If you recall the Star Trek television show, when someone would become ill or was injured, “Bones” (the ship’s doctor) would wave a diagnostic scanner over their bodies and immediately receive an accurate diagnosis. Then, Bones would often cure his patient with a single shot.
At the time the show was made, such diagnosis and treatment might have been called futuristic. Today, the future is starting to arrive. Artificial intelligence (AI) is changing how we diagnose and treat patients, and we’re making incredible advances in this area quickly. AI’s ability to hold and process huge amounts of information is especially useful today, helping doctors evaluate images.
AI as a Game-Changer
AI is truly a game-changer in terms of medical intervention. For example, AI can input numerous images of a cell mutation present in early-stage liver cancer and then identify these types of cells from a new image (such as an MRI scan) better and faster than even the most experienced doctor or technician.
While AI is currently used as a complementary or even a lead tool, it is not yet being used alone. Invariably, this situation will change, and AI will likely replace radiologists at some point in the future.
Currently, we’re using “narrow AI,” focusing the use on a single task or goal. In the future, we’ll transition to “general AI” where it’s used much more broadly as a diagnostic and decision-making tool for a range of conditions and purposes.
Look at the benefits of AI another way. Imagine a highly experienced doctor, one who has been in practice for 40 years and has treated many patients successfully and learned enough to qualify as an expert. Now imagine being able to draw on the knowledge and skills of 5000 other doctors with the same amount of experience and expertise. And finally, imagine an AI tool that can store all this knowledge in a database, process and evaluate it in terms of a given disease or patient condition and extract the right information for treatment in seconds.
This isn’t a pipedream; it’s the future of AI, and it’s a future that’s approaching rapidly.
The World of Wearables
The tech isn’t there yet, but we’re getting a glimpse of what that tech will be through “wearables.” Currently, the wearables aren’t particularly precise, and the acronym, GIGO (garbage in, garbage out), applies. We haven’t found ways to input all the necessary data/variables that we need to give us a highly useful output.
The problems are many, including such simple issues as a wearable not being worn sufficiently tightly to provide accurate tracking. Another problem is that one size doesn’t fit all. People vary considerably in the data they produce depending on their weight, the way their bodies work, and other factors.
Nonetheless, the wearables market is exploding, and that’s yielding at least some healthspan benefits. Even if the measurables aren’t always accurate, they raise users’ consciousness about important health factors, like heart rate, sleep quality and quantity, blood sugar (continuous blood sugar monitoring), and so on. Ideally, this increased consciousness will result in better diets, more effective and efficient exercise, and a stronger commitment to getting a good night’s sleep. In terms of valuable health data, some wearables can collect fairly precise measurements of temperature, respiration, blood oxygen saturation, heart rate, blood pressure, and electricity-measured ECG readings (sinus rhythm and atrial fibrillation). Paying attention to this data can alert patients and doctors to problems as well as motivate better health-related behaviors.
Invariably, AI and related technologies will advance far beyond the current state of wearables.
The Best Right Now
Let’s take a look at the digital devices that seem to be having the most positive effects on healthspan at the moment.
The Oura ring tracks data conveniently, especially as it relates to EEG, the gold standard for sleep evaluation. It’s become smaller over time—it’s currently the size of a wedding band—and provides additional information about pulse rate, temperature, calories burned, and so on.
Meanwhile, the Apple Watch, which works well with other Apple products, tracks activity with an accelerometer (which helps determine the intensity of your workout) and measures heart rate. This is along the same lines as the Fitbit and Garmin. These devices do a good job of monitoring vital signs continuously during workouts, though for more serious athletes, the chest straps provided by Polar H10 and Wahoo are better.
The BioStamp from MC10 will soon be on the market and is for people who want to dig deep into their health. This device has a sensor that conforms and adheres to at least 25 body locations and offers metrics to evaluate sleep, posture, activity, and vital signs. The measurements are of medical quality and can be used for clinical evaluation. Currently, MC10 has limited distribution for clinical trial use, but given its uniqueness and marketability, it should have a wider distribution in the near future.
The Future is Bright
Of course, this is just the start of the AI’s revolutionary impact on medicine. Right now, the technology exists to increase the speed, accuracy, and efficiency of diagnosis. The real key, though, will be AI’s ability to process data and draw correlations that can be tested. This is already happening with the development of drugs.
We can evaluate a particular virus or bacteria for multiple characteristics that can lead to a possible solution—a new drug or an existing one with the potential actions to combat the virus or bacteria. Without AI, the process of looking at and evaluating the multitude of pathogens and the various drugs and structures to make potential drugs can take decades to identify. With AI, it can be done much more quickly (some evaluations can take only minutes!).
In the current pandemic, scientists have used AI to examine the potential for genetic variants that slowed or accelerated the course of COVID-19. Labs tested these models in vitro to validate AI predictions as to the drugs that would be helpful in slowing virus replication and the mechanism of action.
More significantly, AI has helped generate at least eight different types of COVID-19 vaccines. Without machine-learning systems and computational analyses provided with lightning speed by AI, we would not have produced the vaccine candidates so quickly – especially the non-conventional experimental candidates developed by AI. Without AI, it probably would have taken many months, if not years, to develop them.
Don’t get me wrong; AI is not the complete solution to this and other healthcare challenges. AI isn’t a substitute for the necessary lab and clinical studies; it can’t shorten the time needed to test vaccines on animals and humans. But AI does accelerate theoretical aspects of development and does so at times with blinding speed.
AI can be a game-changer, applicable to preventative, anti-aging, and traditional medicine. It will help us connect the dots between theory, observable and quantifiable data points, especially ones in the area of biological aging and associated biomarkers.