| 英文摘要 |
National Environmental Research Academy’s “Establishment of Green Consumer Chemical Toxicity Prediction Technology” project focuses on environmental specimen analysis. It involves predicting the toxicity of emerging green consumer chemicals detected by instruments. On the other hand, Chemicals Administration’s “Development of a Green Chemical Substitution Diagnostic Module and Decision Support Plan(1/4)” project emphasizes establishing a system for searching, assessing, and screening safe alternative chemicals for high-risk chemicals. This system is designed to provide a rapid means for the Chemicals Administration and the industry to search for and obtain hazard information and risk levels for high-risk chemicals on the regulatory assessment front. Additionally, it offers alternative safe options. The system developed by this project can also assist the Chemicals Administration in searching for hazard assessments of chemical substances.
First, we will collect a large amount of relevant data on chemical toxicity, including existing experimental data, research reports, and literature. This data will be used to establish a comprehensive and reliable toxicity prediction model. We will employ machine learning and statistical analysis techniques to process and mine the data, identifying key features and correlations of chemicals. These features will form the basis of the prediction model and will be used to predict the toxicity potential of new chemicals.
To validate the accuracy and reliability of our prediction model, we will conduct validation using our collected data. The laboratory data will be compared with our prediction results to assess the model’s accuracy. We will also collaborate with relevant organizations and experts to gather their opinions and suggestions, further improving and optimizing our prediction technology.
Lastly, we will establish an open and sustainable platform to provide chemical toxicity prediction services. Through this platform, consumers can access toxicity information on different chemicals, based on their individual needs and health considerations. By integrating data from domestic and international databases, implementing quantitative structure-activity relationship (QSAR) models, and connecting with analytical instruments, we will establish a workflow for green consumer chemical toxicity prediction. Temporary patents will be sought in Taiwan and the United States, and educational training on data searching techniques will be conducted to promote the adoption of green chemistry. Overall, this project aims to bridge the gap between green consumerism and chemical toxicity prediction, ensuring a sustainable approach to chemical product selection and usage.
|