There have been great developments in medical technology in the past few decades. From things such as innovative procedures to creating vaccines to bionic prosthetics – these have not only saved lives but have also improved the quality of life of billions around the world.

The latest medical technology that has been developed involves something that was once only seen in movies – artificial intelligence.

Researchers believe that they can create an artificial intelligence system that can predict when a patient will die based on images on their organs.

At least that is what a new study, published in the Nature journal Scientific Reports, says.

According to this new research, this artificial intelligence system will be able to tell which patients will die almost as accurately as trained specialists.

“Predicting the future of a patient is useful because it may enable doctors to tailor treatments to the individual,” said lead author Dr Luke Oakden-Rayner, a radiologist and PhD student.

What this new study was able to successfully do was show that artificial intelligence can detect things on medical images that some doctors can’t.

“The accurate assessment of biological age and the prediction of a patient’s longevity has so far been limited by doctors’ inability to look inside the body and measure the health of each organ.”

With technology being this precise and analytical, it means that there is a greater chance of early detection and treatment for a large number of serious illnesses and conditions.

At the University of Adelaide, researchers were able to use artificial intelligence to analyse chest scans of 48 patients under 60.

From those scans, the computer based analysis were able to predict which patients would die in the next five years. However, this was done with a 69 per cent accuracy meaning that there is still room for error.

Though the researcher were not entirely sure what the system were specifically detecting in the CT scans, the system was able to analyse patterns and anomalies from a large amounts of data.

“Our research has investigated the use of “deep learning”, a technique where computer systems can learn how to understand and analyse images,” says Dr Oakden-Rayner.

In a comparison between human and machine predictions, a total of 15,957 individual bio-marker features, from seven different tissue segments, were identified by an expert radiologist.

Approximately 2.6 per cent of these images, around 417 of them, showed signs of the patients having less than five years survival.

The “deep learning” model from the artificial intelligence system was shown to closely mirror the predictions made by the radiologist.  

The most sure-fire results were seen in patients with severe chronic diseases – conditions such as congestive heart failure and emphysema.

“Although for this study only a small sample of patients was used, our research suggests that the computer has learnt to recognise the complex imaging appearances of diseases, something that requires extensive training for human experts.”

“Our research opens new avenues for the application of artificial intelligence technology in medical image analysis, and could offer new hope for the early detection of serious illness, requiring specific medical interventions.”

The next part of the research is looking to repeat the previously done research on a larger scale – tens of thousands of medical scans.

The researchers are hoping that this new technology can be used to predict medical conditions, such as the onset of heart attacks.

Though this technology could be very useful, by no means is it a way to replace doctors. If anything, it can be used to assist medical staff with detecting problems that they may have missed and doing further investigations.   

(Visited 162 times, 1 visits today)