HSHui Sun

Hui Sun

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Hui Sun has a total of 28 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Machine Learning–Based Risk Factor Analysis and Prediction Model Construction for the Occurrence of Chronic Heart Failure: Health Ecologic Study

    by Qian Xu, Xue Cai, Ruicong Yu, Yueyue Zheng, Guanjie Chen, Hui Sun, Tianyun Gao, Cuirong Xu, Jing Sun

    Background: Chronic heart failure (CHF) is a serious threat to human health, with high morbidity and mortality rates, imposing a heavy burden on the health care system and society. With the abundance of medical data and the rapid development of machine learning (ML) technologies, new opportunities are provided for in-depth investigation of the mechanisms of CHF and the construction of predictive models. The introduction of health ecology research methodology enables a comprehensive dissection of CHF risk factors from a wider range of environmental, social, and individual factors. This not only helps to identify high-risk groups at an early stage but also provides a scientific basis for the development of precise prevention and intervention strategies. Objective: This study aims to use ML to construct a predictive model of the risk of occurrence of CHF and analyze the risk of CHF from a health ecology perspective. Methods: This study sourced data from the Jackson Heart Study database. Stringent data preprocessing procedures were implemented, which included meticulous management of missing values and the standardization of data. Principal component analysis and random forest (RF) were used as feature selection techniques. Subsequently, several ML models, namely decision tree, RF, extreme gradient boosting, adaptive boosting (AdaBoost), support vector machine, naive Bayes model, multilayer perceptron, and bootstrap forest, were constructed, and their performance was evaluated. The effectiveness of the models was validated through internal validation using a 10-fold cross-validation approach on the training and validation sets. In addition, the performance metrics of each model, including accuracy, precision, sensitivity, F1-score, and area under the curve (AUC), were compared. After selecting the best model, we used hyperparameter optimization to construct a better model. Results: RF-selected features (21 in total) had an average root mean square error of 0.30, outperforming principal component analysis. Synthetic Minority Oversampling Technique and Edited Nearest Neighbors showed better accuracy in data balancing. The AdaBoost model was most effective with an AUC of 0.86, accuracy of 75.30%, precision of 0.86, sensitivity of 0.69, and F1-score of 0.76. Validation on the training and validation sets through 10-fold cross-validation gave an AUC of 0.97, an accuracy of 91.27%, a precision of 0.94, a sensitivity of 0.92, and an F1-score of 0.94. After random search processing, the accuracy and AUC of AdaBoost improved. Its accuracy was 77.68% and its AUC was 0.86. Conclusions: This study offered insights into CHF risk prediction. Future research should focus on prospective studies, diverse data, advanced techniques, longitudinal studies, and exploring factor interactions for better CHF prevention and management.


  • Open Access

    Factors Influencing Poststroke Cognitive Dysfunction: Cross-Sectional Analysis

    by Wu Zhou, HaiXia Feng, Hua Tao, Hui Sun, TianTian Zhang, QingXia Wang, Li Zhang

    Background: Poststroke cognitive impairment (PSCI) is a common and debilitating complication that affects stroke survivors, impacting memory, attention, and executive function. Despite its prevalence, the factors contributing to PSCI remain unclear, with limited insights into how demographic and clinical variables influence cognitive outcomes. Objective: This study investigates the incidence of cognitive impairment in patients with stroke and examines key demographic and clinical factors, such as age, gender, and education level, which contribute to cognitive decline. The aim is to provide a deeper understanding of PSCI to inform early intervention strategies for improving patient outcomes. Methods: A cross-sectional study was conducted on 305 patients with ischemic stroke admitted to Zhongda Hospital, Southeast University, from January 2019 to September 2022. Cognitive function was assessed using the Mini-Mental State Examination (MMSE) within 72 hours of hospital admission. Demographic information, including age, gender, and education level, were collected. Statistical analyses were performed using chi-square tests, independent t tests, and multivariate regression to assess the relationship between cognitive function and key variables. Pearson correlation analysis explored associations among age, education, and MMSE scores. Results: Among the 305 patients with stroke, 16.7% (n=51) were diagnosed with cognitive impairment based on MMSE scores. The prevalence of cognitive impairment was slightly higher in males (17.6%, n=159) than females (15.8%, n=146), but this difference was not statistically significant. A strong negative correlation was found between MMSE scores and age (r=–0.32; P<.01), indicating that older patients had lower cognitive function. Education level showed a positive correlation with MMSE scores (r=0.41; P<.01), with patients with higher educational attainment demonstrating better cognitive outcomes. Cognitive function showed a marked decline in patients older than 60 years, particularly in domains such as memory, attention, and language skills. Conclusions: This study confirms that age and education are significant factors in determining cognitive outcomes after stroke. The results align with existing literature showing that cognitive function declines with age, while higher educational attainment serves as a protective factor. The findings suggest that individuals with greater cognitive reserve, often linked to higher education, are better equipped to cope with the impact of brain injury. However, the study’s reliance on MMSE may have limited its ability to detect domain-specific impairments. Future studies should consider using more sensitive cognitive tools, such as the Montreal Cognitive Assessment (MoCA), to provide a more comprehensive evaluation of PSCI. Cognitive impairment is prevalent among stroke survivors, with age and education level being key factors influencing outcomes. These findings underscore the importance of early detection and targeted interventions to mitigate cognitive decline. Further research with larger samples and more sensitive cognitive assessments is needed to fully understand PSCI and improve rehabilitation strategies for patients with stroke.


