Identification of potentially painful disc fissures in magnetic resonance images using machine-learning modelling.

PURPOSE: It is suggested that non-specific low back pain (LBP) can be related to nerve ingrowth along granulation tissue in disc fissures, extending into the outer layers of the annulus fibrosus. Present study aimed to investigate if machine-learning modelling of magnetic resonance imaging (MRI) data can classify such fissures as well as pain, provoked by discography, with plausible accuracy and precision. METHODS: The study was based on previously collected data from 30 LBP patients (age = 26-64 years, 11 males). Pressure-controlled discography was performed in 86 discs with pain-positive discograms, categorized as concordant pain-response at a pressure ≤ 50 psi and for each patient one negative control disc. The CT-discograms were used for categorization of fissures. MRI values and standard deviations were extracted from the midsagittal part and from 5 different sub-regions of the discs. Machine-learning algorithms were trained on the extracted MRI markers to classify discs with fissures extending into the outer annulus or not, as well as to classify discs as painful or non-painful. RESULTS: Discs with outer annular fissures were classified in MRI with very high precision (mean of 10 repeated testings: 99%) and accuracy (mean: 97%) using machine-learning modelling, but the pain model only demonstrated moderate diagnostic accuracy (mean accuracy: 69%; precision: 71%). CONCLUSION: The present study showed that machine-learning modelling based on MRI can classify outer annular fissures with very high diagnostic accuracy and, hence, enable individualized diagnostics. However, the model only demonstrated moderate diagnostic accuracy regarding pain that could be assigned to either a non-sufficient model or the used pain reference.

Assessment of lumbar disc herniaton using fractional anisotropy in diffusion tensor imaging along with conventional T2-weighted imaging.

OBJECTIVE: To assess the usefulness of diffusion tensor imaging and its fractional anisotropy map along with conventional T2-weighted imaging in evaluating the anisotropic water diffusion variations of annulus fibres involved in herniation disc pathology. MATERIALS AND METHODS: Seventy-five patients with previous medical ethics committee approval and informed consent experiencing low back pain were selected for this prospective randomised blinded trial. Lumbar disc fractional anisotropy maps were obtained acquiring diffusion tensor sequences on a 3T machine. The matrix of nucleus pulposus and structures of annulus fibres were analysed using fractional anisotropy textural features to highlight any presence of lumbar disc herniation. Observer variability and reliability between two neuroradiologists were evaluated. The χ2 test, two-tailed t test and linear regression analysis were used to focus differences in patients' demographic data and magnetic resonance imaging findings. RESULTS: Annular fissures with extrusions were identified using diffusion tensor imaging in 10 out of 17 discs (study group) previously assessed as bulging discs using conventional magnetic resonance imaging. Eighteen extrusions out of 39 (study group) disc levels were identified on diffusion tensor imaging compared to eight extrusions highlighted on T2-weighted imaging (P < 0.01). All eight (study group) disc extrusions evaluated on T2-weighted imaging showed annular fissures on diffusion tensor imaging. Seven out of 14 (study group) protrusions highlighted on T2-weighted imaging had no annular fissures on diffusion tensor imaging; thirty-six disc levels in the control group had no evidence of annular fissures on diffusion tensor imaging (P > 0.01). CONCLUSIONS: The addition of diffusion tensor imaging sequences and fractional anisotropy mapping to a conventional magnetic resonance imaging protocol could be useful in detecting annular fissures and lumbar disc herniation.

Morphologic features of spontaneous annular tears and disc degeneration in the aging sand rat (Psammomys obesus obesus).

The sand rat, a member of the gerbil family, is a valuable small animal model in which intervertebral disc degeneration occurs spontaneously as the animal ages. Radiographic features of cervical and lumbar degeneration resemble those in human spines. We conducted a retrospective analysis of spines of 140 animals 3-41 months old focusing specifically on the presence of annular tears that are not visible by radiography and have not been described previously in the sand rat disc. During degeneration of the nucleus pulposus, notochordal cell death occurs and granular material, which stains with Alcian blue for proteoglycans, accumulates. Lamellar architecture also deteriorates and annular tears occur that are morphologically similar to the concentric, radiating and transdiscal annular tears in human discs. These tears contain granular material that provides a "marker" that can be used to distinguish the annular tears from artefactual separations during sectioning. We observed lamellar degeneration and separation in the annulus fibrosus at 4 months with associated tears that contained granular material in the nucleus. Tears that contained granular material and displacement of the degenerating nucleus were common in cervical and lumbar discs of animals older than 9 months; some specimens showed tears at 4 and 5 months. With advanced degeneration, granular globules were displaced dorsally adjacent to and into the spinal cord area and also ventrally into regions where osteophytes formed. We present morphologic data that expand the utility of this rodent model of spontaneous age-related disc degeneration and provide novel information on annular tears and disc degeneration.