img

How AI is Rooting Out Diseases

Artificial Intelligence may once have been firmly rooted in the world of science fiction, but it is becoming more of a reality with every passing day – and one of the areas in which its uses are being most widely explored is medicine. At the Tej Kohli Cornea Institute we believe technology can be harnessed for the betterment of people everywhere, and we pay close attention to the latest developments in medical technology. Here, we explore the ways in which AI is being used to root out diseases and transform medicine for the better.

The speed with which AI technology has advanced has led to medical applications that could never have been predicted even a few years ago. ‘Deep learning’ algorithms have proved to be a particularly remarkable advancement in the field of medical AI. They have already been shown capable of diagnosing tuberculosis in chest x-ray images with an astonishing degree of accuracy, and are now beginning to be rolled out into several other branches of medicine.

Last month Kohli Ventures discussed how these algorithms could be used to improve and simplify the process of developing new cancer drugs, but they are also being used to help diagnose heart disease and certain types of cancer. Currently these diagnoses are carried out by specialists examining images of the body, whether from an x-ray or CT scan, but AI algorithms could potentially carry out the same task without the risk of human lapses in judgment and knowledge.

For our founder Tej Kohli news of these algorithm’s potential effect on the field of ophthalmology – the branch of medicine dealing with the human eye – has been of particularly significant interest. Google has been using its deep machine learning technology to diagnose diabetic retinopathy, an eye condition common among diabetics. By familiarising their deep learning machine with images of both healthy eyes and those suffering from retinopathy, the machine was then able to outperform human ophthalmologists when it came to actual diagnoses. This could prove to be a huge breakthrough for the 387 million diabetes sufferers across the globe who need readily available access to a specialist ophthalmologist to identify diabetic retinopathy as early as possible.

It is also not just humans who look set to benefit from major steps forward in medical AI technology – researchers have recently developed a new AI system that can analyse the facial expressions of sheep to recognise pain and lead to earlier diagnoses and treatments.

Of course, it is important to note that all of these new AI systems are still in the development phase. Despite their evident degree of success so far, they are not yet at the stage where they can become ubiquitous parts of our everyday medical experience. A great deal of further research and development is needed before they can see widespread use. No one can say for sure how long this may take, but it could be sooner than anyone predicts. Ten years ago would we ever have imagined that medical technology could have progressed as far as it has now? Advancements in the field of AI are gathering pace every day, and there’s no telling what benefits the future will bring.