  • Open Access

    Nutrition Management Miniprograms in WeChat: Evaluation of Functionality and Quality

    by Hui Sun, Yanping Wu, Jia Sun, Wu Zhou, Qian Xu, Dandan Hu

    Background: With the rise in people’s living standards and aging populations, a heightened emphasis has been placed in the field of medical and health care. In recent years, there has been a drastic increase in nutrition management in domestic research circles. The mobile nutritional health management platform based on WeChat miniprograms has been widely used to promote health and self-management and to monitor individual nutritional health status in China. Nevertheless, there has been a lack of comprehensive scientific evaluation regarding the functionality and quality of the diverse range of nutritional miniprograms that have surfaced in the market. Objective: This study aimed to evaluate the functionality and quality of China’s WeChat nutrition management miniprogram by using the User Version of the Mobile Application Rating Scale (uMARS). Methods: This observational study involves quantitative methods. A keyword search for “nutrition,” “diet,” “food,” and “meal” in Chinese or English was conducted on WeChat, and all miniprograms pertaining to these keywords were thoroughly analyzed. Then, basic information including name, registration date, update date, service type, user scores, and functional scores was extracted from January 2017 to November 2023. Rating scores were provided by users based on their experience and satisfaction with the use of the WeChat miniprogram, and functional scores were integrated and summarized for the primary functions of each miniprogram. Moreover, the quality of nutrition management applets was evaluated by 3 researchers independently using the uMARS. Results: Initially, 27 of 891 miniprograms identified were relevant to nutrition management. Among them, 85.2% (23/27) of them offered features for diet management, facilitating recording of daily dietary intake to evaluate nutritional status; 70.4% (19/27) provided resources for nutrition education and classroom instruction; 59.3% (16/27) included functionalities for exercise management, allowing users to record daily physical activity; and only 44.4% (12/27) featured components for weight management. The total quality score on the uMARS ranged 2.85-3.88 (median 3.38, IQR 3.14-3.57). Engagement scores on the uMARS varied from 2.00 to 4.33 (median 3.00, IQR 2.67-3.67). Functional dimension scores ranged from 3.00 to 4.00 (median 3.33, IQR 3.33-3.67), with a lower score of 2.67 and a higher score of 4.33 outside the reference range. Aesthetic dimension scores ranged from 2.33 to 4.67 (median 3.67, IQR 3.33-4.00). Informational dimension scores ranged from 2.33 to 4.67 (median 3.33, IQR 2.67-3.67). Conclusions: Our findings from the uMARS highlight a predominant emphasis on health aspects over nutritional specifications in the app supporting WeChat miniprograms related to nutrition management. The quality of these miniprograms is currently at an average level, with considerable room for functional improvements in the future.


