AI in healthcare electronics and devices in 2020
Over the past few years, new applications of AI have rapidly spread through the medical field, changing the way scientists perform research, doctors run hospitals and patients receive treatment. Some scientists have already used AI tools to develop new vaccines and diagnose patients. Now, AI is poised to revolutionise another sector of the healthcare industry — medical devices.
By Kayla Matthews
Medical electronics that will become available this year, aimed towards both professionals and consumers, feature some of the first applications of AI in medical devices. These MedTech tools may transform the manufacturing and design of future equipment.
Here is how manufacturers are implementing AI in healthcare.
How AI Is transforming medical electronics
Some MedTech devices that incorporate AI are taking advantage of the sensory processing powers the technology can bring, like machine vision and audio signal processing, to provide extra senses for those with impaired sight or hearing.
For example, Envision is a new app that integrates with smartphones, or smart glasses like Google Glass, to turn these devices into an extra pair of eyes for the sight-impaired. By using machine vision, the app can scan the environment and describe it to the user with synthesised speech. The platform can also scan text on signs or packaging and read it out loud.
Another new device, from Canadian medical startup HearThat, uses AI-powered noise filtering to remove background sounds picked up by hearing aids, providing sound amplification with the least amount of noise possible.
In other cases, AI is improving existing gadgets by providing extra insights. The technology's advanced pattern-finding abilities make this possible. One device recently approved by the FDA has integrated AI into existing computed tomography (CT) machines. In these revamped machines, AI reconstructs images and filters out image noise more accurately, providing radiologists with the best possible pictures to work with.
AI will also become increasingly intertwined with the internet of medical things (or IoMT). In some studies, AI has been successful in helping doctors process and analyse the vast amounts of real-time patient data collected by IoMT devices. Soon, physicians could use AI to make IoMT-powered medical observation more practical.
The challenges faced by AI in medtech
AI-powered medical devices will face roadblocks similar to other electronics, like the long and often intensive FDA regulation process for new gadgets. Revised FDA guidelines, designed to provide a framework for the approval of new uses for AI in healthcare, may soon streamline this process, however.
Doctors and other medical professionals should also be prepared to handle false positives and other errors made by AI analysis. For example, one AI tool incorrectly concluded that patients with pneumonia were less likely to die if they had asthma. Misleading or false conclusions like this can easily lead to poor or harmful medical decisions.
Maintaining data security for these MedTech devices will also be a significant challenge. AI-powered medical wearables collect, store and transfer large amounts of sensitive and identifying medical data, as those devices regularly interact with other consumer devices or send data back to clinicians.
Data safety challenges can grow even more because of low-security or unsecured networks — or by use of 5G, which provides fast connection speeds but suffers from a range of unique security issues.
Medical electronics manufacturers will need to be especially aware of the unprecedented security difficulties associated with AI-powered medical devices. They should take all the necessary steps to ensure they are developing the most secure tools possible.
The future of AI-powered medical devices
AI is transforming healthcare through the creation of new medical devices. These tools primarily take advantage of the data processing and pattern-finding abilities of AI to boost their functioning powers.
For example, developers have created equipment such as AI-powered hearing aids that provide cleaner audio amplification and CT scanners that use AI to deliver more accurate images for radiologists.
These new electronics, however, will face some challenges. Device manufacturers who integrate AI in MedTech will need to carefully monitor error risks and security dangers associated with data collection and transfer.
Kayla Matthews is a technology journalist and writer whose work has also been featured on a number of publications. To read more from her, find Kayla's tech blog here.