When former Google employee Blake Lemoine recently made claims that the tech giant was secretly harbouring an artificial intelligence that was showing signs of being self-aware, half the world ignored the claims, brushing them off as the bitterness of a disgruntled employee, the other half took to the conspiracy forums.
AI in healthcare. Credit: metamorworks / Shutterstock
Credit: metamorworks / Shutterstock
Whether or not Lemoine’s claims are true – and if they are (big if) then it’s truly groundbreaking – what it does show is that AI has not yet won full public trust. Not surprising considering the number of cautionary tales that Hollywood has produced over the years.
Nevertheless, AI is increasingly being adopted by a sector that people place their trust in every day – healthcare. From bespoke medical implants to wearable tech, to cancer treatment, to drug discovery, the applications for artificial intelligence in the sector are growing fast. Here, we look at some of the main developments that AI has brought to healthcare over the past few months and years.
Drug discovery
Traditionally, drug discovery and the process of bringing it to market is a lengthy process, often taking more than a decade and at an average cost of $2.6 billion per medicine according to the Tufts Center for the Study of Drug Development.
AI is able to study chemicals in ways that humans simply cannot, automatically predicting complex protein structures and designing new drugs, despite actually having no real scientific understanding.
The tech also has the power to democratise drug design, as it is no longer limited to the major pharmaceutical companies. With the right algorithms, startups can invent thousands of new molecules in just a few hours. Using machine learning, parts of the design process are being automated, helping scientists develop new drugs for notoriously difficult diseases such as cancer, HIV/AIDS and Alzheimer’s, as well as rare genetic conditions.
In April 2021, Oxford-based Exscientia began Phase I trials using its experimental Alzheimer’s treatment, becoming the first company to use an AI-designed molecule in human trials.
The following month, the company entered into an agreement – potentially worth over $1.2 billion – with New York-based pharmaceutical company Bristol-Myers Squibb to accelerate drug discovery in a number of therapeutic areas including immunology and oncology. The same month it also announced its AI had discovered two more molecules, which are also undergoing trials
Read more: Alphabet launches Isomorphic Labs to boost AI drug discovery
In November of the same year, Google's parent company Alphabet Inc. decided to get in on the act and launched Isomorphic Labs, a new artificial intelligence company specialising in the discovery of new drugs.
The company leverages technology developed by its sister company DeepMind to, in its own words “accelerate drug discovery, and ultimately, find cures for some of humanity’s most devastating diseases".
DeepMind made waves last year when it revealed how its AlphaFold2 technology can be used to predict the shape of each and every protein in the human body with near-total accuracy.
All of the AI-developed potential drugs are, however, still in the earliest stages of human trials, and it will take some years to ascertain whether they work or not, while medical regulators have yet to approve an AI-designed drug, at the time of writing.
Nonetheless, pharmaceutical investors are putting billions of dollars into AI and machine learning-based research in recent years, with many believing these technologies are set to become the next frontier in drug development and medical innovation.
Robotics
While the use of robotics in surgery first dates back to the 1980s the use of AI is relatively recent, though its impact has been huge. Robotic surgery is typically less invasive and enables human surgeons to perform complex operations with greater control and precision compared to conventional methods. Statistically, it also leads to fewer complications.
Hugo is a customisable robotic platform designed for non-invasive surgeries. Credit: Medtronic
In October 2021, Ireland-based US healthcare company Medtronic revealed its surgery assistant robot, Hugo, had received clearance for use in urologic and gynecologic surgeries, including hysterectomies, in several European countries.
Robotics is even being used by the National Health Service (NHS) in the UK, a country which has often been accused of being sceptical of automation. In February 2020, the Versius surgical robot carried out its first NHS treatments.
The first procedure was carried out in Edinburgh, followed shortly after by another at Milton Keynes University Hospital. Both the procedures were colorectal operations to treat patients with severe bowel disease or bowel cancer.
Read more: Surgical robot undertakes first NHS work in UK
In May 2022, Professor Mike Reed and Dr Justin Green from the Northumbria Healthcare NHS Trust became the first surgeons in the country to utilise data science and machine learning to create bespoke surgical risk profiles. These profiles aim to allow those on waiting lists to make a more informed decision about potential risks when considering whether to go ahead with a particular surgery or not.
