process through mathematical principles by replacing the biased [Formula: see text] method with a more rigorous [Formula: see text] method, thereby improving the accuracy of gene expression quantification analysis.
Methods: Essentially, the C T value serves as the exponent in a base-2 exponential equation within the logic of comparative C T method. In the traditional [Formula: see text] method, the arithmetic means of raw C T and ΔC T values are directly calculated and the exponential nature of C T data is overlooked, which may introduce systematic bias to the calculation results. We propose a new method, entitled the [Formula: see text] method, in which all calculations are based on the transformation of C T values into [Formula: see text]. This includes computing the relative initial expression levels of target and reference genes within each sample, the relative abundance of the target gene, and its fold change across groups. Statistical comparisons are then performed based on fold change values. By strictly adhering to the exponential nature of of C T values, the biases introduced by arithmetic averaging at the C T or ΔC T level are avoided. We applied this method to multiple RT-qPCR datasets to evaluate the differences between the traditional [Formula: see text] and the proposed [Formula: see text] methods in gene expression quantification, as well as the effect of the differences.
Results: In the original dataset from LIVAK and SCHMITTGEN, the two methods produced similar results. However, in the cadmium exposure experiment, findings from the [Formula: see text] method indicated that 8-hour cadmium exposure caused an increase of irg-6 gene expression in Caenorhabditis elegans from 1.314-fold to 7.125-fold (P = 0.0002). In contrast, findings from the [Formula: see text]method showed a fold change from 1.0 to 4.124 (P = 0.0015), a 70% difference between the two methods.
Conclusion: The [Formula: see text] method provides a mathematically more rigorous approach that more accurately reflects gene expression changes, particularly in experiments with high C T variability. It offers a more reliable computational paradigm for quantitative gene expression analysis.
Objective: To develop a precise method for analyzing urinary peptides based on electromembrane extraction (EME) combined with liquid chromatography-tandem mass spectrometry (LC-MS/MS), and to evaluate its potential applicability in tumor biomarker screening.
Methods: A total of 15 disease-associated peptides were selected as the target analytes. A supported liquid membrane (SLM) composed of n-octanol containing 5% di (2-ethylhexyl) phosphate was employed, with the donor phase being a 1∶1 mixture of urine and 100 mmol/L formic acid and urine, and the acceptor phase being 20 mmol/L formic acid containing 50% dimethyl sulfoxide (DMSO). After EME at 40 V for 15 min, the acceptor phase solution was analyzed by LC-MS/MS. Subsequently, the method, EME combined with LC-MS/MS (EME-LC-MS/MS), was preliminarily validated utilizing urine samples from 12 healthy controls and 7 patients with urinary system tumors.
Results: All 15 peptides exhibited excellent linearity in the range of 0.1-100.0 ng/mL (r ≥ 0.995), with the limits of detection (LODs) being 0.01-0.50 ng/mL and the limits of quantification (LOQs) being 0.03-1.50 ng/mL. The spiked recoveries ranged from 21.0% to 71.2%, with relative standard deviations (RSDs) of 0.8%-20.0% (n = 3). Small-sample analysis of clinical specimens revealed that the concentration of bradykinin 1-5 in the urine were significantly higher in tumor patients (median: 0.65 ng/mL) than that in healthy controls (median: 0.37 ng/mL) (P < 0.05), suggesting its potential as a specific biomarker for urinary system tumors.
Conclusion: The EME-LC-MS/MS method established in the study features simplicity, high efficiency, and high sensitivity, enabling precise determination of trace-level peptides in urine samples. Moreover, this approach provides a reliable methodological basis for disease biomarker screening and promotes the clinical application of electromembrane extraction.
Objective: To investigate the mechanisms underlying regional heterogeneity in the elevating patterns of palatal shelf during mammalian craniofacial development.
Methods: Using a mouse model of embryonic palatal development, we acquired coronal multi-plane slices of the palatal shelves before elevation (early E13.5), during elevation (late E13.5), and after elevation (early E14.5). Hematoxylin and eosin (HE) staining was performed to compare the morphological changes and spatial correlations between the palate and tongue. Immunofluorescence staining of myosin heavy chain 1 (MYH1), a marker found in slow muscle fibers and responsible for muscle contraction and movement, was performed to observe the tongue muscle development characteristics at different stages. We also observed changes in the palatal shelf elevating patterns at early E13.5 in the absence of the tongue through HE-stained in vitro palate organ culture. Further immunofluorescence staining of tenascin-C, an extracellular matrix protein, was performed to evaluate the effect of the tongue on the elevating pattern of the palatal shelf along the anterior-posterior axis.
Results: HE staining results of the coronal multi-plane slices showed that during the elevation period, from the posterior toward anterior, the coronal height of the tongue decreased, lateral inclination and flattening increased, but the sagittal length of the tongue increased. The elevating pattern of the palatal shelf changed from slow remodeling to rapid flipping, and MYH1 was abundantly expressed in both the internal and external muscle bundles of the tongue during this period. According to findings from in vitro cultivation of palatal organs, the posterior part of the palatal shelf elevated without forming new lateral lingual protrusions in the absence of the tongue. The regional expression pattern of tenascin-C was consistent with that observed before elevation. The posterior palate exhibited an elevation pattern similar to that of the anterior region.
Conclusion: The tongue may play a crucial role in shaping the posterior morphological remodeling and distinct elevation patterns of the palatal shelf.
Objective: To evaluate the feasibility and safety of a domestically developed, single-arm single-port robotic system for performing complex gynecological surgeries under extreme conditions, such as ultra-remote locations and high-altitude environments.
Methods: In November and December 2024, a surgeon on the campus of West China Second Hospital, Sichuan University in Chengdu remotely manipulated a domestically developed single-arm, single-port robotic surgical system via a high-speed, low-latency communication network to perform two telesurgical procedures. The first procedure was a transumbilical single-port robot-assisted laparoscopic total hysterectomy, bilateral salpingectomy, and left ovarian cystectomy on a patient with multiple uterine fibroids at the Maternity and Child Health Hospital of Xizang Autonomous Region (distance between Chengdu and Lhasa > 2000 km and altitude difference >3000 m). The second procedure was a transumbilical single-port robot-assisted laparoscopic total hysterectomy, bilateral salpingo-oophorectomy, and sentinel lymph node biopsy on a patient with FIGO stage IA endometrial cancer at Zhujiang Hospital, Southern Medical University in Guangzhou (the distance between Chengdu and Guangzhou > 1500 km). Perioperative data were collected and analyzed.
Results: Both procedures were successfully completed without conversion to laparotomy or the use of additional auxiliary ports. The operative times for the Chengdu-Lhasa and Chengdu-Guangzhou surgeries were 90 minutes and 135 minutes, respectively, with estimated blood loss ≤ 50 mL in both cases. The intraoperative bidirectional network latency remained around 40 ms, and the total end-to-end latency was less than 60 ms. The surgeon reported no perceptible delay in instrumental response. Both patients recovered well postoperatively, and no surgery-related complications or disease recurrence were observed during follow-up until July 2025.
Conclusion: This study provides preliminary evidence supporting the feasibility and safety of a domestically developed single-arm, single-port robotic system for performing complex gynecological surgeries in ultra-remote and high-altitude settings. This technical approach offers a promising solution to address geographic disparities in access to high-quality medical resources and demonstrates significant potential for improving the availability of advanced minimally invasive surgery in remote areas and regions of special settings.
