How AI is saving lives

How AI is saving lives

George Thaw, CEO of Fuel3D

During the last few years, artificial intelligence (AI) and machine learning have taken a huge leap forwards in transforming healthcare. From diagnosing and treating patients to emerging applications in drug research and discovery, the potential is truly vast.

McKinsey[1] estimates that big data and machine learning in pharma and medicine could generate a value of up to $100B annually, based on better decision-making, optimised innovation, improved efficiency of research/clinical trials, and new tool creation for physicians, consumers, insurers, and regulators.

Much in the same way that robots are on the verge of revolutionising manufacturing, AI is making its mark in healthcare by automating repetitive tasks. One such area is using AI to scour through huge amounts of medical imagery. By leveraging the power of deep learning, algorithms can be trained to distinguish anomalies on MRI, CT and advanced 3D scans, more quickly and accurately than the human eye.

Right here in Oxford, lots of pioneering AI applications for interpreting scans are taking place that could change the face of healthcare on a global scale in the not too distant future.

Researchers at Oxford’s John Radcliffe Hospital have created forms of ‘narrow’ artificial intelligence (AI developed to do a specific task), which can diagnose heart disease and lung cancer when present in scans. Data collected from an initial wave of clinical trials suggests that the technology greatly outperforms cardiologists because it can pick up on vital details doctors are unable to see and at less advanced stage of the disease. Further successful trials could result in a nationwide roll-out across the NHS later this year.

Medical imaging company Brainomix which was spun out of Oxford University in 2012 is using AI for the fast diagnosis and treatment of stroke victims. Its software looks for signs of stroke in pictures captured by CT scans. It’s already being used by the NHS to help doctors to make fast decisions so that treatment can be started sooner. Given that every minute counts, the impact for the patient is significant and life changing.

Over here at Fuel3D we’re using AI and machine learning to help speed up drug discovery in cancer research. The journey from lab to patient is a long and costly one and typically takes 12 years. Only five in 5,000 drugs that begin pre-clinical testing ever make it to human testing and just one of these five will ever be approved for human use[2]. Success rates are in decline in pharmaceutical R&D and big data could provide the cure. Last year, the 12 largest bio-pharma companies generated a mere 3.2% return from their drug-research arms compared to 10.1% in 2010. It is now costing these companies almost $2bn to bring a drug to market[3].

This has fuelled us to look at how we can use AI to improve drug discovery in oncology at the earliest stages when the risk of failure is at its highest. Our research team and data analysts are analysing thousands of 3D subcutaneous tumour scans from in-vivo studies. The hope is that our technology platform will learn to spot the earliest signs of success and failure in drug development, so that a decision can be taken more quickly on whether to accelerate or terminate a trial. Ultimately this will help the industry to bring crucial life saving drugs to market more quickly.

Science lives and dies by the strength of its data and so will AI and machine learning. The more data there is, the better the results it will yield, and the more effective its use will be in removing human error and improving the accuracy and speed of decisions in the lab and on the hospital ward.

Despite all the hype, AI and machine learning is not something new, nor is it just the latest fad. With faster computer processing and continued innovation I have no doubt that it will play an increasingly powerful helping hand in medicine and in the pharmaceutical industry so that more lives can be saved.


[1] McKinsey&Company – How big data can revolutionise pharmaceutical R&D

[2] California Biomedical Research Association – New Drug Development Process.

[3] Deloitte – Eighth Annual Pharmaceutical Innovation Study

No Comments

Post A Comment
Contact Us
Thanks for your message
We'll be in touch shortly