Protein Sensing Chip
While many of the technologies being demonstrated were all about convenience and making our high tech lives ever easier, there was one innovation that has massive potential to change the lives of the entire human population, particularly for those in the third world.
Intel has developed a chip that can detect specific proteins, giving near instant results. What does this mean for us? Well, viruses, bacteria and other antibodies all have signature proteins by which they can be identified. By detecting these proteins in a sample of blood, say, you can confirm the presence of the associated disease.
This is nothing new, and there are many established methods for doing this. However, they generally rely on sending a sample away to be analysed in a lab, which takes a long time. So, what Intel has done is use this protein matching method and interfaced it with a silicon chip to create a simple portable test that can be used on site.
The exact science behind how this is done takes a fair bit of explaining but it essentially involves detecting the change in resistance that occurs when the target protein binds with a manufactured protein on the chip. The change in resistance can then be used to trigger an alarm or turn on a light, etc.
The chip can only be used once so the plan is to have a disposable part that comes into contact with the blood (or whatever it is you're testing) and a processing part that displays the results. When the manufacturing process is sufficiently refined it is hoped the chip will be cheap enough to be used en masse around the world.
Dynamic Physical Rendering
As well as all the more immediate research that is aiming at production within the next few years, Intel had a number of 'Blue Sky' projects that have indefinite timelines. There was some interesting stuff on show but the one thing that really caught my eye was what Intel has called Dynamic Physical Rendering (DPR).
The goal of this project is to create a "material" that can change shape under software control and in real time. Think of it as rapid prototyping on steroids.
The material would be composed of millions of tiny spherical robot modules covered in electronic actuators (or, as in the prototype, electromagnets). By applying charges to different actuators, different points on the sphere would be repelled or attracted to similar points on other spheres. The coordinated movement of all the spheres would result in the material assuming a shape.
Each robot will effectively have its own CPU that interacts with all the other robots in an 'intelligent' manner, so rather than specify exactly where and how each robot should move, the controller should just be able to specify a shape and the robots do the rest. The challenges of doing this are similar to those of programming and controlling terascale CPUs, thus the close relationship with the other Intel research teams.
Taking both the physical limitations and the programming challenges of this project into account, I don't see this technology even being demonstrable within the next five or probably even ten years. But, it's still very exciting to see what may be possible in the future.
It was a shame we only had a day to look round all the research projects as there were plenty more I would have liked to look at and many more questions I could've asked the researchers. However, one thing is without doubt, there's going to be some interesting things coming out from Intel in the coming years.