AI Article Synopsis
- The study focuses on using artificial neural networks (ANNs) to simultaneously determine nalidixic acid (NA) and its metabolite 7-hydroxymethylnalidixic acid (OH-NA) through their voltammetric signals.
- A PCR model based on standard sample voltammetric data serves as the training set to effectively calibrate the ANN for analysis.
- Recovery rates for measuring NA and OH-NA in urine range from 91-109% and 82-112%, respectively, indicating better performance than traditional partial least squares (PLS) methods.
Article Abstract
Nalidixic acid (NA) and its main metabolite, 7-hydroxymethylnalidixic acid (OH-NA), are simultaneously determined by applying artificial neural networks (ANNs), to their square wave voltammetric signals. The scores of a PCR model, built with the voltammetric data of a set of standard samples, recorded between -0.70 and -1.0V, are used as training set for the net for each compound. The trained nets (ANNs) are used for the simultaneous determination of NA and OH-NA in urine. The recovery values are comprised between 91 and 109% for NA and between 82 and 112% for OH-NA, being these results better than the results obtained by application of partial least squares (PLS) multivariate calibration.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.talanta.2003.08.005 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!