IoT needs Machine Learning. Our Health depends on it.

Machine learning used to sound futuristic, but now it’s here as an accepted subfield of computer science and artificial intelligence. It involves the construction of systems that can analyze and learn from the data collected from the expanse of devices and sensors connected to the Internet of Things. Most excitingly, it exists now around us, albeit at a simplistic level.  Consider Pandora, which recently announced it will be acquired by Sirius XM. It studies your preferences and predicts from the data which songs you might like.  Amazon is of course the pioneer in this area, always suggesting books and other items based on your buying choices.  In other words, machines are looking at data and finding patterns and similarities that can be used to predict your preferences and behaviors.

Where else can this go? The better question is, “Where can’t it go?”

For instance, health wearables can already track and sense our various bodily functions and movements. In the near future,  biochips or sensors will follow our heart rate, our temperature, and our blood sugar throughout the day.  Machines will be able to read this data, find patterns, and establish normal ranges for each of us. And it will all be done with real time data – before your heart attack occurs. A human at the other end of the data stream would never be able to analyze the data from hundreds of patients, and assess particular abnormalities for each one.  Machines will have no such limitations.

Those machines aggregating this big data stream could establish similarities, correlations and abnormalities and quickly pinpoint health data outside our norms, flag the information and alert our doctors. That’s why telehealth is such a fascinating and transformative piece of the machine learning story.

It’s time to let machines keep us healthy.