Gang Li

Gang Li

Auto-generated profile, based on publicly available information

Gang Li

Institute of Information and Statistics Center, Beijing Center for Disease Prevention and Control, Beijing, China

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 February 2023 in Journal of Medical Internet Research
Verified
Journal Article
Open access
LYLiuyang Yang
Liuyang Yang
GLGang Li
Gang Li
JYJin Yang
Jin Yang
TZTing Zhang
Ting Zhang
JDJing Du
Jing Du
Weizhong YangWeizhong Yang
Weizhong Yang
Background: In megacities, there is an urgent need to establish more sensitive forecasting and early warning methods for acute respiratory infectious diseases. Existing prediction and early warning models for influenza and other acute respiratory infectious diseases have limitations and therefore there is room for improvement. Objective: The aim of this study was to explore a new and better-performing deep-learning model to predict influenza trends from multisource heterogeneous data in a megacity. Methods: We collected multisource heterogeneous data from the 26th week of 2012 to the 25th week of 2019, including influenza-like illness (ILI) cases and virological surveillance, data of climate and demography, and search engines data. To avoid collinearity, we selected the best predictor according to the weight and correlation of each factor. We established a new multiattention-long short-term memory (LSTM) deep-learning model (MAL model), which was used to predict the percentage of ILI (ILI%) cases and the product of ILI% and the influenza-positive rate (ILI%×positive%), respectively. We also combined the data in different forms and added several machine-learning and deep-learning models commonly used in the past to predict influenza trends for comparison. The R2 value, explained variance scores, mean absolute error, and mean square error were used to evaluate the quality of the models. Results: The highest correlation coefficients were found for the Baidu search data for ILI% and for air quality for ILI%×positive%. We first used the MAL model to calculate the ILI%, and then combined ILI% with climate, demographic, and Baidu data in different forms. The ILI%+climate+demography+Baidu model had the best prediction effect, with the explained variance score reaching 0.78, R2 reaching 0.76, mean absolute error of 0.08, and mean squared error of 0.01. Similarly, we used the MAL model to calculate the ILI%×positive% and combined this prediction with different data forms. The ILI%×positive%+climate+demography+Baidu model had the best prediction effect, with an explained variance score reaching 0.74, R2 reaching 0.70, mean absolute error of 0.02, and mean squared error of 0.02. Comparisons with random forest, extreme gradient boosting, LSTM, and gated current unit models showed that the MAL model had the best prediction effect. Conclusions: The newly established MAL model outperformed existing models. Natural factors and search engine query data were more helpful in forecasting ILI patterns in megacities. With more timely and effective prediction of influenza and other respiratory infectious diseases and the epidemic intensity, early and better preparedness can be achieved to reduce the health damage to the population.

23 April 2025 in JACC
Verified
Journal Article
Closed access
SMSailimai Man
Sailimai Man
YZYining Zu
Yining Zu
XYXiaochen Yang
Xiaochen Yang
YDYuhan Deng
Yuhan Deng
DSDan Shen
Dan Shen
LLLiming Li
Liming Li

27 September 2018 in Lipids in Health and Disease
Verified
Journal Article
Open access
PDF file
KFKai Fang
Kai Fang
HLHang Li
Hang Li
AMAijuan Ma
Aijuan Ma
JDJing Dong
Jing Dong
JXJin Xie
Jin Xie
ZDZhong Dong
Zhong Dong
Obesity is known to be a risk factor to a variety of chronic diseases. Weight misperception has an impact on weight-loss attitude and behavior. We aimed to investigate factors associated with weight underestimation, and to assess the effect of hypertension, diabetes and dyslipidemia awareness on weight underestimation and weight management for overweight and obese adults. Data was obtained from the 2011 Beijing Non-communicable disease and risk factors Surveillance (BJNCDRS). A total of 19,932 participants with measures of weight and height were included in the analysis. Self-perception of weight was obtained by asking, “How do you describe your weight?”, and the question for weight management was “Are you taking any actions to control your body weight?”. Multiple logistic regression was used to investigate factors related to weight underestimation. For the underweight, normal weight, and overweight/obese categories, more than half of the participants perceived their weight accurately (63.6, 53.8, 66.2%, respectively). For overweight and obese adults, older age, male, rural residence, lower level of education, lower level of income, absence of hypertension, presence of diabetes and absence of dyslipidemia positively associated with weight underestimation, and awareness of having hypertension and dyslipidemia were negatively associated with weight underestimation (Adjusted OR(95%CI) were 0.70(0.61~ 0.79) and 0.71(0.62~ 0.80), respectively). Awareness of having hypertension and dyslipidemia were significantly associated with weight management (Adjusted OR (95%CI) were 1.42(1.25~ 1.62) and 1.53(1.36~ 1.72), respectively). There was no significant association between awareness of diabetes and weight underestimation(P > 0.05) or weight management(P > 0.05). More than half of the participants perceived their weight accurately. For overweight/obese population, awareness of having hypertension and dyslipidemia could improve weight perception and weight management, whereas awareness of having diabetes might not.

