Subject: Using artificial intelligence in the development of medicine - Comments and suggestions are welcome! Don't hesitate and leave a comment on our comment section down below the article!
Image Credit: ivabalk via pixabay - HDR tune by Universal-Sci
Artificial intelligence has been a topic of much discussion in recent years. Concerning the future of AI, there seems to be a consensus on at least one thing, and that is that we don’t really know what it has in store for us. On the one hand, it is sometimes seen as a potential threat to our way of life, but on the other hand, some people see its potential in solving the problems humanity is facing.
Read more: The future of AI: 'Computers shouldn't think like people'
In the world of medicine and chemistry, at least it seems that there are some helpful developments in progress. Scientists from the University of Cambridge have developed an algorithm that self improves through machine learning. The algorithm has been shown to foretell the results of complex chemical reactions with an accuracy of over 90%. It has been functioning so well that it even outperformed qualified chemists.
One of the most difficult aspects of designing a new type of medicine is that building blocks used to composite complex organic molecules often behave in unanticipated ways. It is essential to have those molecules function in a specifically intentioned way for medicine to work.
Dr. Alpha Lee from Cambridge's Cavendish Laboratory stated that our understanding of chemical reactions is incomplete. Machine learning algorithms can have a better understanding of chemistry as they extract patterns of reactivity from the data of millions of distributed chemical reactions. Humans are simply unable to do such a thing.
Image Credit: TheDigitalArtist via pixabay
The algorithm utilizes pattern recognition tools to understand how chemical groups in molecules react. The researchers looked at chemical reaction prediction as a machine translation problem. The reacting molecules are considered as one ‘language,’ while the product is regarded as a different language. The model then uses the patterns in the text to learn how to ‘translate’ between the two languages.
In addition to very precisely anticipating the correct outcome of unseen chemical reactions, the model also comprehends what it doesn't know. It produces an uncertainty score eliminating incorrect predictions with an accuracy of 89 percent. This uncertainty score can be useful in avoiding the waste of resources and time.
The model is an excellent addition to the tool set of scientists that are designing experiments. Although there is still a lot of work to do, there is a lot of potential in the use of artificial intelligence in the world of medicine and chemistry.
Sources and further reading:
If you enjoy our selection of content please consider following Universal-Sci on social media:
