A new study from a team of Australian researchers tackled the highly intriguing subject of applying artificial intelligence algorithms to the difficult problem of degenerative disc pathology and possible treatments—in a notably novel way.
Authors Mauro Allini, Ph.D. (AO Research, Davos, Switzerland), Ashish Diwan, Ph.D., FRACS, FAOrthA (Department of Orthopedic Surgery, University of New South Wales), W. Mark Erwin, D.C., Ph.D. (Sydney University Faculty of Medicine and Health), Christopher B. Little, Ph.D. (Sydney University Faculty of Medicine and Health) and James Melrose, Ph.D. (University of New South Wales) applied artificial intelligence (AI) algorithms to various animal models of degenerative disc disease to tease out which ones might have predictive value for humans as well as understanding the progression of disc pathology itself.
To begin with, the team noted that degenerative disc disease, referred to as intervertebral disc degeneration (IVD) in the study, is extremely complex. While IVD is prevalent in most patients over 65 years of age, it is not always symptomatic.
Why do spine discs degenerate and why are they sporadically symptomatic?
What is required, the Australian team wrote, is a paradigm shift in how these fundamental questions of disc degeneration and pain are studied. Specifically, to apply artificial intelligence algorithms to histologies of degenerating discs from various animal models.
Their ambitious goals were to uncover patterns of the disc degeneration process heretofore undetected by humans which would then improve our understanding of IVD degeneration leading to novel and more effective potential treatments.
“Implementation of AI in the evaluation of histology data has improved the usefulness of a popular murine IVD model and could also be used in an ovine histopathological grading scheme that has been used to quantify degenerative IVD changes and stem cell mediated regeneration,” wrote the authors.
In addition, the research team posited that animal models, augmented by the application of AI, could effectively evaluate new therapies (antioxidant compounds, for example) for tamping down disc inflammation, degeneration, and patient pain.
One notably intriguing way to use AI in these animal models, noted the team, is as an animal facial recognition pain assessment tool. In other words, correlating the level of pain relief in, for example, a dog, pig, alpaca, or kangaroo model, by mapping changes in facial muscles.
The full study (”An update on animal models of intervertebral disc degeneration and low back pain: Exploring the potential of artificial intelligence to improve research analysis and development of prospective therapeutics“),was published six months ago in the online Journal of Orthopedic Research, Spine.
Finally, after reviewing a number of animal models (and we’ve reproduced one of the charts from this study above) the research team led by Dr. Mauro Allini and Dr. James Melrose, said:
“While currently there is a disconnect to some extent between AI and the novel therapies covered in this review the sheer depth of these studies outlining the beneficial tissue and cell protective properties of the many listed compounds and the illustration of how they might potentially be applied to IVD regeneration is highly suggestive of the likelihood that they may find future application. For this to become a reality more reliable evaluative methods are required for the assessment of IVD tissues and aspects of their regeneration in animal models. We consider that the AI methodology we have described in this review will provide such an improvement that may well allow the therapeutic potential of these novel compounds to be realized in future studies.”
Stay tuned, for sure.
Study funding, incidentally, came from the National Health and Medical Research Council project grants in Australia.
