Evaluating economic efficiency of the national high-tech industrial development districts in the Yangtze River Delta by stochastic frontier analysis.

The High-tech Industrial Development Districts (HIDDs) are technological engines for the regional economy in China. The Chinese government implemented the Torch Initiative to accelerate industrial agglomeration and innovation development by administratively upgrading the local HIDDs into a national list since 1989. The policy intervention emphasizes the labor and capital inputs on science and technology. The study adopts the Stochastic Frontier Analysis (SFA) to understand the economic efficiency of the enlisted national HIDDs in the Yangtze River Delta. The results suggest that the average efficiency trends of the all-in-total sales, product sales and commodity sales are decreasing while that of technology sales is increasing from 2007 to 2019. In the total sales efficiency evaluation, most early-enrolled HIDDs are in the high-efficiency group whereas the newcomers are in the low-efficiency group. The Nantong HIDD has the highest efficiency, followed by Wuxi, Taizhou, Suzhou, Nanjing, Shanghai ZJ, Xiaoshan, Ningbo, Suzhou IP and Jiangyin HIDDs. Huainan, Huaian, Tongling, Shanghai ZZ, Lianyungang and Suqian HIDDs rank very low in the total sales evaluation. Besides, Hangzhou, Nantong and Ningbo HIDDs rank first in the technology, product and commodity sales efficiency evaluation correspondingly. In addition, the increase of S&T labor inputs would decrease the efficiency of the product, commodity and total sales while the increase of the S&T capital inputs is as expected a facilitator to the efficiency of product, commodity and total sales. This study contributes to the discussion on the role of political intervention in technological innovation by breaking down the economic efficiency into three major parts, the efficiency of technology, product and commodity sales. The findings could help policymakers strike a balance between the human and capital inputs in regional innovation capability.

Is bone marrow biopsy and aspiration still mandatory when 18F-FDG PET/CT is available for the initial assessment of bone marrow metastasis in pediatric Ewing sarcoma?

Purpose: To compare the diagnostic value of 18F-fluorodeoxyglucose-positron emission tomography/computed tomography (18F-FDG PET/CT) and bone marrow biopsy and aspiration (BMBA) for evaluating bone marrow metastases (BMM) in newly diagnosed pediatric Ewing sarcoma (ES). Material and methods: To assess the diagnostic accuracy of 18F-FDG PET/CT against BMBA for marrow infiltration in ES patients, a retrospective analysis encompassed 103 ES patients from the Children's Hospital of Chongqing Medical University, spanning nine years, who underwent both 18F-FDG PET/CT and BMBA at the point of diagnosis. Results: The median age of this study was 9.3(15 days to 17.1 years), 52(50.5%) patients were male. Among the cohort, 8 subjects received a BMM diagnosis via marrow cytology or histopathology, concomitant with positive 18F-FDG PET/CT findings. An additional 4 patients were identified with BMM solely through 18F-FDG PET/CT. No cytologically or histologically positive BMM were found in PET/CT-negative patients. Therefore, within this selected sample group, the 18F-FDG PET/CT imaging technique exhibited sensitivity of 100% and specificity of 95.8%. The five-year overall survival rate decreased from 57.5% among the entire cohort of patients to a mere 30% for individuals suffering from BMM. Conclusion: Given these findings, the prevailing reliance on BMBA warrants reevaluation when 18F-FDG PET/CT is available, potentially heralding a shift towards less invasive diagnostic modalities in the management of ES.

Incidence, prevalence, and medical costs of pressure injuries in Taiwan from 2001 to 2015: Results of a retrospective cohort study.

AIMS AND OBJECTIVES: This study aims to analyse the trends in the incidence, prevalence and medical costs of pressure injuries (PIs) among genders in Taiwan. BACKGROUND: The treatment of PIs is complex and costly, often leading to complications and increased mortality. This issue significantly impacts healthcare quality and incurs substantial medical and social costs, warranting attention. METHODS: A retrospective cohort study was conducted using data from Taiwan's National Health Insurance Database to obtain and calculate the incidence, prevalence, and medical costs of PIs in the country between 2001 and 2015 as well as to analyse high-risk groups and the medical care utilisation of patients following the STROBE reporting guidelines. RESULTS: Between 2001 and 2015, 15,327 incident case of PIs were diagnosed. During the study period, the prevalence rate of PIs per 100,000 population rose from 26.3 to 189.6, with approximately 11.5%-16.3% of patients undergoing surgical debridement. The PIs prevalence rate increased by 7.2-fold, and hospitalisation costs accounted for 91.7%-96.0% of the total medical costs. Patients with older age, comorbidities, poorer financial status and lower education levels were found to be likely to develop PIs. These predisposing factors differed between males and females. The prevalence of PIs was higher in patients ≥75 years old than in patients from other age groups. Moreover, PI-related medical expenses have been increasing annually. CONCLUSIONS: In Taiwan, the rising incidence of PIs is driving up medical costs. Effective care and prevention of PIs necessitate a comprehensive plan from the entire healthcare system. RELEVANCE TO CLINICAL PRACTICE: This research fills a gap in the available data on the incidence, prevalence, and medical costs of PIs in Taiwan and Asia. PATIENT OR PUBLIC CONTRIBUTION: The findings can be used to help develop clinical guidelines for preventive education and treatment of PIs.

