Abstract #5037

T2 texture features can detect differences between the discs of patient with and without low back pain

Vahid Abdollah^1,2,3, Eric C Parent⁴, Maria Beketskaia^1,2, and Jeff F. Dunn^1,2,3

¹Radiology, University of Calgary, Calgary, AB, Canada, ²Experimental Imaging Centre, University of Calgary, Calgary, AB, Canada, ³McCaig Institute for Bone and Joint Health, University of Calgary, Calgary, AB, Canada, ⁴Physical Therapy, University of Alberta, Edmonton, AB, Canada

Conventional MRI is often clinically inconclusive in diagnosing the pathology of back pain, as both symptomatic and asymptomatic individuals demonstrate the same abnormal features. We hypothesized that texture driven features could be deployed to identify the underlying pathology of low back pain. Fourteen patients with chronic back pain were matched (age, weight, and gender) with 14 healthy volunteers. A grey-level co-occurrence matrix with one- to four-pixel offset and four directions was constructed to extract texture features. The texture analysis results indicated the discs of the healthy subjects were more uniform (lower contrast) than those of the participants with back pain.

This abstract and the presentation materials are available to members only; a login is required.

Join Here