Related Experiment Video
Updated: Jun 27, 2026
07:50
Facilitating Drug Discovery: An Automated High-content Inflammation Assay in Zebrafish
Published on: July 16, 2012
A predictive framework for assessing naproxen-mediated changes in Nitzschia dubia fucoxanthin levels via machine
Yong Zhou1, Xu Dong2, Mohammad Russel2
1Hubei Key Laboratory of Resource Utilization and Quality Control of Characteristic Crops, College of Life Sciences and Technology, Hubei Engineering University, Xiaogan, 432000, People's Republic of China.
Marine Pollution Bulletin
|June 25, 2026
Summary
This study developed a machine learning dielectric sensor to monitor algal fucoxanthin levels under naproxen stress. The sensor accurately estimated fucoxanthin, showing potential for non-destructive pharmaceutical impact assessment in diatoms.
Area of Science:
- Environmental Science
- Biotechnology
- Analytical Chemistry
Background:
- Non-steroidal anti-inflammatory drugs like naproxen can negatively impact aquatic ecosystems.
- Monitoring physiological responses in algae, such as fucoxanthin production, is crucial for assessing environmental stress.
- Traditional methods for measuring algal compounds can be time-consuming and destructive.
Purpose of the Study:
- To develop and validate a machine learning-driven dielectric sensor framework for estimating fucoxanthin responses in Nitzschia dubia.
- To investigate the effects of acute naproxen stress on algal physiology using dielectric measurements.
- To establish a non-destructive method for monitoring algal health under pharmaceutical contamination.
Main Methods:
- A laboratory-scale dielectric sensor using a coplanar waveguide probe measured dielectric responses (S11, ε') over a broad frequency range.
- Machine learning models, including multivariate statistics, principal component analysis, multiple linear regression, and support vector regression (SVR-RBF), were employed.
- Dielectric data were integrated with optical indices (OD686, chlorophyll-a, fucoxanthin) and analyzed across various naproxen concentrations and cultivation periods.
Main Results:
- The machine learning dielectric approach demonstrated a concentration-dependent naproxen response, with low-dose stimulation and high-dose inhibition.
- Naproxen exposure correlated with changes in dielectric parameters (S11, ε') and optical indices (OD686, fucoxanthin).
- The SVR-RBF model achieved high accuracy (R²=0.913) in estimating fucoxanthin, indicating strong nonlinear relationships between dielectric properties and algal responses.
Conclusions:
- Dielectric spectroscopy coupled with machine learning offers a promising non-destructive method for assessing algal fucoxanthin levels under pharmaceutical stress.
- The developed sensor framework can provide real-time insights into the physiological impact of contaminants like naproxen on marine diatoms.
- This approach facilitates efficient monitoring of water quality and algal health in environments exposed to pharmaceutical pollution.