Eco-environmental changes due to human activities in the Erhai Lake Basin from 1990 to 2020.

Human activities have increased with urbanisation in the Erhai Lake Basin, considerably impacting its eco-environmental quality (EEQ). This study aims to reveal the evolution and driving forces of the EEQ using water benefit-based ecological index (WBEI) in response to human activities and policy variations in the Erhai Lake Basin from 1990 to 2020. Results show that (1) the EEQ exhibited a pattern of initial degradation, subsequent improvement, further degradation and a rebound from 1990 to 2020, and the areas with poor and fair EEQ levels mainly concentrated around the Erhai Lake Basin with a high level of urbanisation and relatively flat terrain; (2) the EEQ levels were not optimistic in 1990, 1995 and 2015, and areas with poor and fair EEQ levels accounted for 43.41%, 47.01% and 40.05% of the total area, respectively; and (3) an overall improvement in the EEQ was observed in 1995-2000, 2000-2005, 2005-2009 and 2015-2020, and the improvement was most significant in 1995-2000, covering an area of 823.95 km2 and accounting for 31.79% of the total area. Results also confirmed that the EEQ changes in the Erhai Lake Basin were primarily influenced by human activities and policy variations. Moreover, these results can provide a scientific basis for the formulation and planning of sustainable development policy in the Erhai Lake Basin.

Establishment of low-cost production platforms of polyhydroxyalkanoate bioplastics from Halomonas cupida J9.

Microbial production of polyhydroxyalkanoate (PHA) is greatly restricted by high production cost arising from high-temperature sterilization and expensive carbon sources. In this study, a low-cost PHA production platform was established from Halomonas cupida J9. First, a marker-less genome-editing system was developed in H. cupida J9. Subsequently, H. cupida J9 was engineered to efficiently utilize xylose for PHA biosynthesis by introducing a new xylose metabolism module and blocking xylonate production. The engineered strain J9UΔxylD-P8xylA has the highest PHA yield (2.81 g/L) obtained by Halomonas with xylose as the sole carbon source so far. This is the first report on the production of short- and medium-chain-length (SCL-co-MCL) PHA from xylose by Halomonas. Interestingly, J9UΔxylD-P8xylA was capable of efficiently utilizing glucose and xylose as co-carbon sources for PHA production. Furthermore, fed-batch fermentation of J9UΔxylD-P8xylA coupled to a glucose/xylose co-feeding strategy reached up to 12.57 g/L PHA in a 5-L bioreactor under open and unsterile condition. Utilization of corn straw hydrolysate as the carbon source by J9UΔxylD-P8xylA reached 7.0 g/L cell dry weight (CDW) and 2.45 g/L PHA in an open fermentation. In summary, unsterile production in combination with inexpensive feedstock highlights the potential of the engineered strain for the low-cost production of PHA from lignocellulose-rich agriculture waste.

Assessment of optimal combinations of therapeutic probiotics for depression, anxiety, and stress.

BACKGROUND: Accumulating data show that probiotics may be beneficial for reducing depressive, anxiety, and stress symptoms. However, the best combinations and species of probiotics have not been identified. The objective of our study was to assess the most effective combinations and components of different probiotics through network meta-analysis. METHOD: A systematic search of four databases, PubMed, Web of Science, Cochrane, and Embase, was conducted from inception to 11 January 2024. The GRADE framework was used to assess the quality of evidence contributing to each network estimate. RESULTS: We deemed 45 trials eligible, these included 4053 participants and 10 types of interventions. The quality of evidence was rated as high or moderate. The NMA revealed that Bifidobacterium exhibited a greater probability of being the optimal probiotic species for improving anxiety symptoms (SMD = -0.80; 95% CI -1.49 to -0.11), followed by Lactobacillus (SMD = -0.49; 95% CI -0.85 to -0.12). In addition, for multiple strains, compared with the other interventions, Lactobacillus + Bifidobacterium (SMD = -0.41; 95% CI -0.73 to -0.10) had a positive effect on depression. CONCLUSION: The NMA revealed that Lactobacillus and Bifidobacterium had prominent efficacy in the treatment of individuals with anxiety, depression, and combination of Lactobacillus + Bifidobacterium had a similar effect. With few direct comparisons available between probiotic species, this NMA may be instrumental in shaping the guidelines for probiotic treatment of psychological disorders.

