In the past, disease was a crapshoot. Few of us knew with a high degree of certainty that a specific serious medical condition was in our future, much less when the disease would strike. This will soon change. Thanks to the artificial intelligence and machine learning revolution, most of us will become “previvors”—that is, we will know which of the 10,000 known human diseases are in our future long before we develop symptoms. Over the next decade those predictions will become both more accurate and more foresighted—increasing the time we have to effect change and seek professional help.
The impact of previvors on the medical field can already be seen in the communities of those with the BRCA1 and BRCA2 mutations and among healthy individuals carrying the HIV virus. Both of these groups coalesced into large consumer-activist organizations advocating for novel treatments and compelling regulatory agencies to speed adoption of promising drugs and interventions. Unfortunately, medicine’s ability to forecast diseases often outpaces breakthroughs for effective interventions and cures. Facilitated by social media, connected groups of previvors will band together to share peer-to-peer information—some valuable, some junk science.
As future research illuminates clear genetic connections between certain genes and illnesses, an increasing number of people will become previvors. For the first time, the FDA recently approved direct-to-consumer genetic test kits that will reveal risk markers for different diseases. As such consumer test kits become popular and increasingly accurate, the medical world should expect a flood of new patients worried about diseases they don’t yet have.
Testing for specific disease-related genes is only the beginning.
Machine learning and artificial intelligence are already making remarkable new advances in disease forecasting. Much of the data needed for advanced AI disease forecasting already exists. Researchers are harvesting petabytes of patient records and medical images. Adding to that mountain of data are the thousands of clinical trials and the tens of thousands of human genomes that have been partially or fully sequenced.
Researchers at Mount Sinai Hospital in New York recently used machine learning on a collection of 700,000 patient records and found they could outperform traditional ways of forecasting disease for a wide variety of conditions, including diabetes, schizophrenia and various cancers.
Currently half of men and two-thirds of women who die suddenly of coronary heart disease have no previous warning signs.
Using 200,000 patient records and new machine learning tools, researchers at Sutter Health in California were able to predict heart failure 9 months earlier than doctors using traditional methods.
University of California medical centers are currently using AI to harvest insights from over 13 million patient records.
CONDITION |
CURRENT DIAGNOSIS
|
KEY PREVIVORS
|
POSSIBLE PREVIVORS
|
Breast cancer |
BRCA1 and BRCA2 mutation. Breast removal. | AI algorithms will allow women to know with more certainty when or if breast removal is necessary. | Gene therapy using CRISPR will remove health threats encoded in BRCA genes and keep them from being passed to future generations. |
Parkinson’s disease |
Disease diagnosed at onset of symptoms. Family history increases likelihood. | Our interactions with touch screens will pick up early signs of the condition. | Early deep-brain stimulation— either through wearable or implantable devices—will be employed at earliest signs of the condition. |
Alzheimer’s disease |
No specific test exists to diagnose Alzheimer’s. Family history increases likelihood. Low efficacy drug treatments. | AI algorithms analyzing polygenic risks and brain imaging will predict the disease in early adulthood. | Optogenetic stimulation of interneurons through implantable devices may decrease amyloid-beta production before symptoms appear. |
Celiac disease |
A blood test and intestinal biopsy can provide a diagnosis. Dietary changes. | Sensors in toothbrushes and toilets will monitor and predict all gut-related conditions. | A sensor and drug delivery device placed into the digestive tract meters out pneumococcal vaccine for ongoing treatment. |
Type 2 diabetes |
Prediabetes blood testing can give type 2 previvors a decade to make behavior changes. | Lifestyle, diet and blood monitoring through sensors and the internet of things will provide new data for doctors and patients. | Personalized diet designed for individual genome. Constant blood and metabolism monitoring. |
Cardiovascular disease |
Diagnosis is attained through blood tests, X-rays and electrocardiograms—often after a dramatic heart event. | Implanted vascular flow bots and heart rhythm monitors warn of coming danger. | Patients check into hospitals before life-threatening cardiovascular events. |
Lots of diseases are preventable, but they happen so slowly that people get worse without realizing it. If we can use deep learning as a powerful tool to give patients a wake-up call, we’d be able to prevent diseases when there’s still time.
—Professor Narges Razavian
New York University Langone School of Medicine
Data: Increasing Variety and Complexity
THE SOURCES OF DATA PUT TO USE BY AI WILL BOTH WIDEN AND DEEPEN OVER THE NEXT DECADE
The amount of data created by the human body is potentially limitless, and the collection of that data has already moved outside hospitals and doctors’ offices.
Fitbits and watches that collect health information on sleep, activity and heart rate are just the beginning. Sophisticated mobile cardiac monitoring devices and implantable blood chemistry sensors will produce always-on patient data. Such devices will first be deployed on patients with serious illnesses, but eventually they will become so small, noninvasive and inexpensive that they will be used by everyone. Information will be instantly sent through our mobile devices to the cloud to be monitored on health dashboards at lifestyle-disease companies that send alerts to your doctor when diseases are forecast in your future.
As real-time data collection becomes widespread, new groups of previvors will emerge. Eventually, AI will be able to combine lifestyle and environmental data, genetic propensities and biometric tracking to forecast most major diseases sometimes years in advance.
These predictive diagnostics, real-time data and genomic analysis will combine to form a new data standard:
the Universal Health Record.
BY THE TIME PATIENTS BEGIN SHOWING UP AT DOCTORS’ OFFICES IN DRIVERLESS CARS (ANOTHER AI ADVANCE), MANY OF THEM WILL BE EXPECTING TREATMENTS FOR DISEASES THEY DON’T YET HAVE
This will likely lower healthcare costs, as fewer patients will present in acute condition, but will require a rethinking of treatment offerings. The promise of AI disease forecasting—particularly when it comes to diseases affected by lifestyle—is that the longer the patients have to change their behavior, the more chance they have of avoiding or deferring the illness.
As effective preventive interventions come online, more and more people will want to know their previvor status. Early detection will become an obsession. To avoid costly acute illnesses, wearing sensors and having blood labs done monthly won’t just be covered by your insurance provider—it’ll be mandatory.
Soon sensors—some just outside the body, some inside—won’t just monitor the body, they’ll actually intervene. Many people will be wearing some equivalent of an automatic insulin pump for their previvor condition. Already, over 100,000 Parkinson’s patients have pulse generators implanted in their chests that are wired up to their brains to control tremors. In this way, medicine will be mimicking the body’s own homeostatic systems, ever shortening the loop between
BLACK BOX DIAGNOSIS
The long-term outlook for AI and previvorship is a little spooky. As AI machines mine more data, patients will learn that they are headed for certain diseases without any clear understanding of how the AI algorithm made that determination. This opacity is what technologists refer to as the “dark secret” of AI: the way it looks for patterns in data becomes so complicated that not even the computer scientists who design the algorithms can reverse engineer the patterns it ultimately recognizes.
With life-and-death stakes, accuracy will outweigh our desire for human comprehension. A computer may one day tell us to expect a heart attack within the next year but not be able to tell us exactly how or why that prediction was made. “We can build these models,” one researcher said, “but we don’t know how they work.”
Whether it’s an investment decision, a medical decision or maybe a military decision, you don’t want to just rely on a ‘black box’ method. It is a problem that is already relevant, and it’s going to be much more relevant in the future.
—Tommi Jaakkola
Professor at MIT specializing in machine learning