Data fusion of near-infrared and mid-infrared spectroscopy for rapid origin identification and quality evaluation of Lonicerae japonicae flos
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.This study introduces a data fusion strategy using near-infrared (NIR) and mid-infrared (MIR) spectroscopy for identifying the origin and evaluating the quality of Lonicerae japonicae flos (LJF). The combined spectral data fusion models significantly improved accuracy and practical value in natural product quality control.
Area Of Science
- Analytical Chemistry
- Spectroscopy
- Chemometrics
Background
- Lonicerae japonicae flos (LJF) is a valuable traditional medicine requiring robust quality control.
- Current methods for LJF authentication and quality assessment can be time-consuming and labor-intensive.
- Developing rapid and accurate analytical techniques is crucial for ensuring the efficacy and safety of LJF.
Purpose Of The Study
- To develop a data fusion strategy integrating near-infrared (NIR) and mid-infrared (MIR) spectroscopy for LJF.
- To enable rapid origin identification and quality evaluation of LJF.
- To demonstrate the superiority of data fusion models over single-spectroscopy methods.
Main Methods
- Development of a high-level data fusion model using the soft voting method for origin identification.
- Construction of a data fusion model employing a weighted average algorithm to convert spectral data to liquid chromatography (LC) data.
- Utilizing Euclidean distance and adjusted cosine similarity to assess the accuracy of spectral-to-LC data conversion.
Main Results
- The high-level data fusion model achieved high performance for origin identification (95.5% accuracy, 0.910 Kappa value).
- The spectral-to-LC data conversion model demonstrated high similarity to real chromatographic data (Euclidean distance 247.990, adjusted cosine similarity 0.996).
- Data fusion models consistently outperformed models based on single NIR or MIR data.
Conclusions
- Multispectral data fusion using NIR and MIR spectroscopy is effective for rapid LJF origin identification and quality control.
- The developed data fusion strategy offers significant practical value for the quality management of natural products.
- This approach provides a foundation for advanced analytical methods in the field of herbal medicine authentication.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
