Health Data Meets Artificial Intelligence And Machine Learning
Data holds the key to future medical revolution. The more data we collect, the more we learn from it, and the closer we come to discovering revolutionary new treatments and cures.
The challenge has been speed. How quickly can we collect health data from various procedures and observations? How quickly can we analyze that data to extract new insights? How quickly can we learn as more new data reveals new understanding? How long will it take before all this is turned into meaningful medications and procedures that save lives?
And the answer has always been “not soon enough.” Until now.
Artificial Intelligence, Machine Learning And Cognitive Computing Join The Medical Team
Lots of data starts out offering incomplete value. As we collect more data, more answers are revealed. The problem is that this process of correlating and analyzing can take years before doctors and researchers can generate useful solutions from all the gathering data.
Machines learn faster.
At their most elemental level, all data sets are empirical in nature. They are what they are and are not open to interpretation. As such, they remain usefully objective and only rise in value as they are added to other forms of empirical data. It’s important to maintain the distinction between this unchanging data and the subjective nature of human opinions, interpretations and observations.
Effective health care requires the development of understanding based on two very different types of data. Structured data is found in very neat tables of columns and rows. Each column is another field of data, part of a more complete record. Each row in the table represents another record. This structure makes it easy for digital devices to ingest the data and process it.
What has been trickier has been all the unstructured data we encounter. CAT scans, MRIs and other types of diagnostic imaging are unstructured. There are no records containing precise fields. There are images that must be interpreted. The same is true of written information such as physician notes, diagnostic evaluations, medical research articles and much more.
While it’s easy to imagine computers managing structured records and fields of data, most people couldn’t imagine a computer “reading” an MRI, an article or written notes. Even if they could scan and record the image and written content, they certainly couldn’t develop an understanding of what they were reading or evaluate the images. Right?
Three major advances in technology have contributed to remarkable new capabilities for digital devices to collect, evaluate, understand and apply understanding to recommendations that help health care professionals make better decisions and deliver superior care.
Today, computers can read images and written information, evaluate what they’ve read and seen, draw conclusions, and make recommendations. Decision support at its finest.
No matter how much evaluation they or their human operators perform, and no matter how subjective their interpretations may be, the actual base data always remains objective. It can always be added to and re-evaluated. In other words, data sets can continue to learn and grow.
Examples Of AI In Action
Computer algorithms are quickly making their way into health care from startups and major cloud players. Pager, a startup founded by one of the tech gurus behind Uber, uses AI to enable its online dispatch system to more rapidly match a physician and a patient for a personalized house call.
Google, which has recently made a massive commitment to health care through its DeepMind Health initiative, is also using AI algorithms to enable researchers to better anonymize patient data and find nuggets of valuable information to combat disease.
And according to Mercom Capital (via Healthcare IT News), health care technology companies received over $8 billion in funding last year. Funding specifically related to AI surpassed $400 million. With these figures in mind, look for an explosion of use cases in the health care space.
Surfacing Real Value
As we add more data and our systems collect and capture more data, they continue to learn. As they learn they grow in value. New data may often surface new insights that create significant value in the improvement that can be made in treatment. Smarter learning systems create more value over time.
Most important to remember is how much faster AI, ML and cognitive technologies can accomplish their mission than humans. They process new data into information at rates much faster than humans. Time to value is slashed, and new discoveries emerge at record rates.
Don’t Misinterpret Regulation
The common misperception is that the Health Information Portability and Accountability Act (HIPAA) represents very restrictive regulations that inhibit the sharing of information with the goal of protecting the private health information (PHI) of all patients. Nothing could be further from the truth.
The two most highly significant words in the name of the legislation are “portability” and “accountability.” When you think about it, you quickly realize that patients are best served when their PHI can be shared readily and securely, where it’s needed, when it’s needed. Intelligent systems now make it possible to properly share specific types of information with first responders who can put it to use. An ambulance crew gets relevant PHI. The fire department gets structural information about the site. And police receive any pertinent information regarding potential perpetrators. Most important, each group only gets what it needs.
Bottom line: intelligent systems applied intelligently learn, grow and enable far superior performance far faster than people could accomplish alone. All they need is data.