Posted by 
Artificial intelligence has a lot of potential to improve many areas on our lives. Algorithms that can learn from vast databases can help improving problem solving at rates we didn’t even consider possible. However, a new artificial intelligence system from the University of Queensland is so versatile it could improve insulin dosing for diabetics and transform the way aeroplane engine wear-and-tear is monitored.
Artificial intelligence system would allow for a much more accurate dosage of medicine, making the treatment more precise and with fewer side effects. Image credit: Biswarup Ganguly via Wikimedia (CC BY 3.0)
Artificial intelligence is typically characterized as an algorithm, which can learn, analyse and execute by itself. It changes as it is learning. Usually artificial intelligence systems are rather specialized – they can complete one or another task, pre-programmed by people. In other words, it is not some human-like application like movies would like you to believe. And the world-first machine-intelligent artificial pancreas developed in Queensland seems to be a similar achievement. It is able to recommend the best insulin dosage to keep each individual patient’s blood glucose levels under control. Because it studies a huge amount of data, this system is extremely precise.
Diabetes treatment has a lot of side effects. This AI system will help minimizing some of them just by allowing for a more accurate treatment. In other words, the algorithm would calculate the exact dosage, making sure that the patient is not taking too much medicine, which could cause adverse side effects. However, the same artificial intelligence system can be used for something else as well. Scientists were invited to apply it to aviation turbine engines and their related systems. In this situation artificial intelligence also analyses a vast amount of data, which helps it predicting aviation engine component degradation.
It is quite interesting that the artificial intelligence system is virtually the same in both applications. However, it is not that different of a job. In both cases the algorithm has to analyse a huge database and use its information to predict movement of the variables. In one area it helps creating a more precise treatment and in another – a better plan of service intervals. Dr Ingo Jahn, leader of the research team, said: “We’re able to use EMI’s breakthrough AI technology to predict aviation engine component degradation and plan services to improve performance. It allows us to evolve computational models of aviation engines as if they were organisms and the AI can explain explicitly what it thinks is happening inside the engine”.
Artificial intelligence is extremely versatile. In the future we will see many more application of it. However, it is important to recognize where and when it is not good to rely on it. It is likely that more research needs to be done to confirm its effectiveness.
Source: University of Queensland