Machine Learning-Enabled NIR Spectroscopy. Part 3: Hyperparameter by Design (HyD) Based ANN-MLP Optimization, Model Generalizability, and Model Transferability
- Title
- Machine Learning-Enabled NIR Spectroscopy. Part 3: Hyperparameter by Design (HyD) Based ANN-MLP Optimization, Model Generalizability, and Model Transferability
- Creator
- Ali H.; Muthudoss P.; Chauhan C.; Kaliappan I.; Kumar D.; Paudel A.; Ramasamy G.
- Description
- Data variations, library changes, and poorly tuned hyperparameters can cause failures in data-driven modelling. In such scenarios, model drift, a gradual shift in model performance, can lead to inaccurate predictions. Monitoring and mitigating drift are vital to maintain model effectiveness. USFDA and ICH regulate pharmaceutical variation with scientific risk-based approaches. In this study, the hyperparameter optimization for the Artificial Neural Network Multilayer Perceptron (ANN-MLP) was investigated using open-source data. The design of experiments (DoE) approach in combination with target drift prediction and statistical process control (SPC) was employed to achieve this objective. First, pre-screening and optimization DoEs were conducted on lab-scale data, serving as internal validation data, to identify the design space and control space. The regression performance metrics were carefully monitored to ensure the right set of hyperparameters was selected, optimizing the modelling time and storage requirements. Before extending the analysis to external validation data, a drift analysis on the target variable was performed. This aimed to determine if the external data fell within the studied range or required retraining of the model. Although a drift was observed, the external data remained well within the range of the internal validation data. Subsequently, trend analysis and process monitoring for the mean absolute error of the active content were conducted. The combined use of DoE, drift analysis, and SPC enabled trend analysis, ensuring that both current and external validation data met acceptance criteria. Out-of-specification and process control limits were determined, providing valuable insights into the models performance and overall reliability. This comprehensive approach allowed for robust hyperparameter optimization and effective management of model lifecycle, crucial in achieving accurate and dependable predictions in various real-world applications. Graphical Abstract: [Figure not available: see fulltext.]. 2023, The Author(s).
- Source
- AAPS PharmSciTech, Vol-24, No. 8
- Date
- 2023-01-01
- Publisher
- Springer Science and Business Media Deutschland GmbH
- Subject
- Artificial Neural Network Multilayer Perceptron (ANN-MLP); data-driven modelling; design of experiments (DoE); hyperparameter optimization; model generalizability; model lifecycle management; model transferability; near infrared (NIR); process monitoring; statistical process control (SPC); target drift detection
- Coverage
- Ali H., Christ (Deemed to Be University), Karnataka, Bangalore, 560029, India; Muthudoss P., A2Z4.0 Research and Analytics Private Limited, Tamilnadu, Chennai, 600062, India, NuAxon Bioscience Inc., Bloomington, 47401-6301, IN, United States, School of Pharmaceutical Sciences, Vels Institute of Science, Technology & Advanced Studies (VISTAS), Velan Nagar P.V. Vaithiyalingam Road Pallavaram 600117, Tamilnadu, Chennai, India; Chauhan C., The Machine Learning Company, Pune, India; Kaliappan I., School of Pharmacy, Hindustan Institute of Technology and Science (HITS), Padur, Tamilnadu, Chennai, 603 103, India; Kumar D., Department of Pharmaceutical Engineering & Technology, IIT (BHU), Uttar Pradesh, Varanasi, 221011, India; Paudel A., Research Center Pharmaceutical Engineering GmbH (RCPE), Inffeldgasse 13, Graz, 8010, Austria, Graz University of Technology, Institute of Process and Particle Engineering, Inffeldgasse 13/3, Graz, 8010, Austria; Ramasamy G., Christ (Deemed to Be University), Karnataka, Bangalore, 560029, India
- Rights
- All Open Access; Hybrid Gold Open Access
- Relation
- ISSN: 15309932; PubMed ID: 38062329
- Format
- Online
- Language
- English
- Type
- Article
Collection
Citation
Ali H.; Muthudoss P.; Chauhan C.; Kaliappan I.; Kumar D.; Paudel A.; Ramasamy G., “Machine Learning-Enabled NIR Spectroscopy. Part 3: Hyperparameter by Design (HyD) Based ANN-MLP Optimization, Model Generalizability, and Model Transferability,” CHRIST (Deemed To Be University) Institutional Repository, accessed February 27, 2025, https://archives.christuniversity.in/items/show/13906.