Analytical Chemistry
Bioinformatics and machine learning-based simultaneous spectrophotometric identification of Montelukast sodium and Desloratadine in a commercial film-coated tablet
Desloratadine (DL) and montelukast sodium (MTS) are combined in a pharmaceutical preparation for the treatment of allergic rhinitis and asthma. DL is a competitive H1-receptor antagonist and helps the management of allergic reactions and relieves allergy symptoms, while MTS is a leukotriene receptor antagonist that inhibits the effects of inflammatory mediators. This study introduces a machine learning-assisted UV-Vis spectrophotometric method for the simultaneous quantification of DL and MTS in a commercial film-coated tablet, addressing limitations due to spectral overlap.
Five machine learning regression models (Support Vector Regression, Ridge Regression, Lasso Regression, Elastic Net, and Linear Regression) were evaluated. Ridge Regression (λ=0.1) was selected for its balance of accuracy, computational efficiency, and robustness. The method was applied using Aircomb® film-coated tablets, containing 5 mg DL and 10.4 mg MTS, ensuring high precision.
The developed method demonstrated high recovery rates (99.25% for DL and 101.0% for MTS with minimal relative standard deviation (≤ 1.59%). Sustainability assessments using Analytical GREEnness Metric (AGREE) and Complex Green Analytical Procedure Index (ComplexGAPI) confirmed its alignment with green analytical chemistry principles. Ridge Regression (λ=0.1) provided accurate and reproducible results, making it suitable for routine pharmaceutical analysis.
This study highlights the potential of machine learning-assisted UV-Vis spectrophotometry as a cost-effective and environmentally friendly alternative for pharmaceutical quality control. The method minimizes solvent consumption while ensuring analytical precision. Future research may explore non-linear models, such as artificial neural networks, to enhance predictive performance and broaden its applicability.
Executive Impact: Transforming Pharmaceutical Analysis
Traditional spectrophotometric methods face limitations in quantifying Desloratadine (DL) and Montelukast sodium (MTS) simultaneously in pharmaceutical preparations due to significant spectral overlap, necessitating complex pre-separation steps or more expensive, time-consuming techniques like HPLC. This creates a need for a cost-effective, environmentally friendly, and efficient analytical method.
The study proposes a machine learning-assisted UV-Vis spectrophotometric method, leveraging Ridge Regression (λ=0.1) for simultaneous quantification of DL and MTS. This approach addresses spectral overlap limitations, provides high accuracy and precision, and aligns with green analytical chemistry principles by minimizing solvent consumption and waste.
0
Reduction in Solvent Consumption
0
Improvement in Analysis Speed
0
Cost Savings per Analysis
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
99.25%
DL Recovery Rate
101.0%
MTS Recovery Rate
| Feature | ML-assisted UV-Vis | Traditional HPLC |
|---|---|---|
| Cost-Effectiveness |
|
|
| Analysis Time |
|
|
| Solvent Consumption |
|
|
| Waste Generation |
|
|
| Spectral Overlap Handling |
|
|
| Accuracy & Precision |
|
|
| Environmental Impact |
|
|
Machine Learning-Assisted Spectrophotometry Workflow
Spectroscopic Measurements (UV-Vis)
→
Data Pre-processing (SVD, Noise Reduction)
→
Machine Learning Model Training (Ridge Regression)
→
Hyperparameter Optimization (λ=0.1)
→
Model Validation & Performance Evaluation
→
Simultaneous Quantification of Analytes
Application in Pharmaceutical Quality Control
The developed Ridge Regression (λ=0.1) model was successfully applied to Aircomb® film-coated tablets. This demonstrated its capability for accurate and reproducible quantification of Desloratadine (5 mg) and Montelukast sodium (10.4 mg) in a real-world pharmaceutical product. The method's high recovery rates (DL: 99.25%, MTS: 101.0%) and low relative standard deviation (≤ 1.59%) confirm its suitability for routine quality control, offering a greener and more cost-effective alternative to traditional methods like HPLC.
Key Outcome: Enhanced QA/QC Efficiency
Advanced ROI Calculator
Estimate your potential efficiency gains and cost savings by implementing AI-driven analytical solutions in your enterprise.
Estimated Annual Savings
$0
Annual Hours Reclaimed
0
Your AI Implementation Roadmap
A structured approach to integrating machine learning into your analytical workflows, ensuring seamless transition and maximized benefits.
Phase 1: Data Preprocessing & SVD
Cleaning and dimensionality reduction of spectral data using Singular Value Decomposition to prepare for model training. (Estimated: 2-3 weeks)
Phase 2: Model Training & Optimization
Training five regression models (SVR, Ridge, Lasso, Elastic Net, Linear) and optimizing hyperparameters (e.g., Ridge Regression λ=0.1). (Estimated: 3-4 weeks)
Phase 3: Validation & Performance Evaluation
Testing models against internal and external validation sets, assessing MAE, RMSE, R², and adjusted R² to select the best-performing model. (Estimated: 2-3 weeks)
Phase 4: Real-world Application & Sustainability Assessment
Applying the selected model (Ridge Regression) to commercial tablets and conducting AGREE/ComplexGAPI assessments. (Estimated: 1-2 weeks)
Ready to Transform Your Analytical Chemistry?
Our experts are ready to guide you through implementing cutting-edge AI solutions for superior analytical precision and efficiency.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat