Abstract
Pharmaceutical development constitutes a highly complex process with several steps that need to be completed for bringing a new safe and effective drug into the market. This process, that takes years and requires a significant financial investment, is associated with high risk, since most of the investigated drugs fail to successfully complete the clinical evaluation stage and gain marketing authorization. The recent advances in the area of artificial intelligence (AI), including machine learning, fostered the improvement of computational methods meant to assist scientists in performing many laborious processes, thus reducing cost and time of laboratory work. Specifically, the progress achieved in the area of deep learning regarding the efficient training of multilayered architectures of neural networks powered numerous solutions dealing with nonlinearities in the data. This chapter evolves around two axes. First, to present an overview of the application and contribution of AI in the various drug discovery and development stages. Second, to propose a conceptual framework meant to serve as a systematic way for the analysis of AI-based approaches. The proposed framework can be used as a guideline for an end-to-end profiling of the studies of interest. This is conducted by considering the most important modeling dimensions spanning from the input data up to the respective output. The chapter concludes with a series of observations about the current situation in terms of AI utilization in pharmaceutical development, putting at the same time under the spotlight a series of limitations and challenges.
Original language | English |
---|---|
Title of host publication | Novel Formulations and Future Trends |
Subtitle of host publication | Recent and Future Trends in Pharmaceutics, Volume 3 |
Publisher | Elsevier |
Pages | 415-451 |
Number of pages | 37 |
Volume | 3 |
ISBN (Electronic) | 9780323918169 |
ISBN (Print) | 9780323972451 |
DOIs | |
Publication status | Published - 1 Jan 2024 |
Keywords
- artificial intelligence
- deep learning
- drug development
- Drug discovery
- drug screening
- effectiveness prediction
- formulation development
- machine learning
- neural networks
- toxicity prediction