Of late, I have been reading a great deal about cognitive computing. Cognitive computing refers to a whole new level of capability, that allows computers to find patterns in huge quantities of data. A key feature of cognitive computing is the ability to learn how to process data, based on previous experience. If this sounds like the way in which humans learn a task, that is the point.
I have previously written about IBM Watson, which is presently the primary example of cognitive computing. No programmer sat down and wrote specific instructions, i.e. software code, that tells IBM Watson how to find a specific pattern in medical journals. The programmers behind IBM Watson wrote a system that can learn. After that, the programmers stepped back and other specialists began to teach IBM Watson based on examples, once again in a very similar way to teaching a child how to, for example, work out a math problem.
If you ask one of the developers of IBM Watson how exactly the system identified a specific pattern, the developer will not be able to tell you. The developer will be able to explain how learning systems attain knowledge. But he or she will not be able to point at a specific line of software, which is responsible for Watson being able to see the connection between two apparently unrelated articles.
When you watch a TV show that has a lead detective character, there is often a moment where this detective sees a connection that no one else has spotted. This TV character might suddenly become excited and ruffle through a whole set of files. At some point , there will be that “Eureka” moment when the case is solved. Watson has such Eureka moments on a regular basis. In fact, Watson was designed specifically to have such moments and to make such discoveries. A human simply cannot match this capability due to the mass of data involved.
You will likely also begin to hear more and more about “deep” learning. Deep learning computer systems are large-scale systems that have similar capabilities to those of Watson. There are not many of them, yet. But there is tremendous research and money going into their development. It should be no surprise to anyone that a company that can sell such a capability to others, has an incredibly powerful product in hand. And, in an open world with the free ability to engage in business, other companies will soon rise with their own solutions, to challenge and compete with, in this case, IBM and its Watson system.
Business competition is a form of natural selection that yields the best solutions for the problems we face on a day-to-day basis. It is unfortunately, a harsh environment to work in. Anyone who has tried to start a business, and compete with existing large companies, knows how ruthless this type of competition can be. Still, if a new company can succeed, especially if they have a new type of solution to existing problems, the potential is tremendous. A company like Skype started very small but ended up causing all of the major telephone companies in the world to change the way they provided service. In a world where people can make endless phone calls anywhere in the world for free or for very little money, the major telephone companies must respond or become obsolete. For the consumer, this is the best possible situation.
What happens if the medical world truly acts in the same competitive fashion? What if one hospital in the world develops a new way of treating cancer, but does not publish the results of their work? This would lead to a situation where people would have to choose between dying of cancer or spending tremendous amounts of money to travel to this one hospital to get this new type of care. It is true that other hospitals would work to develop equivalent new types of care. But measuring the success of one type of treatment versus another is sometimes very difficult. Therefore, a patient with a given cancer would struggle to choose the best place to go to, to receive care.
At the moment, top cancer–specialized hospitals, like Sloan Kettering Memorial (SLK), are using IBM Watson to extract the best customized type of care for each patient with cancer. Based on the patient’s personal profile and the petabytes of data about cancer treatments, Watson does what no human can do; it pinpoints the best specific treatment for that patient. No human can read and internalize millions of medical articles that are in some way related to the patient’s cancer. Unquestionably, SLK is truly now able to provide a unique level of care that simply never previously existed.
The expectation is that SLK will publish their work with an explanation of how they achieved their success. This is effectively expected within the medical community. SLK has the right to patent their work, and then sell the services they created to other medical centers. But SLK is expected in some way or form to share their discoveries with others. If SLK were to say that they were keeping all of their discoveries to themselves, and not helping the world’s medical community to treat patients everywhere, it would be seen as unprofessional and even unethical.
Physicians’ first responsibility is to the patient, whether they work in a socialized system of medicine or a capitalistic one. A physician, of course, has the right to earn a salary and even to make a nice living. Especially in the United States where the cost of a medical education is so high, and the cost of malpractice insurance continues to be high throughout the career of the physician, doctors legitimately expect to make a significant income. But there must be a compromise between the potential business success of an individual or group of physicians, and the care they deliver.
It is by no means a wild prediction to say that cognitive computing will be a standard in hospitals across the world within a decade. Every patient anywhere in the world should be able to benefit from such an astounding computer based healthcare system. Today, where remote consulting is becoming much more widespread, patients around the world should be able to access high quality medical information any time and anywhere. If major medical centers were to hide their discoveries, it would definitely delay the worldwide access to these new technologies. Physicians should make sure that such delays do not happen. There is plenty of opportunity to make a great deal of money, while still sharing major discoveries.
In time, I hope to see such systems become totally open, so that an individual in their own home could ask the same kinds of questions that doctors ask today. An individual should soon be able to give a system, like IBM Watson, access to their personal medical record, and then ask a straightforward human language question such as “what is the best treatment for my specific disease”. With this answer in hand, the individual could then explore options for where to receive this care. Perhaps, the necessary care will be accessible even through small medical centers. Since the diagnosis and treatment have already been laid out, the staff in a small facility with effectively only need to write a prescription or infuse a medication. I believe it goes without saying that this would be nothing less than a revolution in healthcare. And it is likely to be a reality within the next decade, or at most within 20 years.
I wish the business world well. It is critical that businesses succeed so that our economies stay healthy. But on the medical side, I pray that the only real competition will be in regards to who can cure more diseases faster.
I welcome one and all to my new website at http://mtc.expert
Thanks for listening