Structural and Functional Disparities within the Human Gut Virome in Terms of Genome Topology and Representative Genome Selection.

Circularity confers protection to viral genomes where linearity falls short, thereby fulfilling the form follows function aphorism. However, a shift away from morphology-based classification toward the molecular and ecological classification of viruses is currently underway within the field of virology. Recent years have seen drastic changes in the International Committee on Taxonomy of Viruses' operational definitions of viruses, particularly for the tailed phages that inhabit the human gut. After the abolition of the order Caudovirales, these tailed phages are best defined as members of the class Caudoviricetes. To determine the epistemological value of genome topology in the context of the human gut virome, we designed a set of seven experiments to assay the impact of genome topology and representative viral selection on biological interpretation. Using Oxford Nanopore long reads for viral genome assembly coupled with Illumina short-read polishing, we showed that circular and linear virus genomes differ remarkably in terms of genome quality, GC skew, transfer RNA gene frequency, structural variant frequency, cross-reference functional annotation (COG, KEGG, Pfam, and TIGRfam), state-of-the-art marker-based classification, and phage-host interaction. Furthermore, the disparity profile changes during dereplication. In particular, our phage-host interaction results demonstrated that proportional abundances cannot be meaningfully compared without due regard for genome topology and dereplication threshold, which necessitates the need for standardized reporting. As a best practice guideline, we recommend that comparative studies of the human gut virome always report the ratio of circular to linear viral genomes along with the dereplication threshold so that structural and functional metrics can be placed into context when assessing biologically relevant metagenomic properties such as proportional abundance.

The mediating role of community identity in the relationship between social class and life satisfaction: Evidence from Chinese community residents.

Research has shown that social class is an important predictor of life satisfaction. However, the underlying mechanism for this relationship is yet to be fully elucidated. The study examined the underlying mechanism based on the social identity approach to health. Study 1 recruited 577 community residents to complete self-report questionnaires. Correlation analysis showed that social class, community identity, and life satisfaction were positively correlated with each other. Regression analysis showed that community identity mediated the relationship. To increase the replicability and derive causal inference of the results, Study 2 was a randomized control trial (N = 76) that used the resource-availability task to manipulate subjective social class, and found that life satisfaction in the lower-class group was significantly lower than that in the control group. Further analysis showed that social class predicted life satisfaction through the mediating role of community identity. The findings provide potential strategies to enhance community residents' life satisfaction.

Understanding the efficiency of "political attention and governance action" on marine waste discharge in the coastal provinces in China.

This study examines the "black box" of the policy process by evaluating the efficiency of "political attention and governance action" as a two-stage continuum and investigating the influencing factors in marine waste discharge governance. The results of empirical analysis in the 11 Chinese coastal provincial regions suggest that the overall efficiency fluctuated from 1999 to 2016 with a minor reduction. The developed provinces such as Jiangsu, Zhejiang, Shanghai and Guangdong have low efficiency in overall evaluation. In the first stage, the transferring efficiency from political attention to actions does not change considerably despite fluctuations in 2012-2016. The second-stage efficiency, or the efficiency of the governance actions, has been declining since 2008 except for a brief spike in 2014. There exists a feedback impact between the two stages. Economic competition, as measured by the average GDP of neighbouring provinces, has more negative impacts on the first-stage efficiency.

An Integrated Method for Tunnel Health Monitoring Data Analysis and Early Warning: Savitzky-Golay Smoothing and Wavelet Transform Denoising Processing.

