By Udita Shukla
The 2015 sci-fi movie, Tomorrowland, depicted a disillusioned scientist, Frank Walker (played by George Clooney) as the father of a computer algorithm that could predict the age of the world and subsequently the apocalypse. We might not have reached such a milestone but our current scientific community seems to have taken a gigantic leap towards the realisation of something quite similar.
From a bolt out of the blue, a group of international collaborators at the University of Adelaide, Australia, have put forth to the world a remarkable and incredible discovery: predicting the number of years left in a person’s life!
Science behind the magic
The feat has been rendered possible by feeding a computer algorithm with chest x-rays of forty-eight patients. A previously conducted experiment predicted which patients would die within the coming five years, with an astonishing sixty-nine percent accuracy. The technique of medical image analysis is not unheard of in the healthcare industry, but what betrays the incredulity associated with the marvel is that the computers are so smart that researchers admit they do not know exactly what the machines were looking at to make the predictions, noting they are processing massive amounts of data to detect subtle changes.
Although the sample of subjects was limited in scope, the researchers involved have conjectured that computer systems have acquired the ability to identify the probability of occurrence of diseases. This is a milestone in both medical and artificial intelligence where the ‘thinking robots’ can generate medical outcomes, something which is still elusive to their human counterparts.
The findings, now published in the journal Scientific Reports, say that the innovation has implications for the early diagnosis of grave illnesses and subsequent medical intervention. In the words of Dr Luke Oakden-Rayner, lead author, radiologist and a doctoral student at the University of Adelaide’s School of Public Health, “Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual”.
What lies in store for us
The upcoming step in this research is to leverage the results by working on tens of thousands of patients with a varied spectrum of clinical conditions, and come up with concrete self-learning “doctors in a box”; artificial robots which will revolutionise the realm of Medicine and significantly add to the intricacies of Machine Learning.
Up until now, determining a person’s biological age and lifespan has been hindered by the inability to peek inside the human body and assess the health of each organ; the aforementioned research addresses just that and much more: it brings ‘deep learning’ closer to comprehending images with an accuracy experienced ever before!.
The present age has already had its rendezvous with robotic, artificial limbs, virtual assistants like Siri and Cortana, and suchlike; who knows what else will we witness in this recklessly-changing twenty-first century?
Featured image source: Shutterstock
Stay updated with all the insights.
Navigate news, 1 email day.
Subscribe to Qrius