The future of Artificial Intelligence in healthcare
Since Artificial Intelligence was first introduced into healthcare in 1956, it has come a long way. From MelaFind technology to virtual assistants, there are a variety of AI applications that are well placed to potentially save lives and improve patient care.
A group of scholars from Dartmouth College came up with the term over 60 years ago. It was showcased during the opening remarks at the CDW-G’s AI Showcase at Rutgers University. The group wanted to examine the possibility of using machines to solve problems that humans usually solved with their organic intelligence. Now, AI has moved well beyond those early stages of research and development and it is now used in our daily lives.
The co-presenter and CDW solution architect for AI and deep learning, Jeremy Wise, noted that we are getting to the point where machines are able to tell us things we don’t know. Paul Weber, associate dean for continuing medical education at Rutgers Robert Wood Johnson and New Jersey medical schools, showed during a presentation on the present forthcoming clinical applications of AI that it is certainly true that healthcare is evolving AI wise.
Dr. Weber gave examples of several cases where he has seen AI being used in the healthcare field, including patient communication, diagnosis and treatment options, and the coordination of care.
He suggested that we can train machines to display human-like intelligence and apply that to healthcare. Human intelligence isn’t quite there yet, but it is close.
Clinical applications of AI today and in the future
There are many applications of AI available on the market now, or they are awaiting approval. These applications can potentially save lives and improve patient care.
The applications use pattern recognition, like natural language processing and robotics, which includes translation and speech recognition. Machines will learn the technique which trains the software algorithms to learn from and act upon the data to continually improve. Examples showing the latest tools that use AI to supplement various areas in healthcare and medicine are:
- Virtual assistants: This technology is Driven by AI and can help people with Alzheimer’s in their daily lives. An example is Brian Leblanc, 59, who was diagnosed with early-onset Alzheimer’s disease in 2014. He began using Alexa on his Amazon Echo Dot for reminders to bathe, take medication, and eat. Dr. Weber suggests that this has enabled Mr. Leblanc to have more control over his life.
- MelaFind: This is a technology that uses infrared lights to assess pigmented lesions. Dermatologists can study irregular moles and diagnose severe skin cancers, such as melanoma, by looking at algorithms. While this new technology cannot replace a biopsy, it can help with seeing the early identifiers.
- Robotic-assisted therapy: Robotics and AI are used to assist patients with their recovery from a stroke at the Bionik Laboratories in Toronto and Watertown, Mass. The robotic hand and arm use electronic algorithms to discover the movements that patients cannot carry out during therapy and guides them. The therapy can help patients to perform better movements per hour than they would be able to with a physical therapist alone.
- Caption Guidance: This AI-powered software was just approved by the Food and Drug Administration. The software can help healthcare professionals to take echocardiographic pictures of a patient’s heart that are of satisfactory diagnostic quality, without specialized training. The machine teaches the software to find high-quality 2D ultrasound pictures of the heart. The software can also record videos, which is a huge change in the way heart disease is diagnosed.
What medical professionals should consider before adopting AI
It is important for medical professionals to understand the set regulations for the efficient development and deployment of these technologies. Approval also needs to be sought from the regulatory authorities before the AI applications are released. Nowadays, agencies plan out their regulatory framework and how it can be adjusted to the new technologies. For example, a new framework was introduced by the FDA (Food and Drug Administration) last year that allows it to pre-approve the manufacture of AI-powered software. The framework allows for more testing and faster approval, so the turnaround is faster, just like the smartphone tech industry.
Healthcare professionals need to prioritise patient security and privacy when using AI applications. After all, the practitioner-patient relationship is the most important thing, despite the growing AI presence. Patients still want to speak to their practitioners, and AI shouldn’t replace human interaction; there should still be someone in charge of someone’s care.