The Future of AI in Healthcare
The healthcare industry is set to see some major changes in the coming years. AI systems focused on improving areas such as diagnostics and treatment are being widely developed, many of which are showing encouraging results.
A great example of this emerging technology is a diagnostics system created by Deepmind. It is capable of detecting over 50 forms of eye disease with 94% accuracy by analysing eye scans. This level of accuracy matches the performance of top medical experts in the field, making it a clear demonstration of AI’s potential.
Another approach to using AI for diagnostics comes in the form of chat bots. These intelligent bots use Natural Language Processing to understand the information patients give, which can be via text or voice. The bot can then compare symptoms the patient presents against a database of medical conditions, and offer specific medical advice accordingly. Babylon Health is a prime example of this. Their chat bot is featured on the website’s homepage if you want to try it out for yourself.
As mentioned, diagnostics are not the only area of healthcare AI is helping to improve. Surgery is also seeing an influx of systems and robotics aimed at improving safety and efficiency. In 2017, a study reviewed 379 orthopaedic operations. 250 of these were robot guided, whilst the remaining 79 relied on a surgeon using X-Rays to make accurate incisions. This study found the robot guided surgeries were five times less likely to result in complications when compared to the surgeons operating alone.
A similar piece of surgical technology, named STAR (or Smart Tissue Autonomous Robot), is an AI system designed to make precise incisions during surgery. This system is intended to minimise healthy tissue damage whilst accurately cutting away the problematic area, such as a tumor. It is also capable of stitching tissue together, and actually performed better than a surgeon in a test which involved sewing pig intestines back together.
Whilst AI holds huge potential to advance and improve healthcare, it’s still in its infancy. Many of the systems in development have a long way to go before they will be considered safe for use on humans. There are also ethical factors to consider when using AI for high risk procedures, the key one being who is liable if something goes wrong.
We can be confident that the next decade will bring some significant developments and challenges for AI in healthcare.