Artificial intelligence generates novel 3D printing formulations
Source abstract
Abstract
Formulation development is a critical step in the development of medicines. The process requires human creativity, ingenuity and in-depth knowledge of formulation development and processing optimization, which can be time-consuming. Herein, we tested the ability of artificial intelligence (AI) to create de novo formulations for three-dimensional (3D) printing. Specifically, conditional generative adversarial networks (cGANs), which are generative models known for their creativity, were trained on a dataset consisting of 1437 fused deposition modelling (FDM) printed formulations that were extracted from both the literature and in-house data. In total, 27 different cGANs architectures were explored with varying learning rate, batch size and number of hidden layers parameters to generate 270 formulations. After a comparison between the characteristics of AI-generated and human-generated formulations, it was discovered that cGANs with a medium learning rate (10−4) could strike a balance in generating formulations that are both novel and realistic. Four of these formulations were fabricated using an FDM printer, of which the first AI-generated formulation was successfully printed. Our study represents a milestone, highlighting the capacity of AI to undertake creative tasks and its potential to revolutionize the drug development process.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: 3D Printing in Biomedical Research · Cell Image Analysis Techniques · Innovative Microfluidic and Catalytic Techniques Innovation
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (80)
Show source-provided references
1. 3D printing in personalized medicines: A focus on applications of the technology10.1016/j.mtcomm.2023.105875 ↗
2. 3D printing for chemical, pharmaceutical and biological applications10.1038/s41570-018-0058-y ↗
3. 3D printing as a transformative tool for microneedle systems: Recent advances, manufacturing considerations and market potential10.1016/j.addr.2021.03.007 ↗
4. Looking into the future: toward advanced 3D biomaterials for stem-Cell-based regenerative medicine10.1002/adma.201705388 ↗
5. McCloskey · 2023 · 16 · 613
6. Recent advances in enabling technologies in 3D printing for precision medicine10.1002/adma.201902516 ↗
7. The applications of 3D printing in wound healing: the external delivery of stem cells and antibiosis10.1016/j.addr.2023.114823 ↗
8. 3D printing as a promising tool in personalized medicine10.1208/s12249-020-01905-8 ↗
9. Complex formulations, simple techniques: Can 3D printing technology be the Midas touch in pharmaceutical industry?10.1016/j.ajps.2018.11.008 ↗
10. 3D printing promotes the development of drugs10.1016/j.biopha.2020.110644 ↗
11. 3D printing of oral drugs: a new reality or hype?10.1080/17425247.2017.1371698 ↗
12. Integration of 3D printing with dosage forms: A new perspective for modern healthcare10.1016/j.biopha.2018.07.167 ↗
13. Harnessing artificial intelligence for the next generation of 3D printed medicines10.1016/j.addr.2021.05.015 ↗
14. Advancing pharmacy and healthcare with virtual digital technologies10.1016/j.addr.2021.114098 ↗
15. Revolutionizing Drug Formulation Development: The Increasing Impact of Machine Learning10.1016/j.addr.2023.115108 ↗
16. Algorithmic fairness in artificial intelligence for medicine and healthcare10.1038/s41551-023-01056-8 ↗
17. Artificial intelligence for natural product drug discovery
18. Metallurgy, mechanistic models and machine learning in metal printing10.1038/s41578-020-00236-1 ↗
19. Nanoparticle synthesis assisted by machine learning10.1038/s41578-021-00337-5 ↗
20. Towards artificial intelligence-enabled extracellular vesicle precision drug delivery10.1016/j.addr.2023.114974 ↗
21. The significance of artificial intelligence in drug delivery system design10.1016/j.addr.2019.05.001 ↗
22. Machine learning directed drug formulation development10.1016/j.addr.2021.05.016 ↗
23. Machine learning using multi-modal data predicts the production of selective laser sintered 3D printed drug products10.1016/j.ijpharm.2023.122628 ↗
24. Machine learning and machine vision accelerate 3D printed orodispersible film development10.3390/pharmaceutics13122187 ↗
25. Machine learning predicts electrospray particle size10.1016/j.matdes.2022.110735 ↗
26. Machine learning methods for prediction of food effects on bioavailability: A comparison of support vector machines and artificial neural networks10.1016/j.ejps.2021.106018 ↗
27. Machine learning-based quantitative prediction of drug exposure in drug-drug interactions using drug label information10.1038/s41746-022-00639-0 ↗
28. Real-time coating thickness measurement and defect recognition of film coated tablets with machine vision and deep learning10.1016/j.ijpharm.2022.121957 ↗
29. Image-based simultaneous particle size distribution and concentration measurement of powder blend components with deep learning and machine vision10.1016/j.ejps.2023.106611 ↗
30. Prediction of enhanced drug solubility related to clathrate compositions and operating conditions: Machine learning study10.1016/j.ijpharm.2023.123458 ↗
31. Advancing oral delivery of biologics: Machine learning predicts peptide stability in the gastrointestinal tract10.1016/j.ijpharm.2023.122643 ↗
32. Application of unsupervised and supervised learning to a material attribute database of tablets produced at two different granulation scales10.1016/j.ijpharm.2023.123066 ↗
