基于机器学习的全氟化合物识别方法及装置
By employing a machine learning-based method for identifying perfluorinated compounds, utilizing a multimodal neural network model and a pre-defined compound classification library, the problem of poor accuracy in identifying perfluorinated compounds is solved. This method enables efficient screening and structural identification of unknown or structurally diverse perfluorinated compounds, and possesses high-throughput and automated analysis capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGZHOU INST FOR ADVANCED STUDY UCAS
- Filing Date
- 2026-04-10
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for identifying perfluorinated compounds have limited ability to determine structures when there is uncertainty in the formation of mass spectrometry fragments, making it difficult to effectively identify unknown or structurally diverse perfluorinated compounds.
A machine learning-based method for identifying perfluorinated compounds was developed. By acquiring mass spectrometry data, extracting mass spectrometry feature information, training a multimodal neural network model, and combining it with a pre-defined compound classification library, the method can achieve structural identification and hazard assessment of perfluorinated compounds.
It enables efficient screening and hierarchical structure identification of unknown or structurally diverse perfluorinated compounds, possesses high-throughput and automated analysis capabilities, and meets the needs for rapid screening and systematic evaluation of samples from complex environments.
Smart Images
Figure CN122024934B_ABST