Revisiting Computer-Aided Tuberculosis Diagnosis DOI
Yun Liu, Yu-Huan Wu, Shichen Zhang

et al.

IEEE Transactions on Pattern Analysis and Machine Intelligence, Journal Year: 2023, Volume and Issue: 46(4), P. 2316 - 2332

Published: Nov. 7, 2023

Tuberculosis (TB) is a major global health threat, causing millions of deaths annually. Although early diagnosis and treatment can greatly improve the chances survival, it remains challenge, especially in developing countries. Recently, computer-aided tuberculosis (CTD) using deep learning has shown promise, but progress hindered by limited training data. To address this, we establish large-scale dataset, namely X-ray (TBX11 K) which contains 11 200 chest (CXR) images with corresponding bounding box annotations for TB areas. This dataset enables sophisticated detectors high-quality CTD. Furthermore, propose strong baseline, SymFormer, simultaneous CXR image classification infection area detection. SymFormer incorporates Symmetric Search Attention (SymAttention) to tackle bilateral symmetry property discriminative features. Since may not strictly adhere property, also Positional Encoding (SPE) facilitate SymAttention through feature recalibration. promote future research on CTD, build benchmark introducing evaluation metrics, evaluating baseline models reformed from existing detectors, running an online challenge. Experiments show that achieves state-of-the-art performance TBX11 K dataset.

Language: Английский

Age-period-cohort study of active pulmonary tuberculosis in eastern China: analysis of 15-year surveillance data DOI Creative Commons
Dan Luo, Fengying Wang, Songhua Chen

et al.

BMC Public Health, Journal Year: 2025, Volume and Issue: 25(1)

Published: Feb. 17, 2025

Tuberculosis remains a major public health challenge worldwide. This study aimed to determine the long-term trends in notification rate of tuberculosis Zhejiang Province, and assess potential independent risks associated with age, time period, birth cohort. Data on all pulmonary (PTB) cases Province from 2008 2022 were collected Information Management System. Long-term quantitatively decomposed variables determined using joinpoint regression model age-period-cohort model. Between 2022, total 323,745 PTB notified Province. Overall, declined, notable turning points 2010 2019. Net drift analysis revealed an average annual decrease 5.7% (95% CI: -6.8%, -4.6%; P < 0.01), faster declines among males than females. Age effects showed peaks rates people aged 15–29 65–74. Period peaked during 2008–2012 (RR = 1.4; 95% 1.3, 1.5; 0.01). Cohort indicated decreasing later cohorts, highest risk observed 1928–1932 cohort 11.0, 7.2, 16.8; 0.01) relative protective cohorts born after 1978–1982. Notification declined consistently across various age groups Young individuals older 65–74 identified as high-risk requiring active intervention. Additionally, pre-1978 had relatively higher PTB. These findings provided valuable insights into characteristics patients aiding relevant authorities formulating appropriate policies implementing targeted preventive control measures.

Language: Английский

Citations

0

Oral health conditions and unmet need among Chinese male adults: a national oral health survey DOI Creative Commons
Hui Li, Xiaohui Wen, Shiting Li

et al.

BMC Public Health, Journal Year: 2025, Volume and Issue: 25(1)

Published: March 3, 2025

Based on the present global burden of oral diseases, unmet dental needs affect a more significant population worldwide. It is characterized by need for care but receiving delayed or no care. Contributing factors include lack knowledge about health, its consequences, and availability services. We to find out scale problem Chinese male adults. Therefore, aim was determine relationship between presence conditions quality life-related health. A total 46,439 adults were recruited study August 2022 October 2023. The questionnaire administered using standard procedures. This sought gather various types information, including basic demographic details, health status, specific needs. To gain insight into frequency in our population, thorough clinical examination performed. For data interpretation evaluation, descriptive statistics Chi square tests utilized, with level 5%. results indicated that one most prevalent surface calculus, which noted 22,263 (47.94%) participants. rate bleeding gums 6289 (13.54%), range diseases such as ulcer (1488; 3.2%), trauma (1342; 2.89%), gingivitis (814; 1.75%), pericoronitis (786; 1.69%), cheilitis (527; 1.13%). reported 22.41% prevalence caries, an additional 15.59% participants having untreated caries. portion (54.62%) expressed strong healthcare treatment, 79.11% self-reporting their regular checkups. Furthermore, 81.09% desire receive guidance. Approximately one-fourth had revealing age, occupation location influenced Most checkups, support, stressing importance accessibility By specifically targeting age groups addressing common conditions, can be enhanced. Planning strategies should consider maximize impact.

Language: Английский

Citations

0

Analysis of reported incidence of chronic hepatitis B in Chaoyang District, Beijing during 2005–2022 with age-period-cohort model DOI Creative Commons
Qian Li, Xingquan Zhao, Meng Han

et al.

