Do you remember the battle of man vs. machine in the world of chess? The project of creating a computer to beat the greatest human chess player Garry Kasparov? At first, Deep Thought was handily defeated. But then IBM went back to work and developed Deep Blue, which was able to challenge Kasparov through its ability of continuous learning.
By now, the question is not whether artificial intelligence is able to beat a chess player, but how it is going to reform complete industries and scientific areas. IBM Watson, the company’s advanced artificial intelligence program is transforming healthcare into a quantifiable service where every bit of information is available and physicians only have to go through their personalized reports instead of reading through dozens of papers for every patient’s case.
Lately, there is a tremendous amount of information about AI and IBM Watson, and talking about how artificial intelligence will change our lives is important but the best way to show its potentials and pitfalls is by showing it in practice.
This is the reason why I asked medical professionals how it feels like when they arrive to the hospital and Watson is already there sifting through millions of pieces of information.
Dr. Iain Hennessey, Clinical Director of Innovation at Alder Hey Children’s Hospital, who works with Watson on a daily basis, said that he started using IBM Watson as part of a science and technology facilities council project being run by the Hartree Centre. The UK is spending £300 million pounds to develop its capabilities in this area and we are one of the first use cases, said Hennessey.
Martijn G.H. Van Oijen, PhD is an Associate Professor at the Academic Medical Center – University of Amsterdam. As a clinical epidemiologist, he worked with IBM Watson for Oncology in preparation for a research proposal studying the role of Digital Decision Support tools. With several clinical colleagues of the department of Medical Oncology, they studied Watson for Oncology’s approach in approximately 400 surrogate patients with breast, lung or colorectal cancer.
Van Oijen told me that he had several preconceptions about IBM Watson before he started using it. First and foremost, he was concerned about the fact that Watson for Oncology is a cloud-based tool. Thus they have to make sure to protect the privacy of all information transferring between our electronic health records and Watson for Oncology. He also thought that as he works in a tertiary care referral center, he will probably be more involved in training the system. And thirdly, Van Oijen believed integration in electronic health records remains difficult.
What are the tasks in which Watson is an invaluable asset for them?
Hennessey said that although the applications are legion, they have chosen to concentrate on giving patients information and receiving feedback. Alder Hey is famous for its caring approach to healthcare, it is one of their “strongest ups”. Watson can provide a natural language interface for the delivery of general and patient specific information. This has proven very interesting as children seem to be more forthcoming in their interaction with Watson than with a doctor or nurse, explained the clinical director.
Van Oijen believes that Watson for Oncology could result in a reduction of costs and efforts. He thinks that Watson for Oncology in its current form is perfectly designed for use in preparation for and during multidisciplinary tumor boards. According to him, it is the ideal additional discipline, providing up-to-date support during the tumor board meeting, but can also triage (prioritize) on beforehand the order in which patients are discussed during the tumor board meetings. External experts may then only be consulted for the first patients, which would result in a reduction of costs and efforts.
What are the tasks in which Watson cannot replace or even assist clinicians?
Both practitioners thought that AI cannot be a substitute for communication. Right now, the technology is in its infancy, but improving all the time, said Hennessey. He added that it means its main application is in the field of lower level Q&As. Higher level conversation and reassurance is not quite there yet and it cannot help with physical tasks or procedures either.
How do they see the use of cognitive computing in medicine in the coming 5-10 years?
Van Oijen thinks that within 5-10 years, cognitive computing systems in the field of oncology will assist all tumor boards and individual clinicians with interpretation, display in usable dashboards of imaging and ‘omics’ (such as genomics, metabolomics or proteomics) data per patient and treatment recommendations.
According to Hennessey, the most significant change will happen in the field of patient empowerment. He thinks AI will facilitate earlier and more comprehensive harvesting of patient symptoms. Why wait until you have seen the doctor to tell your story? This will free up more time for practitioners to formulate plans and explain rather than basic info gathering, he explained.
The clinical director emphasized that harvesting unstructured data and non-traditional sources of health data (social media, energy use, weather, and traffic) is also an area of particular interest. Another is the reduction of bureaucracy and improving how we interface with electronic patient records, which is currently very difficult due to poor user interfaces.
When we just talk about disruptive innovations in healthcare, we tend to overhype the importance of technologies. When we see them in everyday practice, things become clearer. The fear that technologies such as Watson will replace physicians has no basis.
Its role, just as the role of physicians in healthcare, will evolve and change constantly. We need to be ready for re-inventing ourselves several times in our career as medical professionals. Instead of holding the key to the ivory tower, we need to be guides in the digital jungle. Instead of running after information, we should let the information come to us. Not being able to change must be really scary in a world where the only constant factor is change.