1
|
Honti B, Farkas A, Nagy ZK, Pataki H, Nagy B. Explainable deep recurrent neural networks for the batch analysis of a pharmaceutical tableting process in the spirit of Pharma 4.0. Int J Pharm 2024; 662:124509. [PMID: 39048040 DOI: 10.1016/j.ijpharm.2024.124509] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/16/2024] [Revised: 07/19/2024] [Accepted: 07/21/2024] [Indexed: 07/27/2024]
Abstract
Due to the continuously increasing Cost of Goods Sold, the pharmaceutical industry has faced several challenges, and the Right First-Time principle with data-driven decision-making has become more pressing to sustain competitiveness. Thus, in this work, three different types of artificial neural network (ANN) models were developed, compared, and interpreted by analyzing an open-access dataset from a real pharmaceutical tableting production process. First, the multilayer perceptron (MLP) model was used to describe the total waste based on 20 raw material properties and 25 statistical descriptors of the time series data collected throughout the tableting (e.g., tableting speed and compression force). Then using 10 process time series data in addition to the raw material properties, the cumulative waste, during manufacturing was also predicted by long short-term memory (LSTM) and bidirectional LSTM (biLSTM) recurrent neural networks (RNN). The LSTM network was used to forecast the waste production profile to allow preventive actions. The results showed that RNNs were able to predict the waste trajectory, the best model resulting in 1096 and 2174 tablets training and testing root mean squared errors, respectively. For a better understanding of the process, and the models and to help the decision-support systems and control strategies, interpretation methods were implemented for all ANNs, which increased the process understanding by identifying the most influential material attributes and process parameters. The presented methodology is applicable to various critical quality attributes in several fields of pharmaceutics and therefore is a useful tool for realizing the Pharma 4.0 concept.
Collapse
Affiliation(s)
- Barbara Honti
- Department of Organic Chemistry and Technology, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary
| | - Attila Farkas
- Department of Organic Chemistry and Technology, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary
| | - Zsombor Kristóf Nagy
- Department of Organic Chemistry and Technology, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary
| | - Hajnalka Pataki
- Department of Organic Chemistry and Technology, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary
| | - Brigitta Nagy
- Department of Organic Chemistry and Technology, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary.
| |
Collapse
|
2
|
Gholap AD, Uddin MJ, Faiyazuddin M, Omri A, Gowri S, Khalid M. Advances in artificial intelligence for drug delivery and development: A comprehensive review. Comput Biol Med 2024; 178:108702. [PMID: 38878397 DOI: 10.1016/j.compbiomed.2024.108702] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/03/2024] [Revised: 05/12/2024] [Accepted: 06/01/2024] [Indexed: 07/24/2024]
Abstract
Artificial intelligence (AI) has emerged as a powerful tool to revolutionize the healthcare sector, including drug delivery and development. This review explores the current and future applications of AI in the pharmaceutical industry, focusing on drug delivery and development. It covers various aspects such as smart drug delivery networks, sensors, drug repurposing, statistical modeling, and simulation of biotechnological and biological systems. The integration of AI with nanotechnologies and nanomedicines is also examined. AI offers significant advancements in drug discovery by efficiently identifying compounds, validating drug targets, streamlining drug structures, and prioritizing response templates. Techniques like data mining, multitask learning, and high-throughput screening contribute to better drug discovery and development innovations. The review discusses AI applications in drug formulation and delivery, clinical trials, drug safety, and pharmacovigilance. It addresses regulatory considerations and challenges associated with AI in pharmaceuticals, including privacy, data security, and interpretability of AI models. The review concludes with future perspectives, highlighting emerging trends, addressing limitations and biases in AI models, and emphasizing the importance of collaboration and knowledge sharing. It provides a comprehensive overview of AI's potential to transform the pharmaceutical industry and improve patient care while identifying further research and development areas.
