Machine learning and deep learning are already showing promise as tools to aid in the diagnosis and management of ankylosing spondylitis (AS), but a new review article highlights that the field is still in its early stages and more research is needed.
Paper, “Machine learning and deep learning for the diagnosis and treatment of ankylosing spondylitis – a scoping review“teeth, Journal of Clinical Orthopaedics and Trauma.
Machine learning is a type of artificial intelligence that works by feeding a computer a large dataset and a set of mathematical rules or algorithms that the computer uses to identify patterns in the data. The computer can then apply the “learned” patterns to understand future datasets. Deep learning is a specific form of machine learning that uses layers of algorithms to mimic the way neurons in the human brain fire.
AI for AS
Here, three Indian scientists reviewed studies published between 2013 and 2023 that explored the use of machine learning and deep learning in AS.
This review highlights over a dozen studies that have investigated these techniques for various applications in AS. Some studies have used machine learning to develop tools to aid in the diagnosis of AS and distinguish it from other inflammatory diseases such as rheumatoid arthritis and osteoarthritis, generally with reasonable accuracy. Other researchers have applied computer-based analyses to track disease progression and predict patient response to various treatments such as biological disease-modifying antirheumatic drugs and TNF inhibitors. Other studies have used deep learning to identify potential biomarkers for the disease.
The researchers concluded that machine learning/deep learning has been “effectively used to identify and screen for AS, and to distinguish it from other similar conditions.”
“These techniques are also useful for predicting and treating the progression of AS,” the researchers said, but noted limitations in the available studies. In particular, most studies used relatively small datasets to train their machine learning algorithms, many of which were from a single center. Because machine learning works by identifying patterns in a dataset, it is generally most accurate when there is a large amount of data to “learn” from. Studies with larger datasets are needed, the researchers said.