A tunnel health monitoring (THM) system ensures safe operations and effective maintenance. However, how to effectively process and denoise several data collected by THM remains to be addressed, as well as safety early warning problems. Thus, an integrated method for Savitzky-Golay smoothing (SGS) and Wavelet Transform Denoising (WTD) was used to smooth data and filter noise, and the coefficient of the non-uniform variation method was proposed for early warning. The THM data, including four types of sensors, were attempted using the proposed method. Firstly, missing values, outliers, and detrend in the data were processed, and then the data were smoothed by SGS. Furthermore, data denoising was carried out by selecting wavelet basis functions, decomposition scales, and reconstruction. Finally, the coefficient of non-uniform variation was employed to calculate the yellow and red thresholds. In data smoothing, it was found that the Signal Noise Ratio (SNR) and Root Mean Square Error (RMSE) of SGS smoothing were superior to those of the moving average smoothing and five-point cubic smoothing by approximately 10% and 30%, respectively. An interesting phenomenon was discovered: the maximum and minimum values of the denoising effects with different wavelet basis functions after selection differed significantly, with the SNR differing by 14%, the RMSE by 8%, and the r by up to 80%. It was found that the wavelet basis functions vary, while the decomposition scales are consistently set at three layers. SGS and WTD can effectively reduce the complexity of the data while preserving its key characteristics, which has a good denoising effect. The yellow and red warning thresholds are categorized into conventional and critical controls, respectively. This early warning method dramatically improves the efficiency of tunnel safety control.

A model for predicting physical function upon discharge of hospitalized older adults in Taiwan-a machine learning approach based on both electronic health records and comprehensive geriatric assessment.

Background: Predicting physical function upon discharge among hospitalized older adults is important. This study has aimed to develop a prediction model of physical function upon discharge through use of a machine learning algorithm using electronic health records (EHRs) and comprehensive geriatrics assessments (CGAs) among hospitalized older adults in Taiwan. Methods: Data was retrieved from the clinical database of a tertiary medical center in central Taiwan. Older adults admitted to the acute geriatric unit during the period from January 2012 to December 2018 were included for analysis, while those with missing data were excluded. From data of the EHRs and CGAs, a total of 52 clinical features were input for model building. We used 3 different machine learning algorithms, XGBoost, random forest and logistic regression. Results: In total, 1,755 older adults were included in final analysis, with a mean age of 80.68 years. For linear models on physical function upon discharge, the accuracy of prediction was 87% for XGBoost, 85% for random forest, and 32% for logistic regression. For classification models on physical function upon discharge, the accuracy for random forest, logistic regression and XGBoost were 94, 92 and 92%, respectively. The auROC reached 98% for XGBoost and random forest, while logistic regression had an auROC of 97%. The top 3 features of importance were activity of daily living (ADL) at baseline, ADL during admission, and mini nutritional status (MNA) during admission. Conclusion: The results showed that physical function upon discharge among hospitalized older adults can be predicted accurately during admission through use of a machine learning model with data taken from EHRs and CGAs.

The effect of political turnover on corporate ESG performance: Evidence from China.

This paper aims to investigate the effect of political turnover on corporate ESG performance in China. By analyzing data from Chinese A-share-listed companies between 2010 and 2020, we have discovered that changes in the municipal party committee secretary or the mayor of the prefecture-level city where a firm is located have a detrimental effect on corporate ESG performance. Compared with the change of the party committee, the change of mayor has a more pronounced negative impact on ESG performance. The reason behind this negative effect is primarily attributed to policy uncertainty, which leads to a decrease in governmental subsidies and an increase in ineffective under-investment by companies, consequently resulting in decreased corporate ESG performance. Furthermore, we have also observed that the adverse influence of political turnover on corporate ESG performance is relatively mitigated in SOEs, politically connected firms, and tertiary industries. These findings contribute to a deeper understanding of the relationship between political uncertainty and corporate behavior, particularly in emerging markets.

Unsupervised blind image quality assessment via joint spatial and transform features.

A novel unsupervised blind image quality assessment (BIQA) method, which requires no mean opinion scores for model training is presented in this paper. The method employs joint spatial and transform features as quality degradation metrics, specifically, phase congruency, gradient magnitude (GM), and GM and Laplacian of Gaussian response and local normalized coefficient are extracted as spatial features, and Karhunen-Loéve transform coefficient and discrete cosine transform coefficient are modeled as transform features. Both spatial and transform features are well analyzed to remove the redundancy, and then fitted to the multivariate Gaussian model for no-reference image quality assessment. Extensive experiments conducted on seven IQA databases demonstrate the superiority of the proposed method over the state-of-the-art both supervised and unsupervised BIQA methods.

Implementation of an Attention Mechanism Model for Facial Beauty Assessment Using Transfer Learning.