Collapse
Affiliation(s)
- Amol D Gholap
- Department of Pharmaceutics, St. John Institute of Pharmacy and Research, Palghar, Maharashtra, 401404, India.
| | - Md Jasim Uddin
- Department of Pharmaceutical Technology, Faculty of Pharmacy, Universiti Malaya, 50603, Kuala Lumpur, Malaysia.
| | - Md Faiyazuddin
- School of Pharmacy, Al-Karim University, Katihar, Bihar, 854106, India; Centre for Global Health Research, Saveetha Institute of Medical and Technical Sciences, Tamil Nadu, India.
| | - Abdelwahab Omri
- Department of Chemistry and Biochemistry, The Novel Drug and Vaccine Delivery Systems Facility, Laurentian University, Sudbury, ON, P3E 2C6, Canada.
| | - S Gowri
- PG & Research, Department of Physics, Cauvery College for Women, Tiruchirapalli, Tamil Nadu, 620018, India
| | - Mohammad Khalid
- James Watt School of Engineering, University of Glasgow, Glasgow G12 8QQ, UK; Sunway Centre for Electrochemical Energy and Sustainable Technology (SCEEST), School of Engineering and Technology, Sunway University, No. 5, Jalan Universiti, Bandar Sunway, 47500 Selangor Darul Ehsan, Malaysia; University Centre for Research and Development, Chandigarh University, Mohali, Punjab, 140413, India.
| |
Collapse
|
3
|
Sun S, Alkahtani ME, Gaisford S, Basit AW, Elbadawi M, Orlu M. Virtually Possible: Enhancing Quality Control of 3D-Printed Medicines with Machine Vision Trained on Photorealistic Images. Pharmaceutics 2023; 15:2630. [PMID: 38004607 PMCID: PMC10674815 DOI: 10.3390/pharmaceutics15112630] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/18/2023] [Revised: 11/01/2023] [Accepted: 11/14/2023] [Indexed: 11/26/2023] Open
Abstract
Three-dimensional (3D) printing is an advanced pharmaceutical manufacturing technology, and concerted efforts are underway to establish its applicability to various industries. However, for any technology to achieve widespread adoption, robustness and reliability are critical factors. Machine vision (MV), a subset of artificial intelligence (AI), has emerged as a powerful tool to replace human inspection with unprecedented speed and accuracy. Previous studies have demonstrated the potential of MV in pharmaceutical processes. However, training models using real images proves to be both costly and time consuming. In this study, we present an alternative approach, where synthetic images were used to train models to classify the quality of dosage forms. We generated 200 photorealistic virtual images that replicated 3D-printed dosage forms, where seven machine learning techniques (MLTs) were used to perform image classification. By exploring various MV pipelines, including image resizing and transformation, we achieved remarkable classification accuracies of 80.8%, 74.3%, and 75.5% for capsules, tablets, and films, respectively, for classifying stereolithography (SLA)-printed dosage forms. Additionally, we subjected the MLTs to rigorous stress tests, evaluating their scalability to classify over 3000 images and their ability to handle irrelevant images, where accuracies of 66.5% (capsules), 72.0% (tablets), and 70.9% (films) were obtained. Moreover, model confidence was also measured, and Brier scores ranged from 0.20 to 0.40. Our results demonstrate promising proof of concept that virtual images exhibit great potential for image classification of SLA-printed dosage forms. By using photorealistic virtual images, which are faster and cheaper to generate, we pave the way for accelerated, reliable, and sustainable AI model development to enhance the quality control of 3D-printed medicines.
Collapse
Affiliation(s)
- Siyuan Sun
- UCL School of Pharmacy, University College London, 29-39 Brunswick Square, London WC1N 1AX, UK; (S.S.); (M.E.A.); (S.G.)
| | - Manal E. Alkahtani
- UCL School of Pharmacy, University College London, 29-39 Brunswick Square, London WC1N 1AX, UK; (S.S.); (M.E.A.); (S.G.)
