Artificial intelligence (AI) and computer algorithms are being used to predict lifespan and detect malignant tumours
Researchers at the University of Adelaide have recently used artificial intelligence to predict five-year mortality in older individuals (age >60 years), using chest CT imaging.
They used the same type of AI behind the concept of driverless cars – called “deep learning” technology – to predict models for survival and mortality, and compared these with visual scans.
The results, published in Scientific Reports this month, reveal the patients who were predicted to survive longer than five years in the AI models appeared visually healthier than those predicted to die within five years.
“We have presented a novel application of medical image analysis as a proof of concept and to motivate the use of routinely collected, high-resolution radiologic images as sources of high-quality data for precision medicine,” said the authors.
“The overall goal of precision medicine is to inform useful predictive models of health and disease.
“We have shown the first proof of concept experiments for a system that is capable of predicting five-year mortality in older (age >60 years) individuals who have undergone chest CT imaging.”
Lead author and radiologist Luke Oakden-Rayner, a PhD student at the University of Adelaide’s School of Public Health, told InDaily that the technology was able to analyse images in a way that doctors cannot.
“Instead of focusing on diagnosing diseases, the automated systems can predict medical outcomes in a way that doctors are not trained to do, by incorporating large volumes of data and detecting subtle patterns,” said Mr Oakden-Rayner.
“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.”
Meanwhile CSIRO researchers have developed a new algorithm to generate an accurate representation of the vasculature of the brains and livers of mice at various stages of cancer growth, preserving the length and shape information of the blood vessel and its branches.
This new development was done using a technique called end-point constraints, with such end-points critical in preserving the geometrical features of new blood vessels, including branching patterns and the lengths of terminal vessels.
The accurate quantification of vasculature changes, particularly the number of terminal vessel branches, can play a critical role in accurate assessment and treatment of cancer.
Cancer Council Australia CEO Professor Sanchia Aranda said the capacity of cancers to form new blood vessels (angiogenesis) was a critical feature enabling cancer cells to spread to other parts of the body.
“This exciting project seeks to bring to life the tumour micro-environment through 3D synchrotron images of the vessels and will help to advance our understanding of this critical cancer progression process,” Professor Aranda said.
“The hope is that through improved understanding new opportunities to disrupt angiogenesis will be identified and open pathways to new treatments.
“If we can stop cancers spreading we can reduce the number of people who die from the disease.”
“Our robust algorithms for the early detection and quantification of angiogenesis could potentially be a great step forward in the detection and treatment of cancer,” said lead researcher Dr Dadong Wang.
Main image: A comparison of the enhanced blood vessel prior to skeletonisation (LHS) and after the end-point skeletonisation process (RHS).