Search

Diabetic macular edema (DME) is a significant complication of diabetes that impacts the eye and is a primary contributor to vision loss in individuals with diabetes. Early control of the related risk factors is crucial to reduce the incidence of DME. Artificial intelligence (AI) clinical decision-making tools can construct disease prediction models to aid in the clinical screening of the high-risk population for early disease intervention. However, conventional machine learning and data mining techniques have limitations in predicting diseases when dealing with missing feature values. To solve this problem, a knowledge graph displays the connection relationships of multi-source and multi-domain data in the form of a semantic network to enable cross-domain modeling and queries. This approach can facilitate the personalized prediction of diseases using any number of known feature data. In this study, we proposed an improved correlation enhancement algorithm based on knowledge graph reasoning to comprehensively evaluate the factors that influence DME to achieve disease prediction. We constructed a knowledge graph based on Neo4j by preprocessing the collected clinical data and analyzing the statistical rules. Based on reasoning using the statistical rules of the knowledge graph, we used the correlation enhancement coefficient and generalized closeness degree method to enhance the model. Meanwhile, we analyzed and verified these models’ results using link prediction evaluation indicators. The disease prediction model proposed in this study achieved a precision rate of 86.21%, which is more accurate and efficient in predicting DME. Furthermore, the clinical decision support system developed using this model can facilitate personalized disease risk prediction, making it convenient for the clinical screening of a high-risk population and early disease intervention.

Background: Cancer patients receiving Western medical treatment, frequently seek Traditional Chinese Medicine (TCM) to alleviate adverse effects and prolong survival. Objective: This study evaluated the association between the use of TCM and cancer survival rate. Research into the effect of TCM on patient survival is limited, this analysis focused on 3 patterns of TCM use. Methods: Three retrospective cohorts with different patterns of TCM use were selected from the National Health Insurance Research Database of Taiwan and analyzed. Patients with newly diagnosed cancer between 1997 and 2012 were classified into groups of prediagnosis, postdiagnosis, and continuous TCM use associated with awareness of cancer diagnosis. All demographic and clinical data were analyzed. Results: After propensity score matching, longevity of the postdiagnosis and continuous TCM user was significantly longer than the non-TCM user. The adjusted hazard ratios of death in postdiagnosis and continuous TCM use groups (0.59 and 0.61, respectively) were lower than the non-TCM use group. Conclusion: The analysis suggests that cancer patients using TCM in conjunction with Western medical treatment exhibited a higher survival rate than patients not using TCM treatment.

Due to the many factors that affect commodity prices, price forecasting has become a problematic research point. With the development of machine learning and artificial in-telligence, some advanced ensemble algorithms and deep learning prediction methods based on time series have high accuracy and robustness. These algorithms have gradually become the inevitable choice for solving price prediction problems. Based on the National Bureau of Statistics of China data from January 2012 to December 2021, this study pro-poses deep learning combined forecasting model based on neural networks to predict wheat prices and fill the research gap in agricultural product price forecasting. Researchers utilize Python and Selenium to realize the automatic data acquisition of web pages to achieve the purpose of data collection and calculation. The final price result curve pre-dicted by the price prediction model based on LSTM deep learning agrees with the actual price curve, and the mean square error MSE is only 0.00026. It shows that this prediction model based on time series influenced by multiple factors has an excellent application prospect in price prediction. [ABSTRACT FROM AUTHOR]