An important consideration in medical plastic surgery is the evaluation of the patient's facial symmetry. However, because facial attractiveness is a slightly individualized cognitive experience, it is difficult to determine face attractiveness manually. This study aimed to train a model for assessing facial attractiveness using transfer learning while also using the fine-grained image model to separate similar images by first learning features. In this case, the system can make assessments based on the input of facial photos. Thus, doctors can quickly and objectively treat patients' scoring and save time for scoring. The transfer learning was combined with CNN, Xception, and attention mechanism models for training, using the SCUT-FBP5500 dataset for pre-training and freezing the weights as the transfer learning model. Then, we trained the Chang Gung Memorial Hospital Taiwan dataset to train the model based on transfer learning. The evaluation uses the mean absolute error percentage (MAPE) value. The root mean square error (RMSE) value is used as the basis for experimental adjustment and the quantitative standard for the model's predictive. The best model can obtain 0.50 in RMSE and 18.5% average error in MAPE. A web page was developed to infer the deep learning model to visualize the predictive model.

Research on Early Identification Model of Intravenous Immunoglobulin Resistant Kawasaki Disease Based on Gradient Boosting Decision Tree.

BACKGROUND: To construct a predictive model for intravenous immunoglobulin (IVIG) resistant Kawasaki disease (KD) based on the gradient boosting decision tree (GBDT), so as to early identify children with IVIG resistance and actively take additional treatment to prevent adverse events. METHODS: The case data of KD children hospitalized in the Pediatric Department of Lanzhou University Second Hospital from October 2015 to July 2020 were collected. All KD patients were divided into IVIG responsive group and IVIG resistant group. GBDT was used to explore the influencing factors of IVIG-resistant KD and to construct a prediction model. Then compared with previous models, the optimal model was selected. RESULTS: In the process of GBDT model construction, 80% of the data were used as the test set, and 20% of the data were used as the validation set. Among them, the verification set was used to adjust the hyperparameters in GDBT learning. The model performed best with a hyperparameter tree depth of 5. The area under the curve of the GBDT model constructed based on the best parameters was 0.87 (95% CI: 0.85-0.90), the sensitivity was 72.62%, the specificity was 89.04%, and the accuracy was 61.65%. The contribution degree of each feature value to the model was total bilirubin, albumin, C-reactive protein, fever time, and Na in order. CONCLUSION: The GBDT model is more suitable for the prediction of IVIG-resistant KD in this study area.

A study on the changing trend and influencing factors of hospitalization costs of schizophrenia in economically underdeveloped areas of China.

The public health problems caused by schizophrenia are becoming increasingly prominent and can place a huge economic burden on society. This study takes Gansu Province as an example to analyze the level and changing trend of the economic burden of schizophrenia inpatients in economically underdeveloped areas of China. Using a multi-stage stratified cluster sampling method, 39,054 schizophrenics from 197 medical and health institutions in Gansu Province were selected as the research objects, and their medical expenses and related medical records were obtained from the medical information system. The rank sum test and Spearman rank correlation were used for univariate analysis. Quantile regression and random forest were used to analyze the influencing factors. The results show that the average length of stay of schizophrenics in Gansu Province of China was 52.01 days, and the average hospitalization cost was USD1653.96 from 2014 to 2019. During the six years, the average hospitalization costs per time decreased from USD2136.85 to USD1401.33. The average out-of-pocket costs per time decreased from USD1238.78 to USD267.68. And the average daily hospitalization costs increased from USD38.18 to USD41.25. The main factors influencing hospitalization costs are length of stay, proportion of medications, and schizophrenic subtype. The hospitalization costs per time of schizophrenics in Gansu Province have decreased but remain at a high level compared to some other chronic non-communicable diseases. In the future, attention should be paid to improving the efficiency of medical institutions, enhancing community management, and promoting the transformation of the management model of schizophrenia.

Current and potential trends in the bioactive properties and health benefits of Prunus mume Sieb. Et Zucc: a comprehensive review for value maximization.

Prunus mume Sieb. Et Zucc (P. mume) is an acidic fruit native to China (named Chinese Mei or greengage plum). It is currently cultivated in several Asian countries, including Japan ("Ume"), Korea (Maesil), and Vietnam (Mai or Mo). Due to its myriad nutritional and functional properties, it is accepted in different countries, and its characteristics account for its commercialization. In this review, we summarize the information on the bioactive compounds from the fruit of P. mume and their structure-activity relationships (SAR); the pulp has the highest enrichment of bioactive chemicals. The nutritional properties of P. mume and the numerous uses of its by-products make it a potential functional food. P. mume extracts exhibit antioxidant, anticancer, antimicrobial, and anti-hyperuricaemic properties, cardiovascular protective effects, and hormone regulatory properties in various in vitro and in vivo assays. SAR shows that the water solubility, molecular weight, and chemical conformation of P. mume extracts are closely related to their biological activity. However, further studies are needed to evaluate the fruit's potential nutritional and functional therapeutic mechanisms. The industrial process of large-scale production of P. mume and its extracts as functional foods or nutraceuticals needs to be further optimized.

