If artificial intelligence (AI) has a big say in what we should watch next on YouTube or any other multimedia streaming platform, then perhaps it can also recommend what we should eat or drink next—or more likely not eat or drink next—to stay in the best of health.
Already the applications of AI are going beyond tips for maintaining a healthy lifestyle, and AI-powered software is becoming an integral part of some medical diagnosis procedures.
Thanks to deep learning algorithms and neural networks, AI solutions have become very good at pattern recognition, which after all, lies at the core of what a human doctor does for figuring out the root cause of a patient’s ailment.
Doctors, in essence, examine all the symptoms a patient is exhibiting—often with the help of medical imaging, bloodwork, and pathological tests—and then compare the systems with the telltale signs of likely diseases and conditions that they have learned about as medical students or cases that they have come across in the medical literature.
That is why doctors with more hands-on experience are generally more adept at diagnosis: they have “seen it all.”
However, a global system of interconnected AI doctors, powered with Big Data, can know about a million more cases than any individual physician could.
Once the data recorded from a patient is fed to an AI diagnosis system, it can go through many possible scenarios in a split second, ruling out all the unlikely ones in light of previously reported cases across the world to finally establish a pattern in the data.
AI’s abilities in pattern recognition with Big Data also makes it a good tool for the interpretation of medical images such as X-rays, CT scans, and MRIs. In the past, doctors had to compare medical images with examples recorded in the literature or medical reference books to see what is at fault.
Medical solutions powered with AI and deep learning algorithms can compare each image with thousands if not millions of reported cases.
In 2017, Pranav Rajpurkar and colleagues developed an algorithm for the analysis of chest X-rays at Stanford University that, using a dataset of over 100,000 X-rays, could detect pneumonia with higher rates of success than practicing radiologists.
This is not to say that human doctors are dispensable, as they have an important quality that AI—at least for the time being—lacks: intuition. However, human doctors are prone to human fallibility such as exhaustion, lack of concentration, and professional burnout.
It is not uncommon for doctors in the developing world to work back-to-back shifts, sometimes performing dozens if not hundreds of visits and ward rounds, all in the course of a single work day.
In contrast, according to a 2011 report published by Medscape, US physicians on average spend 13-16 minutes with each patient, a figure which is much shorter in busy hospitals in the developing world. Under such conditions, the use of AI solutions could be a big help to overworked human doctors and possibly save lives by facilitating early diagnosis.
Vital signs monitoring is one possible use of AI systems. Yes, we know that the monitoring of vital signs such as blood pressure, body temperature, or heart rate by electronic means is old news; NHS England’s Technology Enabled Care Services (TECS) is an on-going program on that front, and there have been similar initiatives in Japan for the tele-monitoring of vital signs in high-risk patients on remote islands.
However, we should note that an AI doctor will be more than a wearable gadget to monitor one’s vital signs and raise the alarm if something goes wrong. A global AI health system will watch you—and millions of others like you—every single second of every single day, figuring out every possible correlation and link between all variables which are at play.
Is a fondness for sugar in our tender years really a predictor of alcoholism in adulthood? Are people with a higher heart rate more likely to succumb to the sudden cardiac arrest syndrome? Is the shape of our brainwaves in REM sleep related to our psychological health? We do not know the answer to these questions yet, but, in a few years’ time, deep learning AI solutions will know for a fact.
And, they will learn it on their own by observing millions and millions of cases without any direct instructional input from us.
All health problems, especially the more serious ones such as cancer and heart conditions, develop following a very clear pattern with early warning signs, which alas go unnoticed because no one is currently fully aware of these signs, nor is anyone there to detect such subtle warning signals when they occur.
But, the application of artificial intelligence in medicine may be about to change all that.