Artificial Intelligence (AI) and biotech are technologies that have been developing in parallel at an exponential rate.  Trillions of data points and exabytes of information are being generated by many thousands of research scientists all over the world, and the task of extruding meaningful conclusions which can move us forward scientifically has become increasingly and massively more complex. 

Happily, the rapid evolution of AI and ML has meant that, increasingly we are able to address that complexity. This is yet another area where science fact increasingly really does look like science fiction and where what we are capable of doing or aspiring to do is truly extraordinary.

Transforming drug discovery 

AI and ML can be used to improve drug discovery and drug design, to optimize clinical trials before they begin, to repurpose existing drug molecules, to hone manufacturing processes, to speed up diagnosis from medical imaging and a great deal else besides. Much of this can and will strip significant cost out of the industry, too.

The British company Exscientia is one of the leaders in the field. As former CEO Andrew Hopkins, explains, AI gives the company “the ability to search a much vaster chemical space than traditional processes could hope to handle.” AI can look for patterns that are too complicated for a human to recognize. The same can be true when looking at medical images. 

Various computer-aided detection (CAD) systems have been approved by the FDA already in this area. They can be used in conjunction with mammography to help radiologists detect breast cancer, for example. Such systems use AI algorithms to analyze images and highlight areas that may be indicative of cancer. 

Similar approaches have been approved for use in automated retinal image analysis to screen for diabetic retinopathy and refer patients to ophthalmologists for further evaluation if required, to interrogate CT scans to help clinicians work out whether a lung nodule is likely to be cancerous or benign, and to detect and classify brain hemorrhages.

AI-powered drug repurposing

Another exciting application for the technology is in repurposing existing compounds. Precision Life is another highly innovative British company in the field. In June 2022, it published research in the journal Cell Patterns which analyzed the drug pipelines of 177 biopharma companies using ML.

 Its goal was to find other potential uses for existing drugs. This broad approach had yielded some success when addressing coronavirus during the pandemic, for example. As Precision Life’s CEO, Dr. Steve Gardner, explained at the time, “The successful global effort to find effective interventions for the most severe COVID-19 patients demonstrated the value of reusing current drugs in new indications.” 

Precision Life and others like it wanted to go further and undertake a similar analysis for a much wider range of drugs and diseases. If we can find other uses for existing drugs, this could save the biotech industry and healthcare systems many billions of dollars and, according to Gardner, “provide a faster, cheaper and derisked route to the approval of new therapies, with major benefits to patients.” 

Using sophisticated “combinatorial analytics” from a huge amount of data, Precision Life’s work has identified 477 “repositioning opportunities” across no fewer than 35 diseases. 

Big tech’s foray into AI / biotech convergence

It isn’t just relatively new, small, innovative companies working in this area either. Apple, Google, IBM, Microsoft and plenty of other large companies are also significant players. Google Health is particularly focused on genomic analysis and AI-enabled imaging and diagnostics. Microsoft has a multi-year strategic alliance with companies such as Novartis and with the British gene and cell therapy company Oxford Biomedica.

Apple would like to leverage more than 1.5 billion iPhone users globally to collect and analyze medical data which could be of enormous value to researchers everywhere in the years ahead. Apple has also shipped more than 200 million Apple Watches since they first came to market and is fairly ambitious about the technology it would like to embed in that product in future when it comes to healthcare. 

Apple and other firms like it will drive growth in various wearable technologies in the years to come which will provide even more useful, rich data for any number of biotech industry initiatives using AI. All of this activity is another example of the importance of “convergence” as a theme overall.

A bold biotech future

AI and ML will have an exciting role to play in the future development of the biotech industry. Many of those involved in the research believe that eventually it may even be possible to develop drugs “100 percent in silico” – that is to say, by just using computers.Exscientia puts it on the landing page of its website: “In the future all drugs will be designed with AI.”

Another possible outcome from the realms of science fiction, still some years away, could be to do away with the need for animal testing – once we have sufficient confidence in the accuracy of the computer models. A 2019 study using a database of 10,000 chemicals and 800,000 studies was able to beat animal testing at predicting toxicity already, for example. AI can already be used today to reduce the number of animals needed for testing by screening molecules in silico and optimizing the design of experiments as a result.

There is also the promise of increasingly personalized medicine. In the not-too-distant future, it may be possible for a diagnostic technology to take a sample from a patient (saliva, blood or stool, for example), draw an extraordinarily accurate conclusion about their condition very quickly, and then even design and produce a genetically and biologically personalized treatment on the spot, or within a few hours or days, as the technology develops.