How Artificial Intelligence Could Help Humans Live Longer

MIT is one of the world’s premier research universities, responsible for all manner of innovative and exciting developments since their founding in 1861. Now MIT researchers have developed a machine-learning algorithm that registers 3D images (like brain scans) 1000-plus times more quickly than traditional methods.

Traditionally, 3D images are created via a technique called medical image registration. This process overlays two images, like MRIs, to compare and contrast anatomical information in great detail – an especially useful tool for doctors to gauge progress with a patient or treatment.

MRI (magnetic resonance imaging) scans consist of hundreds of 2D images, all stacked on top of each other to form 3D images. These images, called “volumes”, contain millions of pixels, or “voxels”. Aligning voxels from multiple volumes is a complex process, made more so by variables like spatial orientations and machine types. Adrian Dalca, a postdoc at Massachusetts General Hospital and CSAIL and co-author of the paper, describes it as “wiggling” the images until the images fit each other. “Mathematically, this optimization procedure takes a long time,” said Dalca – potentially hundreds of hours, if analyzing scans from large populations of data.

This delay is because the algorithms involved never learn from the information they analyze, instead of dismissing all data regarding voxel location after each pair of images. The new algorithm, VoxelMorph, corrects this flaw, registering information from thousands of pairs of images – “Information you should be able to carry over,” explained Guha Balakrishnan, an MIT grad student, and paper co-author – to learn how to align images and estimate optimal alignment parameters. Once the algorithm “learns”, it maps all pixels from one image to another at once, vastly reducing registration times to a couple of minutes via a standard computer.

VoxelMorph uses a common machine-learning approach called a CNN or convolutional neural network. The CNN network is augmented by a spatial transformer, which captures similarities in voxels between MRI scans. It learns from groups of voxels, which it then uses to develop optimized parameters that can be used on any scan pair. All information is gathered in the training phase, with future registrations executed using a single, easily-computed function evaluation.

Another benefit to VoxelMorph is the data is “unsupervised” – it does not require additional information beyond image data to make an accurate reading. Each registration is “smooth”, or without any image distortion or holes, and can be calculated within roughly two minutes via a traditional CPU, or under a second with a graphics processing unit.

Enhanced speed opens a variety of potential application – scanning other parts of the body, for example, or using image registration in close to real-time. The result is a better experience for patients and a powerful tool in doctors’ pockets, all thanks to machine learning.
 

If You’re Wondering When A.I. Will Start Making Market Predictions…
 

Guess what – it already is. Hedge funds and large institutional investors have been using Artificial Intelligence to analyze large data sets for investment opportunities, and they have also unleashed A.I. on charts to discover patterns and trends. Not only can the A.I. scan thousands of individual securities and cryptocurrencies for patterns and trends, and it generate trade ideas based on what it finds. Hedge funds have had a leg-up on the retail investor for some time now.

Not anymore. Tickeron has launched a new investment platform, and it is designed to give retail investors access to sophisticated AI for a multitude of functions:

• Finding stock patterns in the market
• Finding trends in the stock market
• Testing portfolios to see if they are well-diversified
• Back-testing statistics to see how different stock patterns generated trading results
• Making Predictions for price movements in the future, with “A.I. Rank” and level of confidence in the trade.

And much more. No longer is AI just confined to the biggest hedge funds in the world. It can now be accessed by everyday investors. Start a free trial today on tickeron.com.