Development and application of a deep learning-based comprehensive early diagnostic model for chronic obstructive pulmonary disease.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a frequently diagnosed yet treatable condition, provided it is identified early and managed effectively. This study aims to develop an advanced COPD diagnostic model by integrating deep learning and radiomics features. METHODS: We utilized a dataset comprising CT images from 2,983 participants, of which 2,317 participants also provided epidemiological data through questionnaires. Deep learning features were extracted using a Variational Autoencoder, and radiomics features were obtained using the PyRadiomics package. Multi-Layer Perceptrons were used to construct models based on deep learning and radiomics features independently, as well as a fusion model integrating both. Subsequently, epidemiological questionnaire data were incorporated to establish a more comprehensive model. The diagnostic performance of standalone models, the fusion model and the comprehensive model was evaluated and compared using metrics including accuracy, precision, recall, F1-score, Brier score, receiver operating characteristic curves, and area under the curve (AUC). RESULTS: The fusion model exhibited outstanding performance with an AUC of 0.952, surpassing the standalone models based solely on deep learning features (AUC = 0.844) or radiomics features (AUC = 0.944). Notably, the comprehensive model, incorporating deep learning features, radiomics features, and questionnaire variables demonstrated the highest diagnostic performance among all models, yielding an AUC of 0.971. CONCLUSION: We developed and implemented a data fusion strategy to construct a state-of-the-art COPD diagnostic model integrating deep learning features, radiomics features, and questionnaire variables. Our data fusion strategy proved effective, and the model can be easily deployed in clinical settings. TRIAL REGISTRATION: Not applicable. This study is NOT a clinical trial, it does not report the results of a health care intervention on human participants.

Global, regional, and national burden and quality of care of multiple myeloma, 1990-2019.

Background: Multiple myeloma (MM) is the second most common haematologic malignancy, presenting a great disease burden on the general population; however, the quality of care of MM is overlooked. We therefore assessed gains and disparity in quality of care worldwide from 1990 to 2019 based on a novel summary indicator - the quality of care index (QCI) - and examined its potential for improvement. Methods: Using the Global Burden of Disease 2019 data set, we calculated the QCI of MM for 195 countries and territories. We used the principal component analysis to extract the first principal component of ratios with the combinations of mortality to incidence, prevalence to incidence, disability-adjusted life years to prevalence, and years of life lost to years lived with disability as QCI. We also conducted a series of descriptive and comparative analyses of QCI disparities with age, gender, period, geographies, and sociodemographic development, and compared the QCI among countries with similar socio-demographic index (SDI) through frontier analysis. Results: The age-standardised rates of MM were 1.92 (95% uncertainty interval (UI) = 1.68, 2.12) in incidence and 1.42 (95% UI = 1.24, 1.52) in deaths per 100 000 population in 2019, and were predicted to increase in the future. The global age-standardised QCI increased from 51.31 in 1990 to 64.28 in 2019. In 2019, New Zealand had the highest QCI at 99.29 and the Central African Republic had the lowest QCI at 10.74. The gender disparity of QCI was reduced over the years, with the largest being observed in the sub-Saharan region. Regarding age, QCI maintained a decreasing trend in patients aged >60 in SDI quintiles. Generally, QCI improved with the SDI increase. Results of frontier analysis suggested that there is a potential to improve the quality of care across all levels of development spectrum. Conclusions: Quality of care of MM improved during the past three decades, yet disparities in MM care remain across different countries, age groups, and genders. It is crucial to establish local objectives aimed at enhancing MM care and closing the gap in health care inequality.

Association of fat mass and fat-free mass with all-cause and cause-specific mortality in Asian individuals: A prospective cohort study.

OBJECTIVE: The study's objective was to investigate the association of fat mass index (FMI) and fat-free mass index (FFMI) with all-cause mortality and cause-specific mortality in the Chinese population. METHODS: A total of 422,430 participants (48.1% men and 51.9% women) from the Taiwan MJ Cohort with an average follow-up of 9 years were included. RESULTS: The lowest (Q1) and highest (Q5) quintiles of FMI and FFMI were associated with increased all-cause mortality. Compared with those in the third quintile (Q3) group of FMI, participants in Q1 and Q5 groups of FMI had hazard ratios and 95% CI of 1.32 (1.24-1.40) and 1.13 (1.06-1.20), respectively. Similarly, compared with those in Q3 group of FFMI, people in Q1 and Q5 groups of FFMI had hazard ratios of 1.14 (1.06-1.23) and 1.16 (1.10-1.23), respectively. In the restricted cubic spline models, both FMI and FFMI showed a J-shaped association with all-cause mortality. People in Q5 group of FFMI had a hazard ratio of 0.72 (0.58-0.89) for respiratory disease. CONCLUSIONS: The mortality risk increases in those with excessively high or low FMI and FFMI, yet the associations between FMI, FFMI, and the risk of death varied across subgroups and causes of death.

6-Year trajectory of fasting plasma glucose (FPG) and mortality risk among individuals with normal FPG at baseline: a prospective cohort study.

BACKGROUND: Higher fasting plasma glucose (FPG) levels were associated with an increased risk of all-cause mortality; however, the associations between long-term FPG trajectory groups and mortality were unclear, especially among individuals with a normal FPG level at the beginning. The aims of this study were to examine the associations of FPG trajectories with the risk of mortality and identify modifiable lifestyle factors related to these trajectories. METHODS: We enrolled 50,919 individuals aged ≥ 20 years old, who were free of diabetes at baseline, in the prospective MJ cohort. All participants completed at least four FPG measurements within 6 years after enrollment and were followed until December 2011. FPG trajectories were identified by group-based trajectory modeling. We used Cox proportional hazards models to examine the associations of FPG trajectories with mortality, adjusting for age, sex, marital status, education level, occupation, smoking, drinking, physical activity, body mass index, baseline FPG, hypertension, dyslipidemia, cardiovascular disease or stroke, and cancer. Associations between baseline lifestyle factors and FPG trajectories were evaluated using multinomial logistic regression. RESULTS: We identified three FPG trajectories as stable (n = 32,481), low-increasing (n = 17,164), and high-increasing (n = 1274). Compared to the stable group, both the low-increasing and high-increasing groups had higher risks of all-cause mortality (hazard ratio (HR) = 1.18 (95% CI 0.99-1.40) and 1.52 (95% CI 1.09-2.13), respectively), especially among those with hypertension. Compared to participants with 0 to 1 healthy lifestyle factor, those with 6 healthy lifestyle factors were more likely to be in the stable group (ORlow-increasing = 0.61, 95% CI 0.51-0.73; ORhigh-increasing = 0.20, 95% CI 0.13-0.32). CONCLUSIONS: Individuals with longitudinally increasing FPG had a higher risk of mortality even if they had a normal FPG at baseline. Adopting healthy lifestyles may prevent individuals from transitioning into increasing trajectories.