Frontiers in Public Health, Journal Year: 2025, Volume and Issue: 13

Published: March 13, 2025

Objective The study aimed to describe the trend of chronic hepatitis B among population in Chaoyang District from 2005 2022 and explore effects age, period, cohort factors on incidence risk. Methods Incidence rates were collected national infectious disease reporting information management system. Descriptive epidemiological methods employed characterize 2022. Trends analyzed using an age-period-cohort model. estimated Intrinsic Estimator operator. Model goodness fit was assessed by introducing variance, AIC, BIC, comparing established model with conventional two-factor models. Results From 2022, reported showed a gradual decrease followed stabilization, declining 49.58/100,000 8.58/100,000 overall, 63.36/100,000 11.69/100,000 males, 35.15/100,000 5.55/100,000 females. age effect coefficient for males increased initially then decreased whereas females, it age. period risk both genders over time. exhibited pattern decrease, increase, generally decreased. APC constructed intrinsic estimator demonstrated best fit, as indicated lower BIC compared Conclusion declined stabilizing after 2013. We found distinct incidence, higher risks observed aged 25–29 75–79, females born 1966–1970. These findings highlight importance targeted surveillance, expanded screening, improved diagnosis treatment reduce long-term burden.

Language: Английский

Citations

0

Experience of implementing metagenomic next-generation sequencing in patients with suspected pulmonary infection in clinical practice DOI Creative Commons
Yuting Lai,

B Y Chen,

Sida Chen

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 20, 2025

Pulmonary infections remain a leading cause of infectious disease-related hospitalizations. Metagenomic next-generation sequencing (mNGS) has emerged as promising diagnostic tool for identifying pathogens responsible pulmonary infections. However, implementing mNGS in clinical practice presents several challenges. We conducted retrospective analysis 97 patients with suspected who were admitted to our hospital and underwent alongside conventional microbiologic tests (CMT) over the past three years. compared efficacy versus CMT assessed applications challenges associated managing detected 63.9% cases, outperforming (27.8%) showing notable improvements Mycobacterium, fungal species, rare pathogens. Antibiotic regimens adjusted 77.4% positive results, improvement observed 93.5%. Of 138 microbial strains initially identified by possible pathogens, 65 (47.1%) reclassified colonizing organisms upon further evaluation, including bacteria fungi commonly Notably, one patient was diagnosed aspiration pneumonia due oral anaerobes, which had categorized normal flora. In conclusion, serves valuable approach infections, enhancing etiologic precision informing management. Nevertheless, comprehensive interpretation mNGS-identified microorganisms is essential achieve accurate diagnosis.

Language: Английский

Citations

0

An Integrated Mycobacterial CT Imaging Dataset with Multispecies Information DOI Creative Commons
Zhilin Han, Yuyang Zhang, Wenlong Ding

et al.

Scientific Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: March 29, 2025

The increasing global incidence of nontuberculous mycobacterial (NTM) pulmonary disease highlights the need for rapid diagnostic methods to guide timely treatment and prevent antibiotic misuse. While bacterial culture remains gold standard diagnosis, its extended turnaround time compromises clinical decision-making. Computed tomography (CT), with high sensitivity lung lesions imaging capabilities, has emerged as a critical tool. AI-assisted CT interpretation shows particular promise improving NTM detection, yet progress been hindered by limited datasets due rarity. We address this gap introducing first comprehensive dataset combining 430 871 tuberculosis cases, supplemented parameters including demographics, symptoms, species data. This resource aims catalyze AI algorithm development differential ultimately enhancing precision in management through advanced machine learning applications.

Language: Английский

Citations

0

Global, regional and national disparities and temporal trends of common autoimmune disease burdens among children and adolescents from 1990 to 2019 DOI Creative Commons
Chen Chen, Fan Yang, Paul Lodder

et al.

BMJ Global Health, Journal Year: 2025, Volume and Issue: 10(4), P. e017187 - e017187

Published: April 1, 2025

Introduction Previous evidence lacked a thorough review of the disparities autoimmune diseases (AD) burdens among countries and regions, which led to an insufficient basis for developing country-specific developmental level relevant preventive measures. This study aimed analyse trends global, regional national burden common ADs in children adolescents from 1990 2019 investigate associations between specific varied country indexes. Methods All data four major were obtained Global Burden Diseases Study 2019. Age period-cohort modelling was conducted disentangle age, period birth cohort effects on AD incidence Local regression smoothing models used fit correlation sociodemographic index (SDI). Pearson’s country-level risk factors disease burden. Results A global increase observed 1.57 million 1.63 0–24 age group. The age-standardised rate overall showed substantial variation with highest high SDI regions. distributions significantly, especially countries. Relative expected associated SDI, distribution by regions depending ADs. Countries higher levels socioeconomic development, better quality life easier access healthcare system lower Conclusions patterns considerably according time generational cohort, across world Incidences significantly correlated indexes involving risks environment, human rights health safety life.

Language: Английский

Citations

0

Coexistent Pulmonary Tuberculosis and Lung Cancer: An Analysis of Incidence Trends, Financial Burdens and Influencing Factors DOI Creative Commons
Fei Qi,

Hongjie Yang,

Yi Han

et al.

