Real-time 3-D ultrasound reconstruction of knee and its implications for patient specific implants and 3-D joint injections
Methods and apparatus for treating a patient. The method includes acquiring a plurality of radio frequency (RF) signals with an ultrasound transducer, each RF signal representing one or more return echoes from a scan line of a pulse-mode echo ultrasound scan. A position of the ultrasound transducer corresponding to each of the acquired RF signals is determined, and a plurality of contour lines generated from the plurality of RF signals. The method estimates a 3-D shape and position of an anatomical feature, such as a joint of patient based on the generated contour lines and corresponding ultrasound transducer positions. An apparatus, or computer includes a processor and a memory with instructions that, when executed by the processor, perform the aforementioned method.
1. A method of generating a 3D patient-specific bone model, the method comprising:
acquiring a plurality of raw signals from an ultrasound scan of a patient's bone using an ultrasound probe;
tracking the acquiring of the raw signals in three dimensional (“3D”) space and generating corresponding tracking data;
transforming each raw signal into an envelope comprising a plurality of peaks by applying an envelope detection algorithm to each signal, each peak corresponding with a tissue interface echo;
identifying a bone echo from the tissue interface echoes of each raw signal;
determining a two dimensional (“2D”) bone contour from the plurality of bone echoes corresponding to each scan location of the ultrasound probe, thereby resulting in 2D bone contours;
transforming the 2D bone contours into an integrated 3D point cloud using the tracking data; and
deforming a non-patient specific 3D bone model corresponding to the patient's bone in correspondence with the integrated 3D point cloud to generate a 3D patient-specific bone model.
2. The method of claim 1 , wherein applying an envelope detection algorithm to each signal comprises applying a moving power filter to each raw signal.
3. The method of claim 1 , wherein tracking the acquisition includes using at least one of an optical position tracking system and an electromagnetic position tracking system.
4. The method of claim 1 , wherein the non-patient specific 3D bone model is utilized to filter noise by thresholding for a distance between a respective point of the integrated 3D point cloud and the non-patient specific 3D bone model.
5. The method of claim 1 , further comprising identifying the 2D bone contour by removing portions of the bone echo that deviate from a continuous portion of the bone echo.
6. The method of claim 1 , wherein the non-patient specific 3D bone model is an average bone model of a plurality of bone models in a statistical atlas.
7. The method of claim 1 , wherein transforming the 2D bone contours into an integrated 3D point cloud further comprises:
transforming the 2D bone contours into a plurality of registered 3D partial point clouds; and,
integrating the plurality of registered 3D partial point clouds to form the integrated 3D point cloud.
8. The method of claim 1 , wherein deforming the non-patient specific 3D bone model comprises comparing the non-patient specific 3D bone model with the 3D point cloud and, based on the comparing, deforming the non-patient specific 3D bone model to match the 3D point cloud.
9. The method of claim 8 , wherein the comparing and deforming are iteratively performed until the comparing results in a deviation that is less than a deviation threshold.
10. The method of claim 1 , wherein the 3D patient-specific bone model includes at least one of a 3D patient-specific model of a bone, a 3D patient-specific model of a joint, and a 3D patient-specific model of cartilage.
11. The method of claim 1 , wherein the ultrasound scan comprises an A-mode ultrasound scan.
12. The method of claim 1 , wherein the ultrasound scan occurs over a plurality of locations the patient's bone.
13. The method of claim 1 , wherein identifying the bone echo includes selecting a last peak having a normalized envelope amplitude above a preset threshold.
14. The method of claim 13 , wherein the envelope amplitude is normalized with respect to a maximum peak existing in the envelope.
15. A method of generating a 3D patient-specific bone model, the method comprising:
acquiring a plurality of signals from an ultrasound scan of a patient's bone and generating tracking data in three dimensional (“3D”) space corresponding to acquisition of the plurality of signals;
transforming each of the plurality of signals into an envelope comprising a plurality of peaks, each peak corresponding with a tissue interface echo;
identifying bone echoes from tissue interface echoes;
determining a plurality of two dimensional (“2D”) bone contours from the bone echoes;
transforming the 2D bone contours into a 3D point cloud using the tracking data; and,
deforming a non-patient specific 3D bone model using the 3D point cloud to generate a 3D patient-specific bone model.
16. The method of claim 15 , further comprising identifying the bone contours by removing portions of the bone echoes that deviate from a continuous portion of the bone echoes.
17. The method of claim 15 , wherein the non-patient specific 3D bone model is an average bone model of a plurality of bone models in a statistical atlas.
18. The method of claim 15 , wherein transforming the 2D bone contours into a 3D point cloud further comprises:
transforming the 2D bone contours into a plurality of registered 3D partial point clouds; and,
integrating the plurality of registered 3D partial point clouds to form the integrated 3D point cloud.
19. The method of claim 15 , wherein deforming the non-patient specific 3D bone model comprises comparing the non-patient specific 3D bone model with the 3D point cloud and, based on the comparing, deforming the non-patient specific 3D bone model to match the 3D point cloud.
20. The method of claim 15 , wherein identifying the bone echoes includes selecting peaks having a normalized envelope amplitude above a preset threshold.