  • Open Access

    Current Status of Outcomes Reported by Patients With Stroke and an Analysis of Influencing Factors: Cross-Sectional Questionnaire Study

    by Jia Sun, Liang Ma, Xiao Miao, Hui Sun, SuSu Zhu, Ran Zhang, LeLe Fan, TingTing Hu

    Background: Stroke is the leading cause of acquired disability and the second leading cause of death worldwide. Its rate of incidence, disability, mortality, and recurrence is high, and the patients experience various symptoms of discomfort, which not only affect their rehabilitation function but also reduce their ability to perform daily activities and their quality of life. Nowadays, with the improvement of China’s medical standards, patients are increasingly attentive to their quality of life and health status. However, diagnostic techniques and effective treatments for patients with stroke are still limited but urgently required. Objective: This study aimed to evaluate the quality of life during hospitalization using a stroke patient-reported outcomes (PROs) scale and additionally to recognize potential factors and risk indicators that may impact recurrent events, facilitating early intervention measures. Methods: This is a registry-based, retrospective observational cross-sectional study on patients with stroke. A convenient sampling method was used to select various indicators of patients. The Stroke-PRO scale was then used to assess patients’ conditions across physical, psychological, social, and therapeutic domains. Multiple linear regression analysis was applied to identify factors influencing stroke PROs, while correlation analysis was conducted to explore the relationship between these outcomes and blood lipid levels. Results: The mean Stroke-PRO score in this study was 4.09 (SD 0.29) points. By multiple linear regression analysis, residence, occupation, physical exercise, Barthel index, Braden scale, National Institutes of Health Stroke Scale scores at admission, and stroke type were the risk factors for reported outcomes of patients with stroke (P<.05). Correlation analysis showed that serum triglyceride, total cholesterol, and low-density lipoprotein were negatively correlated with Stroke-PRO scores in patients with stroke (P<.05), while high-density lipoprotein was positively correlated with patients with stroke (P<.05). The 95% CI was –0.31 to –0.03 for triglyceride, 0.17-0.44 for high-density lipoprotein, –0.29 to –0.01 for cholesterol, –0.30 to –0.02 for low-density lipoprotein, and –0.12 to 0.16 for blood glucose. Conclusions: Patients with stroke have a low level of health, and their reported outcomes need to be improved. Accordingly, nursing staff should pay attention to the quality of life and blood lipid indexes of patients with stroke, actively assess their actual health status, and take early intervention measures to promote their recovery.


  • Open Access

    Machine Learning–Based Risk Factor Analysis and Prediction Model Construction for the Occurrence of Chronic Heart Failure: Health Ecologic Study (Preprint)

    by Qian Xu, Xue Cai, Ruicong Yu, Yueyue Zheng, Guanjie Chen, Hui Sun, Tianyun Gao, Cuirong Xu, Jing Sun

    BACKGROUND Chronic heart failure (CHF) is a serious threat to human health, with high morbidity and mortality rates, imposing a heavy burden on the health care system and society. With the abundance of medical data and the rapid development of machine learning (ML) technologies, new opportunities are provided for in-depth investigation of the mechanisms of CHF and the construction of predictive models. The introduction of health ecology research methodology enables a comprehensive dissection of CHF risk factors from a wider range of environmental, social, and individual factors. This not only helps to identify high-risk groups at an early stage but also provides a scientific basis for the development of precise prevention and intervention strategies. OBJECTIVE This study aims to use ML to construct a predictive model of the risk of occurrence of CHF and analyze the risk of CHF from a health ecology perspective. METHODS This study sourced data from the Jackson Heart Study database. Stringent data preprocessing procedures were implemented, which included meticulous management of missing values and the standardization of data. Principal component analysis and random forest (RF) were used as feature selection techniques. Subsequently, several ML models, namely decision tree, RF, extreme gradient boosting, adaptive boosting (AdaBoost), support vector machine, naive Bayes model, multilayer perceptron, and bootstrap forest, were constructed, and their performance was evaluated. The effectiveness of the models was validated through internal validation using a 10-fold cross-validation approach on the training and validation sets. In addition, the performance metrics of each model, including accuracy, precision, sensitivity, F1-score, and area under the curve (AUC), were compared. After selecting the best model, we used hyperparameter optimization to construct a better model. RESULTS RF-selected features (21 in total) had an average root mean square error of 0.30, outperforming principal component analysis. Synthetic Minority Oversampling Technique and Edited Nearest Neighbors showed better accuracy in data balancing. The AdaBoost model was most effective with an AUC of 0.86, accuracy of 75.30%, precision of 0.86, sensitivity of 0.69, and F1-score of 0.76. Validation on the training and validation sets through 10-fold cross-validation gave an AUC of 0.97, an accuracy of 91.27%, a precision of 0.94, a sensitivity of 0.92, and an F1-score of 0.94. After random search processing, the accuracy and AUC of AdaBoost improved. Its accuracy was 77.68% and its AUC was 0.86. CONCLUSIONS This study offered insights into CHF risk prediction. Future research should focus on prospective studies, diverse data, advanced techniques, longitudinal studies, and exploring factor interactions for better CHF prevention and management.