But it’s not just surgery where robots could revolutionise healthcare. In January 2021, Sophia, an AI-controlled "social robot" created by Hong Kong’s Hanson Robotics was being prepared for mass rollout during the Covid-19 pandemic with the hope it could look after the vulnerable and elderly in the difficult time, without the risk of getting infected itself.
A new technology called Animus from the UK’s Cyberselves, also has applications in the healthcare system, enabling people who are far from hospitalised loved ones to make a “visit” using a combination of robotics and virtual reality.
Detecting diseases
One especially interesting and potentially life-saving development is the increased use of AI in early disease detection. In recent years the use of AI in cancer detection, in particular, has been gaining both interest and investment because of its accuracy in predicting tumour regrowth, detecting breast cancer during screening, cutting hospital error rates and reducing result waiting times.
In May 2021, the NHS began using an AI system name CADU, which was developed by researchers at the University College London. The tech can help doctors detect oesophageal cancer – one of the deadliest forms of cancer - during its early stages when it is still treatable.
In April, Japan’s SoftBank investment group led a $63 million investment in AI Medical Service Inc, a company which makes AI software that can aid in the diagnosis of stomach cancers.
Gastrointestinal cancers currently account for one in four cancer cases and one in three cancer deaths globally, making them highly common and deadly, meaning the investment has the potential to save millions of lives every year.
Read more: SoftBank invests €59m in AI cancer-detecting technology
But it’s not only cancer, in January 2022, a new robotic system designed by the Institute of Bioorganic Chemistry of the Polish Academy of Sciences was also said to speed up Covid-19 diagnoses.
Agamede, as it was named, is able to process as many as 15,000 samples per day, according to the parameters entered by the programmer.
This June, Alder Hey Children's Hospital in Liverpool with Microsoft and Mindwave unveiled a "hospital without walls". The digital platform provides a "hybrid point of access" for patients and aims to revolutionise how care is delivered to young people through both digital and physical care.
The tech was developed with the main goal of tackling the problem of hospital capacity and limited community resources and providing care to the most vulnerable people.
Quality of life
AI and robotics also have the potential to improve the lives of millions of people with debilitating diseases or physical health problems.
Advances in powered medical devices are helping to mobilise patients with less physical support. For example, exoskeletons, which were originally created to prevent strain in factory workers, are becoming increasingly common in the healthcare sector to help patients regain movement in rehabilitation therapies.
Exoskeletons have controls, sensors and motors to mechanically move parts of the body, with an accompanying battery pack that is worn on the patient’s back. Some models require the use of crutches, but those without have to be bulkier to provide additional support.
Widex Moment
The Widex Moment hearing aid. Image credit: Widex
In January 2021, the Moment, a hearing aid made by Widex that uses artificial intelligence engines to improve real-time listening for people with hearing loss, was awarded the Tech for Change Award at CES 2021.
The Moment was chosen by the judges for having reduced the sound delay from 7-10 milliseconds to 0.5 milliseconds, significantly faster and providing the wearer with the most natural hearing experience to date.
Read more: AI-powered hearing aid wins Tech for Change Award at CES 2021
The following month a robotic glove based on AI technology received financial support from Edinburgh Business School’s Incubator and could soon be helping people recover muscle grip in their hands.
The glove is aimed at the millions of people who suffer from hand weakness because of muscle mass loss as they age or due to illnesses like Multiple Sclerosis, Motor Neurone Disease and carpal tunnel syndrome. It is able to aid in a wide range of day-to-day tasks including opening jars, driving and pouring a cup of tea.
Then in March 2022, 3D Systems and Enhatch partnered to 3D print bespoke medical devices and integrate them into the latter's smart surgery networks.
These implants will likely be utilised in cranial or orthopaedic surgeries, meaning potential reconstruction will not deviate too much from how the patient originally looked, will minimise the risk of implants not fitting, and could even increase its longevity, all while cutting the time taken for the procedure and minimising risk to the patient.
Conclusion
While many of these technologies are still in their nascent stages in the necessarily cautious healthcare sector, their potential is clear enough for investors to pour in billions, suggesting that they are confident of seeing a return on investment.
Indeed, Fortune Business Insights valued the medical devices market at $488.9 billion in 2021, and projected growth to as much as $718.9 billion by 2029.
And while there will almost certainly be some reticence among some of the public, as it becomes more common, this will most likely fade away. After all, saving a life is perhaps the ultimate trust-building exercise.
Back to Homepage
Back to Healthcare
Back to Technology & Innovation