Patient Registration For Total Hip Arthroplasty Procedure Using Pre-Operative Computed Tomography (CT), Intra-Operative Fluoroscopy, and/Or Point Cloud Data
A system for computer assisted navigation during surgery includes a computer platform that operates to register a target surgical area of a patient. In certain cases, a process includes: obtaining a pre-op CT image of a pelvic region of a patient and intra-operatively obtaining a point cloud data about the pelvic region with a navigated instrument, generating a 3D bone model which excludes non-targeted area such as a femur, and then merging the 3D bone model to the point cloud to register the target surgical area.
1 . A system for computer assisted navigation during a surgery, comprising a computer platform operative to:
obtain a computed tomography (CT) image of a pelvic region of a patient captured prior to the surgery, wherein the CT image of the pelvic region includes a target surgical area and a non-target surgical area,
generate from the CT image a three-dimensional (3D) CT volume of the target surgical area which excludes the non-target surgical area;
obtain a fluoroscopy image of the pelvic region of the patient captured during the surgery,
merge the 3D CT volume with the fluoroscopy image, and
register the target surgical area based on the merged 3D CT volume and fluoroscopy image.
2 . The system of claim 1 , wherein the surgery includes a total hip arthroplasty.
3 . The system of claim 1 , wherein the target surgical area includes the acetabulum.
4 . The system of claim 1 , wherein the fluoroscopy image includes a plurality of fluoroscopy images of the pelvic region that are orthogonal to each other.
5 . The system of claim 1 , wherein the fluoroscopy image includes the non-target surgical area.
6 . The system of claim 1 , wherein the non-target surgical area includes at least a portion of a femur of the patient.
7 . The system of claim 1 , wherein the CT image contain an image of the patient in a supine position.
8 . The system of claim 1 , wherein applying the set of image merge rules includes applying at least one rule identifying obstructing anatomy in the CT image.
9 . The system of claim 1 , wherein the merge operation includes:
segmenting the CT image using a first artificial neural network (ANN) to identify a volume of a plurality of bones in the CT image, and
generating a 3D bone model for the pelvis which excludes the femur.
10 . The system of claim 9 , wherein the fluoroscopy image includes at least two fluoroscopy images of the pelvic region.
11 . The system of claim 10 , wherein the merge operation further includes:
merging the 3D bone model with the at least two fluoroscopy images of the pelvic region using a second artificial neural network (ANN).
12 . The system of claim 11 , wherein merging the 3D bone model includes:
generating synthetic fluoroscopy images at various projected angles as digitally reconstructed radiographs (DRRs) and comparing the DRRs to the at least two fluoroscopy images,
identifying a best match between the DRRs and the fluoroscopy images, and
registering the location of the target surgical area based on the best match.
13 . The system of claim 11 , wherein the first ANN includes a trained machine learning algorithm that is trained by providing a CT image dataset including annotated target surgical areas and non-target surgical areas.
14 . The system of claim 11 , wherein the second ANN includes a trained machine learning algorithm that is trained by providing a fluoroscopy image data set including annotated target surgical areas and non-target surgical areas.
15 . The system of claim 1 , wherein registering the target surgical area includes generating a registration matrix of the target surgical area.
16 . The system of claim 15 , wherein the computer platform further displays the registration matrix on a user interface in a surgical area during the surgery.
17 . The system of claim 1 , wherein the computer platform is further operative to:
generate a model of the target surgical area based on the registered location.
18 . A computer program product comprising a non-transitory computer readable medium storing instructions executable by at least one processor to perform operations for computer assisted navigation during surgery to:
obtain a computed tomography (CT) image of a pelvic region of a patient captured prior to the surgery, wherein the CT image of the pelvic region includes a target surgical area and a non-target surgical area,
generate from the CT image a three-dimensional (3D) CT volume of the target surgical area which excludes the non-target surgical area;
obtain a fluoroscopy image of the pelvic region of the patient captured during the surgery,
merge the 3D CT volume and the fluoroscopy image, and
register the target surgical area based on the merged 3D CT volume and fluoroscopy image.
19 . The computer program product of claim 18 , wherein the merge includes:
segmenting the CT image using a first artificial neural network (ANN) to identify a volume of a bone in the CT image, and
generating the 3D volume for the identified volume,
wherein the fluoroscopy image includes at least two fluoroscopy images of the pelvic region.
20 . The computer program product of claim 19 ,
wherein the target surgical area includes the acetabulum,
wherein the fluoroscopy image includes a plurality of fluoroscopy images of the pelvic region of the patient captured during the surgery and including the non-target surgical area,
wherein the non-target surgical area includes at least a portion of a femur of the patient, and
wherein the CT image contain an image of the patient in a supine position.