  • Open Access

    Factors Influencing Poststroke Cognitive Dysfunction: Cross-Sectional Analysis (Preprint)

    by Wu Zhou, HaiXia Feng, Hua Tao, Hui Sun, TianTian Zhang, QingXia Wang, Li Zhang

    BACKGROUND Poststroke cognitive impairment (PSCI) is a common and debilitating complication that affects stroke survivors, impacting memory, attention, and executive function. Despite its prevalence, the factors contributing to PSCI remain unclear, with limited insights into how demographic and clinical variables influence cognitive outcomes. OBJECTIVE This study investigates the incidence of cognitive impairment in patients with stroke and examines key demographic and clinical factors, such as age, gender, and education level, which contribute to cognitive decline. The aim is to provide a deeper understanding of PSCI to inform early intervention strategies for improving patient outcomes. METHODS A cross-sectional study was conducted on 305 patients with ischemic stroke admitted to Zhongda Hospital, Southeast University, from January 2019 to September 2022. Cognitive function was assessed using the Mini-Mental State Examination (MMSE) within 72 hours of hospital admission. Demographic information, including age, gender, and education level, were collected. Statistical analyses were performed using chi-square tests, independent t tests, and multivariate regression to assess the relationship between cognitive function and key variables. Pearson correlation analysis explored associations among age, education, and MMSE scores. RESULTS Among the 305 patients with stroke, 16.7% (n=51) were diagnosed with cognitive impairment based on MMSE scores. The prevalence of cognitive impairment was slightly higher in males (17.6%, n=159) than females (15.8%, n=146), but this difference was not statistically significant. A strong negative correlation was found between MMSE scores and age (r=–0.32; P<.01), indicating that older patients had lower cognitive function. Education level showed a positive correlation with MMSE scores (r=0.41; P<.01), with patients with higher educational attainment demonstrating better cognitive outcomes. Cognitive function showed a marked decline in patients older than 60 years, particularly in domains such as memory, attention, and language skills. CONCLUSIONS This study confirms that age and education are significant factors in determining cognitive outcomes after stroke. The results align with existing literature showing that cognitive function declines with age, while higher educational attainment serves as a protective factor. The findings suggest that individuals with greater cognitive reserve, often linked to higher education, are better equipped to cope with the impact of brain injury. However, the study’s reliance on MMSE may have limited its ability to detect domain-specific impairments. Future studies should consider using more sensitive cognitive tools, such as the Montreal Cognitive Assessment (MoCA), to provide a more comprehensive evaluation of PSCI. Cognitive impairment is prevalent among stroke survivors, with age and education level being key factors influencing outcomes. These findings underscore the importance of early detection and targeted interventions to mitigate cognitive decline. Further research with larger samples and more sensitive cognitive assessments is needed to fully understand PSCI and improve rehabilitation strategies for patients with stroke.

  • Open Access

    Current Status of Outcomes Reported by Patients With Stroke and an Analysis of Influencing Factors: Cross-Sectional Questionnaire Study (Preprint)

    by Jia Sun, Liang Ma, Xiao Miao, Hui Sun, SuSu Zhu, Ran Zhang, LeLe Fan, TingTing Hu