- Department of Pharmaceutics, College of Pharmacy, Prince Sattam bin Abdulaziz University, Alkharj 11942, Saudi Arabia
| | - Simon Gaisford
- UCL School of Pharmacy, University College London, 29-39 Brunswick Square, London WC1N 1AX, UK; (S.S.); (M.E.A.); (S.G.)
| | - Abdul W. Basit
- UCL School of Pharmacy, University College London, 29-39 Brunswick Square, London WC1N 1AX, UK; (S.S.); (M.E.A.); (S.G.)
| | - Moe Elbadawi
- UCL School of Pharmacy, University College London, 29-39 Brunswick Square, London WC1N 1AX, UK; (S.S.); (M.E.A.); (S.G.)
- School of Biological and Behavioural Sciences, Queen Mary University of London, Mile End Road, London E1 4DQ, UK
| | - Mine Orlu
- UCL School of Pharmacy, University College London, 29-39 Brunswick Square, London WC1N 1AX, UK; (S.S.); (M.E.A.); (S.G.)
| |
Collapse
|
4
|
Leeming R, Mahmud T, Roberts KJ, George N, Webb J, Simone E, Brown CJ. Development of a Digital Twin for the Prediction and Control of Supersaturation during Batch Cooling Crystallization. Ind Eng Chem Res 2023; 62:11067-11081. [PMID: 37484628 PMCID: PMC10360059 DOI: 10.1021/acs.iecr.3c00371] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2023] [Revised: 06/20/2023] [Accepted: 06/20/2023] [Indexed: 07/25/2023]
Abstract
Fine chemicals produced via batch crystallization with properties dependent on the crystal size distribution require precise control of supersaturation, which drives the evolution of crystal size over time. Model predictive control (MPC) of supersaturation using a mechanistic model to represent the behavior of a crystallization process requires less experimental time and resources compared with fully empirical model-based control methods. Experimental characterization of the hexamine-ethanol crystallization system was performed in order to collect the parameters required to build a one-dimensional (1D) population balance model (PBM) in gPROMS FormulatedProducts software (Siemens-PSE Ltd.). Analysis of the metastable zone width (MSZW) and a series of seeded batch cooling crystallizations informed the suitable process conditions selected for supersaturation control experiments. The gPROMS model was integrated with the control software PharmaMV (Perceptive Engineering Ltd.) to create a digital twin of the crystallizer. Simulated batch crystallizations were used to train two statistical MPC blocks, allowing for in silico supersaturation control simulations to develop an effective control strategy. In the supersaturation set-point range of 0.012-0.036, the digital twin displayed excellent performance that would require minimal controller tuning to steady out any instabilities. The MPC strategy was implemented on a physical 500 mL crystallizer, with the simulated solution concentration replaced by in situ measurements from calibrated attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy. Physical supersaturation control performance was slightly more unstable than the in silico tests, which is consistent with expected disturbances to the heat transfer, which were not specifically modeled in simulations. Overall, the level of supersaturation control in a real crystallizer was found to be accurate and precise enough to consider future adaptations to the MPC strategy for more advanced control objectives, such as the crystal size.
Collapse
Affiliation(s)
- Ryan Leeming
- School
of Chemical and Process Engineering, University
of Leeds, Leeds LS2 9JT, U.K.
| | - Tariq Mahmud
- School
of Chemical and Process Engineering, University
of Leeds, Leeds LS2 9JT, U.K.
| | - Kevin J. Roberts
- School
of Chemical and Process Engineering, University
of Leeds, Leeds LS2 9JT, U.K.
| | - Neil George
- Syngenta, Jealott’s Hill, Bracknell RG42 6EY, U.K.
| | | | - Elena Simone
- Department
of Applied Science and Technology, Politecnico
di Torino, Torino 10129, Italy
| | - Cameron J. Brown
- CMAC
Future Manufacturing Research Hub, University
of Strathclyde, Glasgow G1 1RD, U.K.
