Darren A. Whitaker
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Darren A. Whitaker has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Marion McAfee
Atlantic Technological University Sligo Ireland
4 shared publications - Fraser Buchanan
School of Mechanical and Aerospace Engineering, Queens University Belfast, Ashby Building, Stranmillis Road, BT9 5AH, Belfast, UK.
2 shared publications Mark Billham
2 shared publications- Nimra Munir
Centre for Mathematical Modelling and Intelligent Systems for Health and Environment (MISHE), Atlantic Technological University, ATU Sligo, Ash Lane, F91 YW50 Sligo, Ireland
2 shared publications - Catherine Brigid Potter
Synthesis and Solid State Pharmaceutical Centre (SSPC), Bernal Institute , University of Limerick , Limerick V94 T9PX , Ireland
1 shared publication - Colin C. Seaton
Solid State Pharmaceutical Cluster, Materials and Surface Science Institute, University of Limerick, Limerick, Ireland
1 shared publication
Latest Publications
- Open Access
Interpretable Machine Learning Methods for Monitoring Polymer Degradation in Extrusion of Polylactic Acid
This work investigates real-time monitoring of extrusion-induced degradation in different grades of PLA across a range of process conditions and machine set-ups. Data on machine settings together with in-process sensor data, including temperature, pressure, and near-infrared (NIR) spectra, are used as inputs to predict the molecular weight and mechanical properties of the product. Many soft sensor approaches based on complex spectral data are essentially ‘black-box’ in nature, which can limit industrial acceptability. Hence, the focus here is on identifying an optimal approach to developing interpretable models while achieving high predictive accuracy and robustness across different process settings. The performance of a Recursive Feature Elimination (RFE) approach was compared to more common dimension reduction and regression approaches including Partial Least Squares (PLS), iterative PLS (i-PLS), Principal Component Regression (PCR), ridge regression, Least Absolute Shrinkage and Selection Operator (LASSO), and Random Forest (RF). It is shown that for medical-grade PLA processed under moisture-controlled conditions, accurate prediction of molecular weight is possible over a wide range of process conditions and different machine settings (different nozzle types for downstream fibre spinning) with an RFE-RF algorithm. Similarly, for the prediction of yield stress, RFE-RF achieved excellent predictive performance, outperforming the other approaches in terms of simplicity, interpretability, and accuracy. The features selected by the RFE model provide important insights to the process. It was found that change in molecular weight was not an important factor affecting the mechanical properties of the PLA, which is primarily related to the pressure and temperature at the latter stages of the extrusion process. The temperature at the extruder exit was also the most important predictor of degradation of the polymer molecular weight, highlighting the importance of accurate melt temperature control in the process. RFE not only outperforms more established methods as a soft sensor method, but also has significant advantages in terms of computational efficiency, simplicity, and interpretability. RFE-based soft sensors are promising for better quality control in processing thermally sensitive polymers such as PLA, in particular demonstrating for the first time the ability to monitor molecular weight degradation during processing across various machine settings.
Open AccessInterpretable Machine Learning Methods for Monitoring Polymer Degradation in Extrusion of Polylactic Acid
This work investigates real-time monitoring of extrusion-induced degradation in different grades of PLA across a range of process conditions and machine set-ups. Data on machine settings together with in-process sensor data, including temperature, pressure, and near-infrared (NIR) spectra, are used as inputs to predict the molecular weight and mechanical properties of the product. Many soft sensor approaches based on complex spectral data are essentially ‘black-box’ in nature, which can limit industrial acceptability. Hence, the focus here is on identifying an optimal approach to developing interpretable models while achieving high predictive accuracy and robustness across different process settings. The performance of a Recursive Feature Elimination (RFE) approach was compared to more common dimension reduction and regression approaches including Partial Least Squares (PLS), iterative PLS (i-PLS), Principal Component Regression (PCR), ridge regression, Least Absolute Shrinkage and Selection Operator (LASSO), and Random Forest (RF). It is shown that for medical-grade PLA processed under moisture-controlled conditions, accurate prediction of molecular weight is possible over a wide range of process conditions and different machine settings (different nozzle types for downstream fibre spinning) with an RFE-RF algorithm. Similarly, for the prediction of yield stress, RFE-RF achieved excellent predictive performance, outperforming the other approaches in terms of simplicity, interpretability, and accuracy. The features selected by the RFE model provide important insights to the process. It was found that change in molecular weight was not an important factor affecting the mechanical properties of the PLA, which is primarily related to the pressure and temperature at the latter stages of the extrusion process. The temperature at the extruder exit was also the most important predictor of degradation of the polymer molecular weight, highlighting the importance of accurate melt temperature control in the process. RFE not only outperforms more established methods as a soft sensor method, but also has significant advantages in terms of computational efficiency, simplicity, and interpretability. RFE-based soft sensors are promising for better quality control in processing thermally sensitive polymers such as PLA, in particular demonstrating for the first time the ability to monitor molecular weight degradation during processing across various machine settings.
- Open Access
- 9 September 2021
- Published by MDPI AG in Pharmaceutics
- Vol. 13 (9) , 1432
Machine Learning for Process Monitoring and Control of Hot-Melt Extrusion: Current State of the Art and Future Directions
In the last few decades, hot-melt extrusion (HME) has emerged as a rapidly growing technology in the pharmaceutical industry, due to its various advantages over other fabrication routes for drug delivery systems. After the introduction of the ‘quality by design’ (QbD) approach by the Food and Drug Administration (FDA), many research studies have focused on implementing process analytical technology (PAT), including near-infrared (NIR), Raman, and UV–Vis, coupled with various machine learning algorithms, to monitor and control the HME process in real time. This review gives a comprehensive overview of the application of machine learning algorithms for HME processes, with a focus on pharmaceutical HME applications. The main current challenges in the application of machine learning algorithms for pharmaceutical processes are discussed, with potential future directions for the industry.