10 January 2026 in The Lancet Planetary Health
Verified
Journal Article
Open access
XMXia Meng
Xia Meng
YZYuchang Zhou
Yuchang Zhou
SSSu Shi
Su Shi
SWShuxiao Wang
Shuxiao Wang
MZMaryam Zaid
Maryam Zaid
MZMaigeng Zhou
Maigeng Zhou
The hazardous effects of fine particulate matter (PM2·5) could vary by source. However, evidence regarding the long-term effects of source-specific PM2·5 on mortality is scarce, particularly in countries with high levels of pollution. We aimed to assess the associations between long-term exposure to source-specific PM2·5 and mortality in China, a low-income and middle-income country with high concentrations of PM2·5. This nationwide cohort study used data from the 2010-11 wave of the China Chronic Disease and Risk Factor Surveillance (CCDRFS) project, which surveyed a geographically and socioeconomically representative sample of participants aged 18 years and older across mainland China. Participants were followed up through mortality information from the national Disease Surveillance Point system. Source-specific PM2·5 concentrations were estimated by combining high-resolution concentration predictions generated by a random forest model with estimates of the proportions contributed by each of six PM2·5 sources-industry, energy, transportation, residential, agriculture, and other-derived from the Community Multiscale Air Quality model. A Cox proportional hazards model, in which exposures were treated as time-varying covariates, was used to evaluate the associations between long-term exposure to source-specific PM2·5 and mortality from total non-accidental causes, cardiopulmonary disease, and lung cancer. 98 058 participants were followed from the date of their CCDRFS survey, conducted between January, 2010, and December, 2011, until death or Dec 31, 2020, whichever was earliest. Of these 98 058 participants, 96 955 were included in the analyses. The mean age was 46·5 years (SD 15·0); 52 631 (54·3%) participants were female and 44 324 (45·7%) were male. Almost all anthropogenic sources of PM2·5 were significantly associated with mortality. For each IQR increase, PM2·5 from industry (hazard ratio 1·079 [95% CI 1·024-1·137]), transportation (1·076 [1·034-1·119]), and residential (1·075 [1·029-1·124]) sources had the strongest associations with mortality from total non-accidental causes. For mortality from cardiopulmonary diseases, PM2·5 from transportation (1·106 [1·049-1·165]), industrial (1·085 [1·011-1·163]), and residential (1·070 [1·012-1·130]) sources had the strongest associations, and for mortality from lung cancer, the strongest associations were observed for PM2·5 from agricultural (1·256 [1·089-1·449]), industrial (1·235 [1·049-1·454]), and transportation (1·204 [1·048-1·384]) sources. We show that the mortality risks of PM2·5 exposure can vary according to the source. These findings could contribute evidence for the development of health-oriented policies to enhance the effectiveness of clean air initiatives. National Key Research and Development Program of China, National Natural Science Foundation of China, Shanghai Municipal Science and Technology Major Project, and Shanghai International Science and Technology Partnership Project.

9 February 2015 in PLOS ONE
Verified
Journal Article
Open access
PDF file
GLGang Li
Gang Li
HHHuanhuan Hu
Huanhuan Hu
ZDZhong Dong
Zhong Dong
JXJin Xie
Jin Xie
YZYing Zhou
Ying Zhou
This study aimed to compare hypertension trends in the urban and suburban population, and to examine the use of several self-care behaviors among patients who were aware of their hypertension. We examined the data from three cross-sectional adult populations obtained in 2005, 2008, and 2011, in Beijing. Our analyses indicated that from 2005 to 2011 the standardized rate of hypertension increased from 31.9% to 36.0% (P <0.001) among urban adults, and was relatively stable (40.8% -40.2%) among suburban adults (P = 0.02). About 10% of the patients reported having taken measures to control their weight for hypertension management. As compared to the other patients, the female patients in the urban areas reported the highest rate of regular BP measurement (52.6%). In addition, the patients who reported taking medication regularly increased among the males and females. Most of the women reported nonsmoking (≥95%) and alcohol abstinence (≥90%). The trend of nonsmoking decreased among the urban males. In contrast, the prevalence of nonsmoking increased among the suburban males, though the trend was not statistically significant (P = 0.055). Further, the patient-reported alcohol abstinence was found to exhibit a decreasing trend among the males. We observed an increase in the hypertension prevalence from 2005 to 2011. The rates remained higher for suburban adults than for urban adults. Females generally had better self-care ability as compared to male patients. Further research is needed to promote self-care behaviors in hypertensive patients, especially for male patients.

8 February 2026 in Internal and Emergency Medicine
Verified
Journal Article
Closed access
PDF file
XDXinjiang Dong
Xinjiang Dong
JWJia Wang
Jia Wang
JDJing Du
Jing Du
BWBeibei Wang
Beibei Wang
GLGang Li
Gang Li
TZTong Zou
Tong Zou