Development of a personalized dendritic cell vaccine and single-cell RNA sequencing-guided assessment of its cell type composition.

BACKGROUND AIMS: Dendritic cell (DC)-based immunotherapy is a promising approach to treat cancer; however, there is no consensus on the manufacturing processes. Cell type heterogeneity in products manufactured by various methods is understudied and may elicit safety concerns from the regulatory perspective. METHODS: We characterized the cell type composition of a recently developed DC vaccine, CUD-002, consisting of DCs loaded with mRNA encoding personalized tumor neoantigens (NCT05270720). RESULTS: Using single-cell transcriptomic analysis as an unbiased approach, we found that >80% cells in the final product were DCs and the rest primarily comprised myelocytes and lymphocytes. Subsequent fluorescence-activated cell sorting analyses confirmed these cellular identities. These results indicate that unintended cells originate from leukapheresis, the first step of the manufacturing process, and thus likely safe. Consistently, no overt toxicity or tumorigenicity was observed in mice inoculated with CUD-002. CONCLUSIONS: Considering that leukapheresis is a widely used procedure for collecting diverse peripheral blood cell types to manufacture various cytotherapies, this study establishes a workflow to analyze and address regulatory considerations on cell type heterogeneity.

Human expansion into Asian highlands in the 21st Century and its effects.

Most intensive human activities occur in lowlands. However, sporadic reports indicate that human activities are expanding in some Asian highlands. Here we investigate the expansions of human activities in highlands and their effects over Asia from 2000 to 2020 by combining earth observation data and socioeconomic data. We find that ∼23% of human activity expansions occur in Asian highlands and ∼76% of these expansions in highlands comes from ecological lands, reaching 95% in Southeast Asia. The expansions of human activities in highlands intensify habitat fragmentation and result in large ecological costs in low and lower-middle income countries, and they also support Asian developments. We estimate that cultivated land net growth in the Asian highlands contributed approximately 54% in preventing the net loss of the total cultivated land. Moreover, the growth of highland artificial surfaces may provide living and working spaces for ∼40 million people. Our findings suggest that highland developments hold dual effects and provide new insight for regional sustainable developments.

A model for predicting fall risks of hospitalized elderly in Taiwan-A machine learning approach based on both electronic health records and comprehensive geriatric assessment.

Backgrounds: Falls are currently one of the important safety issues of elderly inpatients. Falls can lead to their injury, reduced mobility and comorbidity. In hospitals, it may cause medical disputes and staff guilty feelings and anxiety. We aimed to predict fall risks among hospitalized elderly patients using an approach of artificial intelligence. Materials and methods: Our working hypothesis was that if hospitalized elderly patients have multiple risk factors, their incidence of falls is higher. Artificial intelligence was then used to predict the incidence of falls of these patients. We enrolled those elderly patients aged >65 years old and were admitted to the geriatric ward during 2018 and 2019, at a single medical center in central Taiwan. We collected 21 physiological and clinical data of these patients from their electronic health records (EHR) with their comprehensive geriatric assessment (CGA). Data included demographic information, vital signs, visual ability, hearing ability, previous medication, and activity of daily living. We separated data from a total of 1,101 patients into 3 datasets: (a) training dataset, (b) testing dataset and (c) validation dataset. To predict incidence of falls, we applied 6 models: (a) Deep neural network (DNN), (b) machine learning algorithm extreme Gradient Boosting (XGBoost), (c) Light Gradient Boosting Machine (LightGBM), (d) Random Forest, (e) Stochastic Gradient Descent (SGD) and (f) logistic regression. Results: From modeling data of 1,101 elderly patients, we found that machine learning algorithm XGBoost, LightGBM, Random forest, SGD and logistic regression were successfully trained. Finally, machine learning algorithm XGBoost achieved 73.2% accuracy. Conclusion: This is the first machine-learning based study using both EHR and CGA to predict fall risks of elderly. Multiple risk factors of falls in hospitalized elderly patients can be put into a machine learning model to predict future falls for early planned actions. Future studies should be focused on the model fitting and accuracy of data analysis.