Prediction of candidates for spinal neuromodulation
Described herein are various implementations of systems and methods for determining likelihood of a patient favorably responding to a neuromodulation procedure based on a quantitative or objective score or determination based on a plurality of indicators of pain (e.g., chronic low back pain stemming from one or more vertebral bodies or vertebral endplates of a patient). The systems and methods may involve application of artificial intelligence techniques (e.g., trained algorithms, machine learning or deep learning algorithms, and/or trained neural networks).
1 . A computer-implemented method of quantitatively predicting likelihood that a particular subject would respond favorably to basivertebral nerve ablation to treat back pain, the computer implemented method comprising:
receiving one or more digital images of at least a portion of a spine of the particular subject;
pre-processing the one or more digital images including denoising and segmenting the one or more digital images;
extracting features from the one or more pre-processed one or more digital images to identify one or more pre-Modic change characteristics, wherein the pre-Modic change characteristics provide initial indications or precursors of potential Modic changes, wherein the extracting features comprises applying a trained neural network to the one or more pre-processed one or more digital images to automatically identify the pre-Modic change characteristics, wherein the trained neural network is trained using digital images from subjects who previously received basivertebral nerve ablation treatment with annotations indicating treatment success or failure; and
determining a score indicative of the likelihood that the particular subject would respond favorably to basivertebral nerve ablation based on the identified pre-Modic change characteristics; and
wherein the computer-implemented method is performed by at least one processor executing computer-executable instructions stored in at least one memory device.
2 . The computer-implemented method of claim 1 , wherein the one or more digital images comprise at least one of magnetic resonance images (MRIs), computed tomography (CT) images, X-ray images, and fluoroscopic images.
3 . The computer-implemented method of claim 1 , wherein the pre-Modic change characteristics comprise at least one of mechanical characteristics or bacterial characteristics.
4 . The computer-implemented method of claim 3 , wherein the mechanical characteristics comprise at least one of loss of soft nuclear material in an adjacent intervertebral disc of the vertebral body, reduced disc height, reduced hydrostatic pressure, microfractures, fissures, spondylodiscitis, Schmorl's nodes, or osteitis.
5 . The computer-implemented method of claim 3 , wherein the bacterial characteristics comprise at least one of detection of bacteria that have entered an intervertebral disc adjacent to a vertebral body, a disc herniation or annulus tear, inflammation, or new capilarisation.
6 . The computer-implemented method of claim 1 , wherein the one or more digital images comprise at least one of T1-weighted MRIs, T2-weighted MRIs, fat suppression MRIs, ultrashort time-to-echo MRIs, and Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) MRIs.
7 . The computer-implemented method of claim 1 , wherein identifying the one or more pre-Modic change characteristics from the one or more digital images comprises applying one or more filters to the one or more digital images.
8 . The computer-implemented method of claim 1 , wherein the score is determined based on at least one of a quantity, a severity, or a distribution of the pre-Modic change characteristics.
9 . The computer-implemented method of claim 1 , further comprising displaying the score on a display device.
10 . The computer-implemented method of claim 1 , further comprising comparing the score to a predetermined threshold to determine the likelihood that the particular subject would respond favorably to basivertebral nerve ablation.
11 . A computer-implemented method of quantitatively predicting likelihood that a particular subject would respond favorably to basivertebral nerve ablation to treat back pain, the method comprising:
receiving one or more digital images of at least a portion of a spine of the particular subject;
pre-processing the one or more digital images including denoising and segmenting the one or more digital images;
applying one or more filters to the pre-processed digital images to identify one or more pre-Modic change characteristics, wherein the pre-Modic change characteristics provide initial indications or precursors of potential Modic changes, wherein the applying one or more filters comprises applying a trained neural network to the one or more digital images to automatically identify the pre-Modic change characteristics, wherein the trained neural network is trained using digital images from subjects who previously received basivertebral nerve ablation treatment with annotations indicating treatment success or failure; and
determining a score indicative of the likelihood that the particular subject would respond favorably to basivertebral nerve ablation based on at least one of a quantity, a severity, or a distribution of the identified pre-Modic change characteristics; and
wherein the computer-implemented method is performed by at least one processor executing computer-executable instructions stored in at least one memory device.
12 . The computer-implemented method of claim 11 , wherein the one or more digital images comprise at least one of magnetic resonance images (MRIs), computed tomography (CT) images, X-ray images, and fluoroscopic images.
13 . The computer-implemented method of claim 11 , wherein the pre-Modic change characteristics comprise at least one of mechanical characteristics or bacterial characteristics.
14 . The computer-implemented method of claim 13 , wherein the mechanical characteristics comprise at least one of loss of soft nuclear material in an adjacent intervertebral disc of the vertebral body, reduced disc height, reduced hydrostatic pressure, microfractures, fissures, spondylodiscitis, Schmorl's nodes, or osteitis.
15 . The computer-implemented method of claim 13 , wherein the bacterial characteristics comprise at least one of detection of bacteria that have entered an intervertebral disc adjacent to a vertebral body, a disc herniation or annulus tear, inflammation, or new capilarisation.
16 . The computer-implemented method of claim 11 , wherein the one or more digital images comprise at least one of T1-weighted MRIs, T2-weighted MRIs, fat suppression MRIs, ultrashort time-to-echo MRIs, and Iterative Decomposition of water and fat with Echo Asymmetry and Least-squares estimation (IDEAL) MRIs.
17 . The computer-implemented method of claim 11 , further comprising displaying the score on a display device.
18 . The computer-implemented method of claim 11 , further comprising comparing the score to a predetermined threshold to determine the likelihood that the particular subject would respond favorably to basivertebral nerve ablation.
19 . A computer-implemented method of quantitatively predicting likelihood that a particular subject would respond favorably to basivertebral nerve ablation to treat back pain, the method comprising:
receiving one or more digital images of at least a portion of a spine of the particular subject;
pre-processing the one or more digital images by denoising and segmenting the one or more digital images;
extracting features from the one or more digital images to identify one or more pre-Modic change characteristics, wherein the pre-Modic change characteristics provide initial indications or precursors of potential Modic changes, wherein the extracting features comprises applying a trained neural network to the one or more digital images to automatically identify the pre-Modic change characteristics, wherein the trained neural network is trained using digital images from subjects who previously received basivertebral nerve ablation treatment with annotations indicating treatment success or failure; and
determining a score indicative of the likelihood that the particular subject would respond favorably to basivertebral nerve ablation based on the identified pre-Modic change characteristics; and
comparing the score to a predetermined threshold to identify the particular subject as a candidate for basivertebral nerve ablation if the score is above the predetermined threshold; and
wherein the computer-implemented method is performed by at least one processor executing computer-executable instructions stored in at least one memory device.
20 . The computer-implemented method of claim 19 , wherein the score is determined based on at least one of a quantity, a severity, or a distribution of the pre-Modic change characteristics.