Jiaqi Han

Background: Vaginal microbiome dysbiosis is associated with chronic inflammation, pathogen infections, adverse pregnancy outcomes, and increased susceptibility to sexually transmitted infections. However, the current detection methods mainly rely on microscopic examination, enzymology, and culture techniques, lacking precise detection of the species of the pathogen. Methods: A multiplex Droplet digital PCR (ddPCR) detection method of the most common targets related to vaginal diseases-related pathogens were established by retrospective study at Shanghai East Hospital. Swabs of 112 vaginosis patients and 102 non-non-vaginosis patients were detected and quantified the pathogen load of targets in vaginal swabs, targeting Lactobacillus, Gardnerella vaginalis, Candida species, Streptococcus agalactiae, and vaginal Trichomonas to comprehensively describe the vaginal microecological status and compared the results with existing cultivation methods and the double fluorescence staining method. Results: ddPCR test possesses a low limit of detection (LOD) of 5 CFU/mL, achieving PPA values of 87.50%-100.00% across all five targets and significantly increases the detection rate of Candida species (NPA 73.46%, Kappa=0.524 PGardnerella vaginalis (NPA 100%, Kappa=0.241 PStreptococcus agalactia (NPA 89.18%, Kappa=0.606 PTrichomonas (Kappa=0.865, P=1.000) and Lactobacillus (Kappa=0.801, P=0.002). In addition, this assay enables further classification and confirmation of the microbiota in specimens showing “increased small bacilli” or “flora inhibition” on conventional assays. Conclusions: Our study is the first to develop a droplet digital PCR method for multi-target detection and absolute quantification of vaginal microbiome. This method balances both detection efficiency and test convenience. It enables accurate target detection while permitting comprehensive evaluation of the overall vaginal microecology, demonstrating favorable analytical performance and broad clinical application prospects.

Background Parkinson’s Disease (PD) diagnosis lacks effective non-invasive markers, complicating early detection and timely intervention. Machine learning (ML) combined with clinical blood biomarkers may provide a feasible approach for early diagnosis and monitoring. Aim This study aims to construct and validate a non-invasive diagnostic model for PD using machine learning and routine clinical blood biomarkers, and identify key biomarkers linked to disease severity. Methods A total of 920 participants (428 PD and 492 non-PD) from two medical centers were included as training and validation sets. Biomarker selection was performed via least absolute shrinkage and selection operator (LASSO) and stepwise regression. Five machine learning models—logistic regression (LR), support vector machine (SVM), decision tree (DT), Naive Bayes (NB) and K-Nearest Neighbor (KNN)—were constructed and compared. The optimal model was interpreted using Shapley values (SHAP), and correlation with PD severity (Hoehn-Yahr stage) was assessed. Results The SVM model demonstrated the best external validation performance (AUC = 0.916, recall = 0.949, F1-score = 0.843). SHAP analysis revealed superoxide dismutase (SOD) contributed the most to the model prediction, followed by gender and uric acid (UA). Furthermore, albumin (ALB) and SOD showed significant negative correlations with PD severity. Conclusion The SVM-based diagnostic model effectively differentiates PD from controls using readily obtainable clinical biomarkers, offering promising clinical utility for PD screening, diagnosis, and progression monitoring.