| |
Collapse
|
5
|
Wu G, Yion WTG, Dang KLNQ, Wu Z. Physics-Informed Machine Learning for MPC: Application to a Batch Crystallization Process. Chem Eng Res Des 2023. [DOI: 10.1016/j.cherd.2023.02.048] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 03/08/2023]
|
6
|
Zheng Y, Wu Z. Physics-Informed Online Machine Learning and Predictive Control of Nonlinear Processes with Parameter Uncertainty. Ind Eng Chem Res 2023. [DOI: 10.1021/acs.iecr.2c03691] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/31/2023]
Affiliation(s)
- Yingzhe Zheng
- Department of Chemical and Biomolecular Engineering, National University of Singapore, 117585, Singapore
| | - Zhe Wu
- Department of Chemical and Biomolecular Engineering, National University of Singapore, 117585, Singapore
| |
Collapse
|
7
|
Song B, Liu T, Yang S, Liu J, Chen J. Data-Driven Operation Modeling and Optimal Design for Batch Cooling Crystallization with a Case Study on β-LGA. Ind Eng Chem Res 2022. [DOI: 10.1021/acs.iecr.2c02937] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/23/2022]
Affiliation(s)
- Bo Song
- Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China
- Institute of Advanced Control Technology, Dalian University of Technology, Dalian 116024, China
| | - Tao Liu
- Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China
- Institute of Advanced Control Technology, Dalian University of Technology, Dalian 116024, China
| | - Siwei Yang
- Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China
- Institute of Advanced Control Technology, Dalian University of Technology, Dalian 116024, China
| | - Jingxiang Liu
- Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China
- School of Marine Electrical Engineering, Dalian Maritime University, Dalian 116026, China
| | - Junghui Chen
- Department of Chemical Engineering, Chung-Yuan Christian University, Chung-Li District, Taoyuan 32023, Taiwan
| |
Collapse
|
8
|
Advanced predictive control for GRU and LSTM networks. Inf Sci (N Y) 2022. [DOI: 10.1016/j.ins.2022.10.078] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
|
9
|
Zhao T, Zheng Y, Wu Z. Feature selection-based machine learning modeling for distributed model predictive control of nonlinear processes. Comput Chem Eng 2022. [DOI: 10.1016/j.compchemeng.2022.108074] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/27/2022]
|
10
|
Alshahrani SM, Saqr AA, Alfadhel MM, Alshetaili AS, Almutairy BK, Alsubaiyel AM, Almari AH, Alamoudi JA, Abourehab MAS. Application of CO2 Supercritical Fluid to Optimize the Solubility of Oxaprozin: Development of Novel Machine Learning Predictive Models. Molecules 2022; 27:molecules27185762. [PMID: 36144490 PMCID: PMC9506598 DOI: 10.3390/molecules27185762] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/01/2022] [Revised: 08/15/2022] [Accepted: 08/17/2022] [Indexed: 11/26/2022] Open
Abstract
Over the last years, extensive motivation has emerged towards the application of supercritical carbon dioxide (SCCO2) for particle engineering. SCCO2 has great potential for application as a green and eco-friendly technique to reach small crystalline particles with narrow particle size distribution. In this paper, an artificial intelligence (AI) method has been used as an efficient and versatile tool to predict and consequently optimize the solubility of oxaprozin in SCCO2 systems. Three learning methods, including multi-layer perceptron (MLP), Kriging or Gaussian process regression (GPR), and k-nearest neighbors (KNN) are selected to make models on the tiny dataset. The dataset includes 32 data points with two input parameters (temperature and pressure) and one output (solubility). The optimized models were tested with standard metrics. MLP, GPR, and KNN have error rates of 2.079 × 10−8, 2.173 × 10−9, and 1.372 × 10−8, respectively, using MSE metrics. Additionally, in terms of R-squared, they have scores of 0.868, 0.997, and 0.999, respectively. The optimal inputs are the same as the maximum possible values and are paired with a solubility of 1.26 × 10−3 as an output.
