echemi logo
Product
  • Product
  • Supplier
  • Inquiry
    Home > Active Ingredient News > Infection > How to break through the speed-limiting step of AI application in new drug research and development-clinical trials?

    How to break through the speed-limiting step of AI application in new drug research and development-clinical trials?

    • Last Update: 2021-06-20
    • Source: Internet
    • Author: User
    Search more information of high quality chemicals, good prices and reliable suppliers, visit www.echemi.com

    Although the development of modern molecular biology such as genomics, proteomics, and bioinformatics has brought great progress to the theory of drug research and development, due to the complexity of the biochemical reactions of drug molecules in the human body, the research and development of new drugs has not escaped the empirical color


    Traditional drug research and development are dominated by pharmacological experts.


    The advantages of AI technology in natural language processing, image recognition, deep learning, and cognitive computing can be applied to all aspects of new drug research and development


    Therefore, AI+ new drug R&D has become a hot spot in current pharmaceutical research and cutting-edge medical entrepreneurship


    Figure 1: The application of AI in all aspects of drug development

    Figure 1: The application of AI in all aspects of drug development

    Source: According to public information

    Source: According to public information

    01 The application of AI in clinical trials is seriously insufficient

    01 The application of AI in clinical trials is seriously insufficient

    There are more than ten links in drug development, but due to factors such as data availability, AI can only be applied to a few links at present


    Figure 2: Distribution of the number of cases of AI application in the segment of new drug research and development in recent years

    Figure 2: Distribution of the number of cases of AI application in the segment of new drug research and development in recent years

    Source: Speedstone Technology

    Source: Speedstone Technology

    02 The greatest value of AI-assisted new drug development: improving the success rate of clinical trials

    02 The greatest value of AI-assisted new drug development: improving the success rate of clinical trials

    In fact, the efficiency improvement or cost reduction in the clinical trial phase has a far greater impact on the investment in new drug research and development than the drug discovery phase


    Figure 3: Improve speed, improve quality, and reduce costs

    Figure 3: Improve speed, improve quality, and reduce costs

    Comparison of the degree of impact on drug R&D revenue

    Comparison of the degree of impact on drug R&D revenue

    Source: Drug Discovery Today

    Source: Drug Discovery Today

    Andreas Bender simulated three situations of accelerating drug discovery speed, reducing costs, and improving the success rate (improving quality) of each link of the whole process through AI technology (assuming that AI improves speed, cost and stage success rate by 20%), right The cost impact of a new drug successfully introduced to the market


    It can be seen from Figure 3 that improving the success rate (especially in all clinical stages) has the greatest impact on the value of the entire R&D project, which far exceeds the benefits of increasing the R&D speed and reducing costs at each stage


    03 Biological complexity limits the computational processing power of AI in clinical trials

    03 Biological complexity limits the computational processing power of AI in clinical trials

    AI applications are concentrated in the drug discovery link because this link is mainly based on chemical processes.


    However, the clinical trial phase is dominated by biological processes, and its complexity poses huge challenges in both data and AI modeling


    The current AI drug development model usually starts from simplicity, and rarely considers the complexity of biology from the beginning, so it often fails in clinical testing


    In addition, AI systems tend to simplify the model while ignoring other issues, such as whether the compound reaches its intended target, whether it can treat a certain phenotype of the disease, and whether its side effects are within an acceptable range, and so on


    For the AI ​​system to succeed, a clear "compound-target-phenotype" link needs to be established


    04 Summary

    04 Summary

    In a nutshell, limited by the complexity of biology and the lack of clinical databases, the application of AI in the field of drug research and development is mainly concentrated in the front-end drug discovery link, but the benefits that it brings to drug research and development are relatively limited


    Reference materials:

    Reference materials:

    1.


    1.


    2.
    Speedstone Technology "White Paper on AI-assisted Drug R&D Actions of 44 Top Pharmaceutical Companies in the World";

    3.
    Bender A.
    et al.
    Artificial intelligence in drug discovery: what is realistic, what are illusions? ——Drug Discovery Today 2020.
    12

    3.
    Bender A.
    et al.
    Artificial intelligence in drug discovery: what is realistic, what are illusions? ——Drug Discovery Today 2020.
    12
    This article is an English version of an article which is originally in the Chinese language on echemi.com and is provided for information purposes only. This website makes no representation or warranty of any kind, either expressed or implied, as to the accuracy, completeness ownership or reliability of the article or any translations thereof. If you have any concerns or complaints relating to the article, please send an email, providing a detailed description of the concern or complaint, to service@echemi.com. A staff member will contact you within 5 working days. Once verified, infringing content will be removed immediately.

    Related Articles

    Contact Us

    The source of this page with content of products and services is from Internet, which doesn't represent ECHEMI's opinion. If you have any queries, please write to service@echemi.com. It will be replied within 5 days.

    Moreover, if you find any instances of plagiarism from the page, please send email to service@echemi.com with relevant evidence.