Not all emergencies should go to the local authorities nor should they all be voice sessions
Not all emergencies should go to the local authorities nor should they all be voice sessions
Consider a user with a portable monitoring device for a chronic health problem, this monitor might detect that the user’s blood chemistry is going out of the expected range and automatically set up a session between the users monitor and the user’s physician or other health care professions - or perhaps even an on-line expert system.
Not all emergencies should go to the local authorities nor should they all be voice sessions
Transcript
[slide515] And we talked earlier about the fact that not all emergencies actually need to be voice sessions. Today, this is becoming particularly popular because we're starting to have monitoring devices that measure your blood sugar levels, if you're a diabetic, and will automatically tell you, this is the amount of insulin. Or if you're going into diabetic shock, to automatically alert the emergency services. But what do we need about reliability? Do these need to be highly reliable services? If so, how do we build them on unreliable infrastructures? And there are lots of questions about, where do we put the processing and intelligence? So I have a colleague who has done a system that monitors, is an expert in cardiac signal processing, to be able to predict when it is that you're going to be having heart problems. By looking at how your heart is behaving electronically right now, or electrically right now, and predicting the future, are you going to have an episode in a short period of time? If so, then it can alert the medical authorities, or perhaps even tell you, so you change your behavior and hopefully avoid the bad episode. A big trick nowadays is doing lots of that processing locally, because I can't rely on the communications. So just recently, there's an article by a former student, a Master's student of mine who works for Microsoft in Italy, all about AI in Internet of Things. How can you do the signal processing and do the reasoning locally in the device to be able to do these things? But it's key, they're not going to be voice sessions. It's going to be data being exchanged.