Much of the press on the future of healthcare states that robots and computers will take over the practice of medicine. Well, no evolution is without challenges – as seen in a recent Wall Street Journal article titled, “Hospital Falters in Bid to Use AI Tech” about the MD Anderson Cancer Center’s experience with IBM Watson’s recent pilot for the diagnosis and treatment for lung cancer.
But this isn’t to say the medical technology revolution currently underway is poised to reverse course. The article does state that the Watson AI system achieved a 90% accurate rate in its predictions (suggesting the same treatment plan for patients as human physicians at MD Anderson).
Artificial intelligence (or AI) and machine learning are coming to healthcare, and will be all around us soon.
What are Machine Learning and Artificial Intelligence or AI?
In computer science, the field of AI research defines itself as the study of intelligent agents – any device that perceives its environment and takes actions that maximize its chance of success at some goal (definition curtesy of Wikipedia). In other words, an AI can acquire data itself, understand, and process that data to take certain actions.
Machine learning, coined by Arthur Samuel in 1959 as the “field of study that gives computers the ability to learn without being explicitly programmed”, evolved from AI and involves mathematical algorithms that can independently learn from and make predictions on data (again, Wikipedia).
Basically, if we can teach a machine or a computer to learn on its own, we may not have to program it to perform complex surgical procedures or recognize tumors.
How does this relate to Healthcare?
If we can teach a machine to recognize tumors – then it may be able to identify tumors so small they are invisible to the naked eye. And detection that early would only increase treatment options and potential outcomes.
This is precisely what is happening. Only recently, news broke that researchers at Stanford had used a deep convolutional neural net algorithm, which started off as Google Brain, to detect skin cancers as well as human dermatologists could.
Likely because of our belief in the benefits machine learning and AI will soon bring to the practice of medicine, my colleague, Ajay K. Gupta, and I have been asked to chair a conference on this topic. The Machine Learning and AI in Healthcare conference will be a 2-day conference taking place in Boston, Mass on May 3rd and 4th. I hope all can join. The tag line for the conference is:
Connecting Technology to Healthcare to Improve Patient Care, Treatment Efficiency, Productivity, and Cost Reduction.
That is essentially the mission of Health Solutions Research, and this synergy is a major reason we have decided to Chair the event. A partial list of the topics we’ll discuss at the event include:
- Addressing HIPAA and Patient Privacy in a Machine Learning environment
- Effective data management for Machine Learning and AI applications
- Natural Language Processing
- The role of Cybersecurity
- How Machine Learning can advance Prevention Techniques
- Chronic Disease Management through Machine Learning
- The value proposition for human-machine interaction
We will also discuss where our GeoHealth Dashboard, which Ajay spoke of in our last blog, fits in this environment. We believe the Dashboard represents an expansion of the overall scope where AI solutions can benefit medicine. Not only can we take patient and health data into consideration, but can include non-health data that is correlated to health outcomes and then perform spatial and temporal analysis on that data.
Learn more about the conference
The bounds of what machine learning and artificial intelligence can bring to healthcare is really endless. Adding GIS and its spatial and temporal analysis capabilities can mean that we are limited only by creativity. All of these topics will be thoroughly discussed in Boston.
This will be an insightful conference and 2 days of conversation that brings together the developers of machine learning and AI technology with the medical providers who will use that tech in patient care.
Please see the conference site for more information, and feel free to contact myself or Ajay (240-731-0756, firstname.lastname@example.org) if you have any questions or would like to attend and possibly present at the conference.