Spinal surgery outcome prediction
A spinal surgery training process includes the steps of capturing a plurality of 2D images for each of a plurality of spines, generating a curve of each spine from the respective 2D images based on locations of select vertebrae in each of the spines, grouping the spines into one of a number of groups based on similarity to produce groups of spines having similarities, performing the capturing, generating, determining and grouping steps at least once prior to surgery and at least once after surgery to produce pre-operative groups and their resultant post-operative groups, and assigning surgical methods and a probability to each of the post-operative groups indicating the probability that a spinal shape of the post-operative group can be achieved using the surgical methods. An outcome prediction process for determining surgical methods can be implemented once the training process is complete.
1. A spinal surgery training system comprising:
a processor configured to:
generate a 3D curve of each of a plurality of spines based on a plurality of pre-operative 2D images of each spine;
apply a statistical clustering method to group each of the plurality of spines into one of a number of clusters based on similarity of the 3D curves; and
for each cluster, determine, based on one or more surgical techniques used to treat the spines in the cluster and associated post-operative results for each surgical technique, probabilities of achieving each of the post-operative results using each surgical technique.
2. The spinal surgery training system of claim 1 , further comprising an imaging device configured to capture the plurality of pre-operative 2D images for each of the plurality of spines.
3. The spinal surgery training system of claim 2 , wherein:
the imaging device is an X-ray machine, a magnetic resonance imaging machine, or an ultrasound machine; and
the imaging device is configured to capture the plurality of pre-operative 2D images from a plurality of different perspectives.
4. The spinal surgery training system of claim 1 , wherein the processor is configured to determine probabilities of achieving each of the post-operative results based on fusion levels of each surgical technique.
5. The spinal surgery training system of claim 1 , wherein the processor is configured to determine probabilities of achieving each of the post-operative results based on post-operative flexibility of each spine.
6. The spinal surgery training system of claim 1 , wherein the processor is configured to generate the 3D curve by:
determining landmarks on each of a plurality of vertebrae based on the plurality of pre-operative 2D images;
determining locations of each of the plurality of vertebrae based on the determined landmarks; and
determining an alignment and an orientation of each of the plurality of vertebrae based on the determined locations.
7. The spinal surgery training system of claim 1 , wherein the processor is further configured to select a surgical technique based on the probabilities of achieving each of the post-operative results using each surgical technique.
8. The spinal surgery training system of claim 7 , wherein the processor is further configured to instruct a surgical robot to perform spinal surgery according to the selected surgical technique.
9. The spinal surgery training system of claim 8 , wherein the processor is integrated into the surgical robot.
10. A method of training an outcome-prediction system, the method comprising:
receiving a plurality of pre-operative 2D images for each of a plurality of spines;
generating a 3D curve of each spine based on the plurality of pre-operative 2D images;
applying a statistical clustering method to group each of the plurality of spines into one of a number of pre-operative clusters based on similarity of the 3D curves; and
for each pre-operative cluster, determining, based on one or more surgical techniques used to treat the spines in the pre-operative cluster and associated post-operative results for each surgical technique, probabilities of achieving each of the post-operative results using each surgical technique.
11. The method of claim 10 , wherein determining, for each pre-operative cluster, the probabilities of achieving each of the post-operative results using each surgical technique comprises:
receiving, for each spine in the pre-operative cluster, a plurality of post-operative images;
applying a statistical clustering method to group each of the plurality of spines into one of a number of post-operative clusters based on similarity of the post-operative images, each of the post-operative clusters having an associated post-operative result; and
determining the probabilities of achieving each of the post-operative results based on a treatment paths from the pre-operative cluster to the post-operative clusters.
12. The method of claim 11 , wherein applying the statistical clustering method to group each of the plurality of spines into one of the number of post-operative clusters further comprises applying the statistical clustering method based on fusion levels of each spine.
13. The method of claim 11 , wherein applying the statistical clustering method to group each of the plurality of spines into one of the number of post-operative clusters further comprises applying the statistical clustering method based on post-operative flexibility of each spine.
14. A spinal surgery outcome prediction system comprising:
a processor configured to:
receive at least two 2D pre-operative images of a spine;
receive a plurality of 3D curves, each 3D curve having associated probabilities of achieving post-operative results using one or more surgical techniques;
generate at least two 2D curves from each received 3D curve;
compare features of the at least two 2D pre-operative images to the generated 2D curves to match the spine to one of the 3D curves; and
based on the probabilities associated with the matched 3D curve, provide a probability of achieving a post-operative result using at least one surgical technique.
15. The spinal surgery outcome prediction system of claim 14 , further comprising an imaging device configured to capture the at least two 2D pre-operative images of the spine.
16. The spinal surgery outcome prediction system of claim 14 , wherein the processor is configured to compare the features of the at least two 2D pre-operative images to the generated 2D curves by:
determining a landmark on each of a plurality of vertebrae in the 2D pre-operative images; and
matching the determined landmarks against vertebra in the generated 2D curves.
17. The spinal surgery outcome prediction system of claim 16 , wherein the landmark comprises a center point of the vertebra.
18. The spinal surgery outcome prediction system of claim 14 , wherein the processor is configured to receive the plurality of 3D curves by:
receiving a plurality of pre-operative 2D images for each of a plurality of spines;
generating a 3D curve of each spine based on a plurality of captured pre-operative 2D images;
applying a statistical clustering method to group each of the plurality of spines into one of a number of clusters based on similarity of the 3D curves; and
for each cluster, determine, based on surgical techniques used to treat the spines in the cluster and associated post-operative results for each surgical technique, probabilities of achieving each of the post-operative results associated with using each surgical technique.
19. The spinal surgery outcome prediction system of claim 18 , wherein the processor is configured to apply the statistical clustering method based on fusion levels of each spine.
20. The spinal surgery outcome prediction system of claim 18 , wherein the processor is configured to apply the statistical clustering method based on post-operative flexibility of each spine.