    BACKGROUND Stroke is the leading cause of acquired disability and the second leading cause of death worldwide. Its rate of incidence, disability, mortality, and recurrence is high, and the patients experience various symptoms of discomfort, which not only affect their rehabilitation function but also reduce their ability to perform daily activities and their quality of life. Nowadays, with the improvement of China’s medical standards, patients are increasingly attentive to their quality of life and health status. However, diagnostic techniques and effective treatments for patients with stroke are still limited but urgently required. OBJECTIVE This study aimed to evaluate the quality of life during hospitalization using a stroke patient-reported outcomes (PROs) scale and additionally to recognize potential factors and risk indicators that may impact recurrent events, facilitating early intervention measures. METHODS This is a registry-based, retrospective observational cross-sectional study on patients with stroke. A convenient sampling method was used to select various indicators of patients. The Stroke-PRO scale was then used to assess patients’ conditions across physical, psychological, social, and therapeutic domains. Multiple linear regression analysis was applied to identify factors influencing stroke PROs, while correlation analysis was conducted to explore the relationship between these outcomes and blood lipid levels. RESULTS The mean Stroke-PRO score in this study was 4.09 (SD 0.29) points. By multiple linear regression analysis, residence, occupation, physical exercise, Barthel index, Braden scale, National Institutes of Health Stroke Scale scores at admission, and stroke type were the risk factors for reported outcomes of patients with stroke (P<.05). Correlation analysis showed that serum triglyceride, total cholesterol, and low-density lipoprotein were negatively correlated with Stroke-PRO scores in patients with stroke (P<.05), while high-density lipoprotein was positively correlated with patients with stroke (P<.05). The 95% CI was –0.31 to –0.03 for triglyceride, 0.17-0.44 for high-density lipoprotein, –0.29 to –0.01 for cholesterol, –0.30 to –0.02 for low-density lipoprotein, and –0.12 to 0.16 for blood glucose. CONCLUSIONS Patients with stroke have a low level of health, and their reported outcomes need to be improved. Accordingly, nursing staff should pay attention to the quality of life and blood lipid indexes of patients with stroke, actively assess their actual health status, and take early intervention measures to promote their recovery.

  • Open Access

    Nutrition Management Mini Programs in WeChat: Evaluation of Functionality and Quality (Preprint)

    by Hui Sun, Yanping Wu, Jia Sun, Wu Zhou, Qian Xu, Dandan Hu

    BACKGROUND Due to the improvement of living standards and the aging of the population, the medical and health field has attracted much attention.Among them, nutrition management has become the focus of domestic research.Nowadays,with the rapid development of the Internet, developers have developed wechat nutrition management mini programs aimed at improving and monitoring the nutritional health status of individuals. However, there has been insufficient scientific evaluation of the functionality and quality of the various nutritional applets that have emerged on the market. OBJECTIVE The purpose of this study was to evaluate the function and quality of China's wechat nutrition management mini program by using uMARS scale, and to explore the relationship between user rating, function rating and user-version mobile application Rating Scale (uMARS). METHODS Search for keywords such as nutrition, diet and meals on wechat platform, screen related mini programs according to inclusion and exclusion criteria, screen mini programs registered, updated and in use from January 2017 to November 2023, and collect basic information about mini programs such as experience score, registration time, update time and service type . The main functions of the mini program were integrated and summarized, and the function score of each mini program was obtained in the form of 1 point for each function. The quality of the mini program for nutrition management was evaluated by the user version Mobile Mini Program Rating Scale (uMARS) by 2 nutrient-related field evaluators and 1 user experiencer. RESULTS Of 891 small programs were searched and 27 small programs related to nutrition management were finally obtained. 85.2% (23/27) of the mini programs provide diet management, record daily diet to assess nutritional status, 70.4% (19/27) of the mini programs provide nutrition knowledge and classroom teaching, 59.3% (16/27) of the mini programs provide exercise management, daily exercise clock. Only 44.4% (12/27) of the mini programs included weight management.The uMARS total quality median score (out of 5) was 3.38(IQR3.13-3.58), indicating that most small programs scored higher than 3 points ,and the overall range 2.85 to 3.88. The dimensions of engagement ranged from 2 to 4.33 (median 3, IQR 2.67-3.67), with 2 being the lowest score for uMARS. The functional dimension ranges from 3 to 4(IQR 3.33-4), but there are two scores outside the normal range, 2.67 and 4.33. The aesthetic dimension ranges from 2.33 to 4.67(median 3.67,IQR3.33-4) The informational dimension ranges from 2.33 to 4.67(median 3.33, IQR 2.67-3.67). CONCLUSIONS This research evaluated the application of wechat mini programs related to nutrition management. The uMARS scale shows that these mini programs are of average quality, and most concentrate on health rather than nutrition. And the quality is uneven, the function is not perfect, urgent improvement. The future development direction should focus on enhancing quality, innovative functions, and encourage the active participation of professionals to provide more scientific and practical personalized diet guidance.