Artificial Intelligence is credited with some amazing abilities. University of Adelaide researchers are now hoping to add to these abilities by creating a robot equipped with AI that can actually take a look at your organs and let you know how long you will live. Okay, it’s not quite at that stage right now, but it’s only a matter of time.

Early diagnosis of most serious conditions results in timely medical intervention. This has been known to save lives. Predicting the course of a disease can allow doctors to create tailor made treatment processes for a patient. Dr Luke Oakden-Rayner,  PhD student with the University of Adelaide’s School of Public Health is a radiologist who wishes to make this diagnosis device a reality.

The idea is to combine deep learning with an ability to analyse images received in order to predict the outcome of the disease. The study of 48 patients’ chests was undertaken to to predict which patients would die within five years. When the results were compared to human predictions made by clinical doctors there was an accuracy rate of 68%

This science experiment shows that it is possible to predict just how long a person will love based on the medical data accumulated by the robot using Artificial Intelligence.

Getting robots to play existing pieces of music is as easy as programming them with what to do at what interval of time. However is it possible for a music playing robot to actually come up with a new piece of music all by itself? That’s what researchers at the Georgia Institute of Technology are trying to find out.

Shimon is a marimba playing robot with four arms and eight sticks. As part of a science experiment to see if Shimon could use deep learning and artificial intelligence to write new music, researchers fed it over five thousand songs ranging from Beethoven to Lada Gaga. Ph.D. student Mason Bretan also added close to two million motifs, riffs and licks of music to the robot’s repertoire.

Mason Bretan says that after Shimon is given the first four measures to use as a starting point, no humans are involved in either the composition or the performance of the music. The robot has composed two pieces of music that are thirty seconds long. Both performances can be seen on YouTube. This is the first time that a robot has generated melody and harmonic structure on its own. Is this the future of music?

It has been observed that many binary star systems exist throughout the galaxies. Scientists from the University of California – Berkeley, have known that stars are usually binary in nature. In a recent study they wanted to ascertain if they were born that way or one captured the other.

They paid attention to the Perseus molecular cloud which is the dense core within which young stars are born. As the astronomers studied this nest egg of baby stars they were able to see that most sun like stars were born as wide binary systems. These are low mass stars.

Eventually these wide binary systems will either split up or become tight binary systems. The studies were based on statistical models which analysed Perseus molecular clouds. It was found that the only model which could produce the accurate current data started with all stars as binary systems.

This gives credibility to the fact that when our sun was born 4.5 billion years ago, it probably did have a twin. It was not an identical twin, and eventually their binary system split up. There may have been as yet unrecorded nemesis which caused the split. It would make an interesting science project to find out more about this twin.