Practical Applications of Artificial Intelligence in Spine Imaging: A Review

Mo-fi-disc scoring system: Towards understanding the radiological tools to better delineate the disease process and enhancing our solutions for low back pain in artificial intelligence era

BACKGROUND

'Mo-fi-disc' is a new scoring system that quantifies degeneration of the lumbar spine and predicts the intensity of low back pain (LBP). However, its association with LBP-related disability is unknown. In the present study, we aimed to analyze whether 'Mo-fi-disc' scoring system could predict LBP-related disability and distinguish patients with LBP from asymptomatic subjects, while the spine medicine marching towards the era of artificial intelligence (AI).

METHODS

This is a cross-sectional analysis of a prospectively collected database. We included age-, gender-, and BMI-matched 132 subjects (patients: 66, asymptomatic subjects: 66). Modic changes (Mo), fatty infiltration in the paraspinal muscles (fi), and intervertebral disc degeneration (disc) were evaluated using 'Mo-fi-disc' scoring system on lumbar spine magnetic resonance imaging. Pain and disability were evaluated with visual analogue scale (VAS) and Oswestry disability index (ODI), respectively.

RESULTS

A Mo-fi-disc score of 5.5 was the most appropriate cut-off value. Mo-fi-disc scoring system had an OR of 1.79 in distinguishing patients with LBP from asymptomatic subjects. One point increment in VAS and ODI had ORs of 1.82 and 1.13 for predicting higher Mo-fi-disc scores.

CONCLUSION

'Mo-fi-disc' scoring system is a useful tool depicting intensity of LBP and LBP-related disability. The cut off value of Mo-fi-disc score is 5.5 to distinguish patients with LBP from asymptomatic subjects. This scoring system, with progressive improvement of its faults, could help clinicians to select appropriate patients for conservative and surgical management in the very near future, in AI-based spine medicine.

IRB APPROVAL NO

ATADEK 2019-12/4.