33. Shifting machine learning for healthcare from development to deployment and from models to data10.1038/s41551-022-00898-y ↗
34. Revealing invisible cell phenotypes with conditional generative modeling10.1038/s41467-023-42124-6 ↗
35. High-content image generation for drug discovery using generative adversarial networks10.1016/j.neunet.2020.09.007 ↗
36. Feedback GAN for DNA optimizes protein functions10.1038/s42256-019-0017-4 ↗
37. De novo generation of hit-like molecules from gene expression signatures using artificial intelligence10.1038/s41467-019-13807-w ↗
38. Spectrum of advancements and developments in multidisciplinary domains for generative adversarial networks (GANs)
39. GANs for medical image analysis10.1016/j.artmed.2020.101938 ↗
40. Deep learning for synthetic microstructure generation in a materials-by-design framework for heterogeneous energetic materials10.1038/s41598-020-70149-0 ↗
41. Into the Unknown: How Computation Can Help Explore Uncharted Material Space10.1021/jacs.2c06833 ↗
42. Machine-learning-assisted de novo design of organic molecules and polymers: Opportunities and challenges10.3390/polym12010163 ↗
43. ProGAN: Protein solubility generative adversarial nets for data augmentation in DNN framework10.1016/j.compchemeng.2019.106533 ↗
44. Will artificial intelligence for drug discovery impact clinical pharmacology?10.1002/cpt.1795 ↗
45. druGAN: an advanced generative adversarial autoencoder model for de novo generation of new molecules with desired molecular properties in silico10.1021/acs.molpharmaceut.7b00346 ↗
46. Potential 2019-nCoV 3C-like protease inhibitors designed using generative deep learning approaches
47. Cell morphology-guided de novo hit design by conditioning GANs on phenotypic image features10.1039/D2DD00081D ↗
48. Shape-based generative modeling for de novo drug design10.1021/acs.jcim.8b00706 ↗
49. Deep generative molecular design reshapes drug discovery10.1016/j.xcrm.2022.100794 ↗
50. Generating 3D molecules conditional on receptor binding sites with deep generative models10.1039/D1SC05976A ↗
51. De novo protein design for novel folds using guided conditional Wasserstein generative adversarial networks10.1021/acs.jcim.0c00593 ↗
52. Generative adversarial networks (GAN) based efficient sampling of chemical composition space for inverse design of inorganic materials10.1038/s41524-020-00352-0 ↗
53. Generative deep neural networks for inverse materials design using backpropagation and active learning10.1002/advs.201902607 ↗
54. Assessing deep generative models in chemical composition space10.1021/acs.chemmater.2c01860 ↗
55. Inverse design of two-dimensional graphene/h-BN hybrids by a regressional and conditional GAN10.1016/j.carbon.2020.07.013 ↗
56. Deep convolutional generative adversarial network with semi-supervised learning enabled physics elucidation for extended gear fault diagnosis under data limitations10.1016/j.ymssp.2022.109772 ↗
57. Tackling mode collapse in multi-generator GANs with orthogonal vectors10.1016/j.patcog.2020.107646 ↗
58. Synthetic demand data generation for individual electricity consumers: Generative Adversarial Networks (GANs)10.1016/j.egyai.2022.100161 ↗
59. Generative adversarial networks (GANs) challenges, solutions, and future directions10.1145/3446374 ↗
60. Effective dynamics of generative adversarial networks10.1103/PhysRevX.13.041004 ↗
61. 3D printed oral theophylline doses with innovative ‘radiator-like’ design: Impact of polyethylene oxide (PEO) molecular weight10.1016/j.ijpharm.2019.04.017 ↗
62. 3D printing tablets: Predicting printability and drug dissolution from rheological data10.1016/j.ijpharm.2020.119868 ↗
63. 2019
64. Creativity—the unconscious foundations of the incubation period10.3389/fnhum.2014.00215 ↗
65. Active machine learning for formulation of precision probiotics10.1016/j.ijpharm.2022.121568 ↗
66. Semantically consistent text to fashion image synthesis with an enhanced attentional generative adversarial network10.1016/j.patrec.2020.02.030 ↗
67. Choshen, L., et al. (2019). “On the weaknesses of reinforcement learning for neural machine translation.” arXiv preprint arXiv:1907.01752.
68. Pros and cons of gan evaluation measures10.1016/j.cviu.2018.10.009 ↗
69. Generative adversarial networks10.1145/3422622 ↗
70. Wasserstein Generative Adversarial Networks
71. Shi, M., et al. (2023). "Cutout with patch-loss augmentation for improving generative adversarial networks against instability." Computer vision and image understanding: 103761.10.1016/j.cviu.2023.103761 ↗
72. CRPGAN: Learning image-to-image translation of two unpaired images by cross-attention mechanism and parallelization strategy10.1371/journal.pone.0280073 ↗
73. An adversarial model for electromechanical actuator fault diagnosis under nonideal data conditions
74. PET/CT imaging of 3D printed devices in the gastrointestinal tract of rodents10.1016/j.ijpharm.2017.11.055 ↗
75. 3D printed, personalized sustained release cortisol for patients with adrenal insufficiency10.1016/j.ijpharm.2022.122466 ↗
76. Mitigating bias in AI at the point of care10.1126/science.adh2713 ↗
77. Artificial intelligence, bias and clinical safety10.1136/bmjqs-2018-008370 ↗
78. A comprehensive survey and analysis of generative models in machine learning
79. Machine learning predicts 3D printing performance of over 900 drug delivery systems10.1016/j.jconrel.2021.07.046 ↗
80. M3DISEEN: A novel machine learning approach for predicting the 3D printability of medicines10.1016/j.ijpharm.2020.119837 ↗