“There is no reason and no way that a human mind can keep up with an artificial intelligence machine by 2035.”
In my book, The Future of Healthcare: Humans and Machines Partnering for Better Outcomes, I spent a lot of time talking about how AI, machine learning and other technologies can revolutionize the field of healthcare. What I didn’t tell you is that they can also revolutionize the field of competitive gaming. Google’s DeepMind has taught itself to play a number of Atari gamesincluding Atari Breakout, and YouTuber SethBling has created a cleverly-named program called MarI/O which uses “neural networks and genetic algorithms [to kick] butt at Super Mario World.”
Another interesting use of artificial intelligence comes to us via Google’s AI team, which has recently discovered “a new method for teaching computers to understand why some images are more aesthetically pleasing than others.” It’s called neural image assessment (NIMA) and it relies on deep learning to train a convolutional neural network (CNN) to predict ratings for different images. And, as TheNextWeb explains, “The NIMA model eschews traditional approaches in favor of a 10-point rating scale. A machine examines both the specific pixels of an image and its overall aesthetic. It then determines how likely any rating is to be chosen by a human.”
It might not seem like this has much to do with healthcare, but it’s important to remember that AI is still a new field and advances in one industry can often be applied to another. In this case, a variant of the NIMA model could be used to determine which patients need the most immediate treatment.
According to a paper from the NIMA project’s researchers, “[Our] approach differs from others in that we predict the distribution of human opinion scores using a convolutional neural network. Our resulting network can be used to not only score images reliably and with high correlation to human perception, but also to assist with adaptation and optimization of photo editing/enhancement algorithms in a photographic pipeline.” In other words, it could help photographers to sort through photos and determine which ones are the best.
Let’s face it — AI is cool. But it’s more than just some flashy new technology. I truly believe that AI will universalize healthcare and remove borders between countries. It’ll usher in a future that we could hardly even have dreamed about back in 1971, when Médecins Sans Frontières was first established. Health data from people in third world countries and developed countries could all be part of one huge database with AI algorithms to personalize care and offer an equal standard of healthcare all over the globe, when it comes to expertise if not when it comes to access to drugs and machinery. The internet has already opened up the borders, and when you open up the data — or any data, for that matter — to unsupervised machine learning, the potential for positive change is greater than ever.
When I talk about AI and machine learning, a lot of people get confused. There still seems to be some uncertainty around the similarities and differences between the two technologies. Fortunately, I recently came across a great article by author Bernard Marr for Forbes in which he explained, “Artificial intelligence is the broader concept of machines being able to carry out tasks in a way that we would consider ‘smart’. Machine learning is a current application of AI based around the idea that we should really just be able to give machines access to data and let them learn for themselves.”
