With the increasing availability of clinical and biomedical big data, machine learning is being widely used in scientific research and academic papers. It integrates various types of information to predict individual health outcomes. However, deficiencies in reporting key information have gradually emerged. These include issues like data bias, model fairness across different groups, and problems with data quality and applicability. Maintaining predictive accuracy and interpretability in real-world clinical settings is also a challenge. This increases the complexity of safely and effectively applying predictive models to clinical practice. To address these problems, TRIPOD+AI (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis+artificial intelligence) introduces a reporting standard for machine learning models. It is based on TRIPOD and aims to improve transparency, reproducibility, and health equity. These improvements enhance the quality of machine learning model applications. Currently, research on prediction models based on machine learning is rapidly increasing. To help domestic readers better understand and apply TRIPOD+AI, we provide examples and interpretations. We hope this will support researchers in improving the quality of their reports.
Objective To estimate the population attributable disease burden (PAD) of low bone mineral density (LBMD) related fractures among Chinese people over 50 years old from 1990 to 2023, using data from the Global Burden of Disease Study 2023 (GBD 2023), and to provide evidence for prevention strategies and health resource allocation. Methods Based on the GBD 2023, the LBMD summary exposure values (SEV), fracture incidence, years lived with disability (YLDs), and LBMD-related falls YLDs of Chinese people over 50 years old from 1990 to 2023 were extracted. PAD was calculated with population attributable fraction (PAF), and an entropy-weight method was applied to evaluate the contribution of individual fracture sites. Temporal trends and sex differences were examined with Joinpoint regression. Results From 1990 to 2023, the age-standardized SEV of LBMD in people over 50 years old showed an overall decline [average annual percent change (AAPC)=−0.564%]. Age-standardized fracture incidence, fracture YLDs rate, and LBMD-related falls YLDs rate all exhibited W-shaped upward trends (AAPC=1.045%, 0.296%, and 0.724%, respectively). PAF-based estimates indicated that LBMD-attributable fracture incidence likewise increased in a “W-shaped” manner (AAPC=0.558%), whereas the corresponding YLDs rate showed an overall W-shaped decline (AAPC=−0.193%). In international comparison, China and the global average displayed broadly concordant directions of change, with greater volatility in China and a progressive narrowing of the gap after 2015. Regarding sex differences, fracture YLDs rates were consistently higher in the males, whereas the other burden indicators were higher in the females; the temporal patterns were similar in both sexes. Entropy weight method identified hip fractures as contributing most to incidence (weight 0.133), and pelvic fractures as the largest contributor to YLDs rate (weight 0.115). ConclusionSince 1990, the LBMD attributable fracture burden in China’s older population has risen, with female and hip or pelvic fractures bearing the heaviest load. Strengthened osteoporosis screening, improved insurance coverage, and targeted health education are urgently needed to curb further increases in disease burden.
ObjectiveTo analyze and compare the incidence, mortality, temporal trends, and cancer spectrum differences between China and the United States (US), providing theoretical support for cancer prevention and control in China. MethodsAge standardized incidence rate (ASIR), age standardized mortality rate (ASMR), and cancer site composition were extracted from GLOBOCAN, Cancer Statistics 2025, the China Cancer Registry Annual Report, and other epidemiological sources. Spatial (urban-rural, sex specific) and temporal distributions were described, and average annual growth rate (AAGR) were calculated. ResultsFrom 2005 onward, China exhibited a modest rise in ASIR, whereas the US showed a decline (AAGR: 0.58 vs –0.42); nevertheless, China’s overall incidence remained lower (2022 ASIR = 201.61/100 000) than that of the US (303.60/100 000). Both countries experienced decreasing ASMR (AAGR: –1.03 vs –1.72). In both nations, male ASIR and ASMR were higher than female. Since 2005, the top three US cancers had remained prostate (men) or breast (women), lung and colorectal cancer. In China, incidences of lung, colorectal, female breast and thyroid cancers had continued to rise, while stomach and liver cancer incidences had declined yet still rank high among men. Urban ASIR in China exceeded rural rates, whereas rural ASMR was higher than urban counterparts. ConclusionsAccelerating population ageing and lifestyle transitions have driven an upward incidence trend in China, accompanied by a shift towards a mixed pattern of traditional and emerging cancer risks. Drawing on US experience, China should intensify tobacco control measures, expand organized screening and early detection programs, implement comprehensive interventions for priority cancers, strengthen primary level capacity and improve treatment access in rural areas, thereby establishing a more effective national cancer prevention and control system.