Cancer Innovation, Journal Year: 2025, Volume and Issue: 4(3)

Published: April 7, 2025

ABSTRACT Background Tuberculosis (TB) and lung cancer (LC) are both major global health threats. However, coexistent pulmonary TB LC (TBLC) is a unique condition for which incidence trends risk factors have not been fully defined. Methods We retrospectively reviewed the medical records of patients with TBLC alone between 2010 2022 at Beijing Chest Hospital, standard authority diagnosis treatment in China. The cumulative rate (CIR) was calculated as number new cases/number cases per 100,000 annually. comparative ratio (IRR) estimated to be patients/TB general population. Logistic regression used explore TBLC. Results CIR has rapidly increased each year since 2014 reached 7027 2022. Patients had higher developing active than population (IRR = 25.21, 95% confidence interval [CI]: 21.54–29.89). Medical expenditure patient 100.60 thousand yuan those 105.60 ( p 0.687). were older (63.61 ± 10.46 vs. 61.08 10.77, < 0.001) male‐to‐female (2.82 1.59, 0.044) alone. A tendency earlier disease onset observed rather majority (44.92%) lesions located upper lobes proportion squamous cell carcinomas (32.24% 27.49%, 0.002). also presented more aggressively, lymph node involvement distant metastases. Multivariate analysis revealed that age, male sex, mediastinal invasion, lung/bone metastases, anemia, hypoalbuminemia, malnutrition, fibrosis, chronic obstructive Conclusions There rise concomitant increase its financial burden China deserves awareness attention.

Language: Английский

Citations

0

Spatiotemporal Patterns of Leukemia Subtype Burden in Asia: A 32-Year Comprehensive Analysis Using the Global Burden of Disease Database DOI
Ning Jiang, R. Yao, Huifang Zhou

et al.

Published: Jan. 1, 2025

Language: Английский

Citations

0

Association of urbanization-related factors with tuberculosis incidence among 1992 counties in China from 2005 to 2019: a nationwide observational study DOI Creative Commons

Yaping Wang,

Xiaoqiu Liu,

Yu‐Hong Li

et al.

Infectious Diseases of Poverty, Journal Year: 2025, Volume and Issue: 14(1)

Published: April 25, 2025

Abstract Background Most high tuberculosis (TB) burden countries are in low- and middle-income regions undergoing rapid urbanization. We aimed to assess the association between urbanization factors TB incidence China. Methods evaluated at county level China from 2005 2019 using a composite index integrating population density, gross domestic product (GDP, per capita), hospital beds 1000 population, nighttime light (NTL), normalized difference vegetation (NDVI). The annual rate number of cases were obtained national Tuberculosis Information Management System, maintained by Chinese Center for Disease Control Prevention. Fixed-effects models used examine incidence. A subgroup analysis was performed dividing counties into four regions: northeast, eastern, central, western. Results total 1992 included this study. Overall, scores associated with reduced (β = − 0.0114, P < 0.001), corresponding 1.1% reduction unit score. Quadratic presented U-shaped relationship an inflection point 52.94 units. For each indicator urbanization, density positively incidence, ratios 11.384 [95% confidence interval ( CI ): 9.337 13.881], 1.015 (95% : 1.011 1.019), respectively, while GDP, NTL, NDVI exhibited protective effects. Central displayed increase trend that score linked 1.8% rise Conclusions Urbanization-related factors, including NTL NDVI, inversely China’s contrasting results highlighted region-specific challenges. Therefore, governments developing should adopt integrated approaches promote both economic growth sustainable development environment during optimize control efforts. Graphical abstract

Language: Английский

Citations

0

Multimodal Radiomics Integrating Deep Learning and Clinical Features for Diagnosing Multidrug-Resistant Tuberculosis in HIV/AIDS Patients DOI Creative Commons

Chang Seob Song,

Ai-Chun Huang,

Chunyan Zhao

et al.

Journal of Global Antimicrobial Resistance, Journal Year: 2025, Volume and Issue: unknown

Published: May 1, 2025

This study aimed to develop and validate a predictive model based on multimodal data, including clinical features, radiomics deep learning distinguish multidrug-resistant tuberculosis (MDR-TB) in HIV/AIDS patients, thereby improving diagnostic accuracy. A retrospective cohort of patients with drug-sensitive (n=164) MDR-TB (n=63) admitted the Fourth People's Hospital Nanning between January 2016 July 2024 was included. The dataset randomly divided into training validation sets at 7:3 ratio. constructed by integrating model, 2.5D multi-instance (MIL) approach. Key predictors-platelet count C-reactive protein-were identified through univariate multivariate logistic regression analysis. integrated achieved highest performance both set (AUC=0.943 0.899, respectively), significantly outperforming individual models. Grad-CAM effectively localized key image regions influencing decision-making, while nomogram quantified contribution weights each predictor, enhancing transparency. Hosmer-Lemeshow (HL) test confirmed good calibration, decision curve analysis (DCA) demonstrated optimal net benefit model. developed this improved efficacy combining clinical, radiomics, providing reliable tool for individualized precision diagnosis treatment.

Language: Английский

Citations

0