- 9 September 2021
- Published by MDPI AG in Pharmaceutics
- Vol. 13 (9) , 1432
Open AccessMachine Learning for Process Monitoring and Control of Hot-Melt Extrusion: Current State of the Art and Future Directions
In the last few decades, hot-melt extrusion (HME) has emerged as a rapidly growing technology in the pharmaceutical industry, due to its various advantages over other fabrication routes for drug delivery systems. After the introduction of the ‘quality by design’ (QbD) approach by the Food and Drug Administration (FDA), many research studies have focused on implementing process analytical technology (PAT), including near-infrared (NIR), Raman, and UV–Vis, coupled with various machine learning algorithms, to monitor and control the HME process in real time. This review gives a comprehensive overview of the application of machine learning algorithms for HME processes, with a focus on pharmaceutical HME applications. The main current challenges in the application of machine learning algorithms for pharmaceutical processes are discussed, with potential future directions for the industry.
- 1 August 2018
- Published by Elsevier in Chemometrics and Intelligent Laboratory Systems
- Vol. 179, 82-84
A simple algorithm for despiking Raman spectra
- 1 August 2018
- Published by Elsevier in Chemometrics and Intelligent Laboratory Systems
- Vol. 179, 82-84
A simple algorithm for despiking Raman spectra
- Open Access
- 1 May 2018
- Published by Elsevier in Polymer Testing
- Vol. 67, 392-398
A UV-Vis spectroscopic method for monitoring of additive particle properties during polymer compounding
- 1 May 2018
- Published by Elsevier in Polymer Testing
- Vol. 67, 392-398
Open AccessA UV-Vis spectroscopic method for monitoring of additive particle properties during polymer compounding
- Open Access
- 16 March 2018
- Published by American Chemical Society (ACS) in ChemRxiv
A Simple Algorithm for Despiking Raman Spectra
Raman Spectroscopy is a widely used analytical technique, favoured when molecular specificity with minimal sample preparation is required.The majority of Raman instruments use charge-coupled device (CCD) detectors, these are susceptible to cosmic rays and as such multiple spurious spikes can occur in the measurement. These spikes are problematic as they may hinder subsequent analysis, particularly if multivariate data analysis is required. In this work we present a new algorithm to remove these spikes from spectra after acquisition. Specifically we use calculation of modified Z scores to locate spikes followed by a simple moving average filter to remove them. The algorithm is very simple and its execution is essentially instantaneous, resulting in spike-free spectra with minimal distortion of actual Raman data. The presented algorithm represents an improvement on existing spike removal methods by utilising simple, easy to understand mathematical concepts, making it ideal for experts and non-experts alike.
- 16 March 2018
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessA Simple Algorithm for Despiking Raman Spectra
Raman Spectroscopy is a widely used analytical technique, favoured when molecular specificity with minimal sample preparation is required.The majority of Raman instruments use charge-coupled device (CCD) detectors, these are susceptible to cosmic rays and as such multiple spurious spikes can occur in the measurement. These spikes are problematic as they may hinder subsequent analysis, particularly if multivariate data analysis is required. In this work we present a new algorithm to remove these spikes from spectra after acquisition. Specifically we use calculation of modified Z scores to locate spikes followed by a simple moving average filter to remove them. The algorithm is very simple and its execution is essentially instantaneous, resulting in spike-free spectra with minimal distortion of actual Raman data. The presented algorithm represents an improvement on existing spike removal methods by utilising simple, easy to understand mathematical concepts, making it ideal for experts and non-experts alike.
- 1 October 2017
- Published by Elsevier in European Journal of Pharmaceutics and Biopharmaceutics
- Vol. 119, 36-46
Investigation of polymorphic transitions of piracetam induced during wet granulation
- 1 October 2017
- Published by Elsevier in European Journal of Pharmaceutics and Biopharmaceutics
- Vol. 119, 36-46
Investigation of polymorphic transitions of piracetam induced during wet granulation
- 13 November 2012
- Published by Royal Society of Chemistry (RSC) in CrystEngComm
- Vol. 15 (12) , 2241-2250
Building multi-component crystals from cations and co-crystals: the use of chaperones
Ternary crystalline complexes consisting of both salts and ionic co-crystals have been created through the crystallisation of the binary co-crystal 3,5-dinitrobenzoic acid–4-(dimethylamino)benzoic acid with group 1 or ammonium cations. The size and charge density of the cation can be used to adjust the protonation level and local geometry of the acid pair. The selectivity and coordination geometry of the chaperone cation may be further adjusted by the inclusion of a crown ether to reduce the number and location of potential binding sites.
- 13 November 2012
- Published by Royal Society of Chemistry (RSC) in CrystEngComm
- Vol. 15 (12) , 2241-2250
Building multi-component crystals from cations and co-crystals: the use of chaperones
Ternary crystalline complexes consisting of both salts and ionic co-crystals have been created through the crystallisation of the binary co-crystal 3,5-dinitrobenzoic acid–4-(dimethylamino)benzoic acid with group 1 or ammonium cations. The size and charge density of the cation can be used to adjust the protonation level and local geometry of the acid pair. The selectivity and coordination geometry of the chaperone cation may be further adjusted by the inclusion of a crown ether to reduce the number and location of potential binding sites.