Public health laboratory testing involves a wide range of sample types, complex matrices, diverse target analytes with varying concentrations, and multiple application contexts with different analytical requirements. As a critical step in public health laboratory analysis and testing, sample pretreatment plays a decisive role in ensuring the reproducibility and efficiency of the analytical methods. It directly affects the accuracy, sensitivity, and reliability of testing results, as well as the feasibility of downstream analyses. Traditional sample pretreatment techniques face persistent challenges, including low efficiency, limited throughput, restricted universal applicability, high organic solvent consumption, and poor compatibility with downstream analytical procedures. These limitations constrain their capacity to meet the evolving demands of research and practice in public health and preventive medicine. In recent years, technological advances have focused on improving efficiency and automation, enhancing selectivity and sensitivity, facilitating online testing capabilities, and promoting environmental sustainability. Sample pretreatment techniques in public health laboratory testing have been undergoing progressive upgrades, and numerous novel technologies have emerged. The paper provides a comprehensive review of new technologies and applications in the field. We focused on the development of new materials, the application of artificial intelligence, connections for online processing, and the approaches tailored to the demands of specific testing settings. We also discussed sample processing for omics analyses and mass spectrometry imaging methods relevant to public health laboratory testing. These advances are expected to support the development of greener and higher-throughput sample pretreatment and foster innovation in the public health laboratory testing system.
Objective: Focusing on gynecological surgery, we constructed a prediction model for surgical duration by extracting features from unstructured surgical planning texts and integrating multimodal data via artificial intelligence technology.
Methods: The clinical data of 34614 patients who underwent gynecologic surgeries at West China Second University Hospital, Sichuan University between January 2022 and October 2024 were collected. An embedding-transformer model was constructed to convert surgical planning texts into a one-dimensional numerical feature, referred to as the step feature. The predictive value of the step feature was assessed by comparing the performance improvements of linear regression, random forest, eXtreme Gradient Boosting (XGBoost), support vector regression, K-nearest neighbor regression, and artificial neural network algorithms in two scenarios-with and without the step feature as an input. The out-of-sample prediction accuracy of the models was assessed using mean absolute error (MAE), root mean squared error (RMSE), and R-squared (R 2). Furthermore, the model interpretability was examined using SHapley Additive exPlanations (SHAP) values.
Results: SHAP results showed that the step feature had the highest predictive contribution. Temporal factors in surgical scheduling also influenced gynecological surgery duration. The XGBoost model demonstrated optimal performance on the test set, significantly improving prediction accuracy with a 40.43% increase in R 2, while reducing MAE and RMSE by 21.27% and 20.13%, respectively, compared to the baseline model without the step feature.
Conclusion: The embedding-transformer model developed in this study effectively extracts features from surgical planning texts and enhances the predictive performance of machine learning models. The XGBoost prediction model can assist hospital administrators in implementing more refined management of gynecological surgeries and improving the utilization efficiency of surgical resources.
Objective: To establish a brucellosis monitoring and testing technique applicable for the rapid field screening of natural epidemic diseases.
Methods: A rapid testing technique for Brucella was developed based on a double-antibody sandwich testing model using gradient-variant quantum dots as fluorescent tracers. The sensitivity, linearity, precision, and specificity of the technique were evaluated using suspensions of standard Brucella strains. Methodological comparisons across different sample types were conducted to assess the consistency of the test results.
Results: The gradient-variant quantum dots detection method was evaluated with standard Brucella strains, exhibiting a sensitivity of 1 × 103 CFU/mL and a linear correlation coefficient (r) of 0.994 (95% CI, 0.933-1.055). The maximum coefficient of variation was 12.94% in repeated tests, showing good specificity. A comparative assessment of 305 clinical samples was conducted using the Brucella gradient-variant quantum dots detection method, the Rose Bengal plate agglutination test (RBT), and the serum agglutination test (SAT), yielding a Kappa value of 0.95, indicating almost perfect agreement. Additionally, a comparative assessment of 110 environmental samples collected on-site was conducted using the Brucella gradient-variant quantum dots detection method and quantitative real-time PCR (qPCR). The Kappa values for aerosol collection fluid, surface wipes, and wool samples were all above 0.83, demonstrating near-perfect agreement. For fecal and soil samples, the Kappa values were above 0.62, indicating substantial agreement.
Conclusion: The Brucella detection method based on gradient-variant quantum dots technology is simple and can be conducted rapidly. The detection method demonstrates high sensitivity, linearity, precision, and specificity. It shows consistent performance in clinical sample testing. It is well-suited for field rapid screening of natural epidemic diseases in field settings and shows good application prospects in the monitoring, prevention, and rapid detection of zoonotic diseases.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：