Background: To optimize seasonal influenza vaccination programs in regions with potentially complicated seasonal patterns, the epidemiological characteristics of seasonal influenza activity in a subtropical city of China were explored. Materials and Methods: Influenza virus data of patients with influenza-like illness (ILI) during 2013–2019 were collected from two sentinel hospitals in a subtropical region of China, Yichang city. The influenza virus positive rate among sampled ILI cases served as a proxy to estimate influenza seasonal characteristics, including periodicity, duration, peaks, and predominant subtypes/lineages. Epidemiological features of different years, seasons and age groups were analyzed, and vaccine mismatches were identified. Results: In total, 8693 ILI cases were included; 1439 (16.6%) were laboratory-confirmed influenza cases. The influenza A positive rate (10.6%) was higher than the influenza B positive rate (5.9%). There were three influenza circulation patterns in Yichang: (1) annual periodicity (in 2013–2014, 2015–2016 and 2018–2019), (2) semiannual periodicity (in 2014–2015), and (3) year-round periodicity (in 2016–2017 and 2017–2018). Summer epidemics existed in two of the six years and were dominated by influenza A/H3N2. Winter and spring epidemics occurred in five of the six years, and A/H1N1, A/H3N2, B/Victoria, and B/Yamagata were codominant. During the study period, the predominant lineages, B/Victoria in 2015-16 and B/Yamagata in 2017–2018, were both mismatched with the influenza B component of the trivalent vaccine. Children 5–14 years old (26.4%) and individuals over 60 years old (16.9%) had the highest influenza positive rates. Conclusions: The seasonal epidemic period and the predominant subtype/lineage of influenza viruses in Yichang city are complex. Influenza vaccination timing and strategies need to be optimized according to the local features of influenza virus activity.

Environmental pollution, aging, emerging infectious diseases, and unhealthy lifestyles are affecting human health and resulting in serious social and economic burdens. The government-led healthcare big data platform in Yichang has continuously worked towards exploring the use of comprehensive health-related information through top-level design and scientific planning. The platform is based on the following principle: “Openness, inclusiveness, and win-win cooperation.” So far, by relying on one-to-one verification, comparison, correlation, and correction with the source, the platform has succeeded in establishing a unique foundation to benefit the people, government, medical care, and health service. Its successes can hopefully act as a positive case for exploring domestic healthcare big data.

Introduction: Modeling on infectious diseases is significant to facilitate public health policymaking. There are two main mathematical methods that can be used for the simulation of the epidemic and prediction of optimal early warning timing: the logistic differential equation (LDE) model and the more complex generalized logistic differential equation (GLDE) model. This study aimed to compare and analyze these two models.Methods: We collected data on (coronavirus disease 2019) COVID-19 and four other infectious diseases and classified the data into four categories: different transmission routes, different epidemic intensities, different time scales, and different regions, using R² to compare and analyze the goodness-of-fit of LDE and GLDE models.Results: Both models fitted the epidemic curves well, and all results were statistically significant. The R² test value of COVID-19 was 0.924 (p < 0.001) fitted by the GLDE model and 0.916 (p < 0.001) fitted by the LDE model. The R² test value varied between 0.793 and 0.966 fitted by the GLDE model and varied between 0.594 and 0.922 fitted by the LDE model for diseases with different transmission routes. The R² test values varied between 0.853 and 0.939 fitted by the GLDE model and varied from 0.687 to 0.769 fitted by the LDE model for diseases with different prevalence intensities. The R² test value varied between 0.706 and 0.917 fitted by the GLDE model and varied between 0.410 and 0.898 fitted by the LDE model for diseases with different time scales. The GLDE model also performed better with nation-level data with the R² test values between 0.897 and 0.970 vs. 0.731 and 0.953 that fitted by the LDE model. Both models could characterize the patterns of the epidemics well and calculate the acceleration weeks.Conclusion: The GLDE model provides more accurate goodness-of-fit to the data than the LDE model. The GLDE model is able to handle asymmetric data by introducing shape parameters that allow it to fit data with various distributions. The LDE model provides an earlier epidemic acceleration week than the GLDE model. We conclude that the GLDE model is more advantageous in asymmetric infectious disease data simulation.

Background: In resource-limited areas, severe shortages of radiologists contribute to high rates of missed pulmonary tuberculosis (PTB) cases when relying solely on conventional chest X-ray (CXR). Although artificial intelligence–powered computer-aided detection (CAD) has proven effective in PTB diagnosis, its real-world performance remains underexplored. Objective: This study aimed to evaluate the real-world diagnostic yield of CAD technology as a triage tool for detecting PTB in primary health care facilities in high-burden areas. Methods: We conducted a retrospective paired-design diagnostic yield study using CXR images collected from 7 county- and 32 township-level health care facilities in Yichang city between 2022 and 2024 year. All images were retrospectively reprocessed with CAD software (JF CXR-1), and the original reports interpreted by radiologists at the time of patient admission were extracted. CAD and radiologist performances were compared using 2 primary evaluation indicators—diagnostic yield among diagnosed cases (DYD) and positive predictive value (PPV). Subgroup analysis (by region, age, sex, health care facility tier, and patient category) and sensitivity analysis were conducted to assess the robustness of the results. Results: Among 93,319 enrolled study patients, including 273 (0.3%) bacteriologically confirmed PTB cases, CAD demonstrated a substantially higher DYD (229/273, 83.9%) than radiologists (70/273, 25.6%), although the PPV was much lower (1.70% vs 10.31%). This high-sensitivity performance achieved an 85.5% (79,804/93,319) reduction (only 13,515 instead of 93,319 CXRs) in radiologist workload via selective review of CAD-positive images, without missing any radiologist-identified PTB cases. Furthermore, probability scores greater than 0.75 were a key threshold for identifying high-risk patients with PTB, and these patients were prioritized for radiologist review. Subgroup analysis further revealed that CAD outperformed radiologists in identifying PTB cases across all scenarios, despite some heterogeneity. CAD performance was significantly better in township-level medical facilities (DYD: 86.7%; PPV: 2%) than in county-level hospitals (DYD: 62.5%; PPV: 0.6%). Conclusions: CAD technology is valuable for detecting PTB in primary health care facilities. Combined with a tiered artificial intelligence prescreening with selective human review strategy, this approach effectively alleviates the workload of radiologists in resource-constrained regions, offering a scalable solution for tuberculosis prevention and control.

Pulmonary tuberculosis (PTB) remains a major public health burden, and delayed diagnosis may prolong transmission before treatment. Township-level evidence