Many people suffer from lower back pain, a condition likely to become more prevalent due to our growing and ageing population. “This leads to increased demand for healthcare and rising costs, making it essential to gain greater insight into the factors that influence back pain,” says Eddo Wesselink.
In recent years, attention has grown for the decline in the health of our back muscles, whether due to loss of muscle mass or fat infiltration, as a potential cause of lower back pain. However, it remains difficult to determine precisely what this means, as there is no clear standard for what constitutes normal muscle health, nor are there efficient and reliable methods of measurement available.
Computer vision
To address this, Wesselink chose to use artificial intelligence in his PhD research at VU Amsterdam to assess muscle health via MRI scans. He developed computer vision models to analyse large and diverse datasets.
The research shows that a decline in muscle health does not always indicate a problem; it can also be observed in individuals without pain. However, it is more prevalent among those with chronic lower back pain compared to those without. Furthermore, muscle health appears to be a predictor of recovery following surgery in patients with central canal stenosis, a condition that causes leg pain, particularly during walking. There is also moderate evidence suggesting that fat infiltration in the back muscles cannot be reduced through exercise therapy.
Publicly available
“Assessing muscle health with MRI scans is extremely time-consuming, which often limits researchers to small study groups or forces them to use less reliable methods,” Wesselink explains. “We have made our AI model publicly available, enabling researchers to conduct detailed analyses of muscle health from MRI scans in a highly efficient manner.”
Additionally, Wesselink has laid the groundwork for a reference standard for muscle health. “If further developed, this could allow healthcare professionals to assess muscle health in relation to what is considered normal for a person’s age, sex, and BMI.” This contributes to a better understanding of muscle health in individuals with lower back pain and could ultimately lead to more personalised and effective care.