Collapse
Affiliation(s)
- Saad M. Alshahrani
- Department of Pharmaceutics, College of Pharmacy, Prince Sattam Bin Abdulaziz University, P.O. Box 173, Al-Kharj 11942, Saudi Arabia
- Correspondence: (S.M.A.); (A.M.A.); (M.A.S.A.)
| | - Ahmed Al Saqr
- Department of Pharmaceutics, College of Pharmacy, Prince Sattam Bin Abdulaziz University, P.O. Box 173, Al-Kharj 11942, Saudi Arabia
| | - Munerah M. Alfadhel
- Department of Pharmaceutics, College of Pharmacy, Prince Sattam Bin Abdulaziz University, P.O. Box 173, Al-Kharj 11942, Saudi Arabia
| | - Abdullah S. Alshetaili
- Department of Pharmaceutics, College of Pharmacy, Prince Sattam Bin Abdulaziz University, P.O. Box 173, Al-Kharj 11942, Saudi Arabia
| | - Bjad K. Almutairy
- Department of Pharmaceutics, College of Pharmacy, Prince Sattam Bin Abdulaziz University, P.O. Box 173, Al-Kharj 11942, Saudi Arabia
| | - Amal M. Alsubaiyel
- Department of Pharmaceutics, College of Pharmacy, Qassim University, Buraidah 52571, Saudi Arabia
- Correspondence: (S.M.A.); (A.M.A.); (M.A.S.A.)
| | - Ali H. Almari
- Department of Pharmaceutics, College of Pharmacy, King Khalid University, Abha 62529, Saudi Arabia
| | - Jawaher Abdullah Alamoudi
- Department of Pharmaceutical Sciences, College of Pharmacy, Princess Nourah bint Abdulrahman University, Riyadh 145111, Saudi Arabia
| | - Mohammed A. S. Abourehab
- Department of Pharmaceutics, Faculty of Pharmacy, Umm Al-Qura University, Makkah 21955, Saudi Arabia
- Department of Pharmaceutics and Industrial Pharmacy, College of Pharmacy, Minia University, Minia 61519, Egypt
- Correspondence: (S.M.A.); (A.M.A.); (M.A.S.A.)
| |
Collapse
|
11
|
Wang Z, Li J, Rangaiah GP, Wu Z. Machine learning aided multi-objective optimization and multi-criteria decision making: Framework and two applications in chemical engineering. Comput Chem Eng 2022. [DOI: 10.1016/j.compchemeng.2022.107945] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
|
12
|
|
13
|
Zheng Y, Zhao T, Wang X, Wu Z. Online Learning‐Based Predictive Control of Crystallization Processes under Batch‐to‐Batch Parametric Drift. AIChE J 2022. [DOI: 10.1002/aic.17815] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/10/2022]
Affiliation(s)
- Yingzhe Zheng
- Department of Chemical and Biomolecular Engineering National University of Singapore Singapore
| | - Tianyi Zhao
- Department of Chemical and Biomolecular Engineering National University of Singapore Singapore
- Joint School of National University of Singapore and Tianjin University, International Campus of Tianjin University, Binhai New City Fuzhou China
| | - Xiaonan Wang
- Department of Chemical and Biomolecular Engineering National University of Singapore Singapore
- Department of Chemical Engineering Tsinghua University Beijing China
| | - Zhe Wu
- Department of Chemical and Biomolecular Engineering National University of Singapore Singapore
| |
Collapse
|
14
|
Sitapure N, Kwon JSI. Neural network-based model predictive control for thin-film chemical deposition of quantum dots using data from a multiscale simulation. Chem Eng Res Des 2022. [DOI: 10.1016/j.cherd.2022.05.041] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/27/2022]
|