Drift correction for mixed reality in surgical environments
AssigneesArthrex, Inc., The Johns Hopkins University
InventorsWenhao Gu, Mathias Unberath, Jonathan Knopf
Life of this patent
Estimating the term…
Abstract
A method for tracking patient features in a surgical field includes capturing an image stream in a field of view demonstrating patient features of a patient anatomy. A first feature set is identified in a tracking region of the field of view of the image stream. The relative motion among a first plurality of the image features in the first feature set is monitored in the image stream over time. Moving features from the first features set are filtered to generate a second feature set. The second feature set defines tracking features. The location and orientation of the patient anatomy is then tracked based on the feature locations of the plurality of tracking features.
Claims21
A method for tracking patient features in a surgical field, the method comprising:
- capturing an image stream with a camera, the image stream comprising a plurality of frames in a field of view demonstrating image features representative of the patient features of a patient anatomy;
- identifying a first feature set comprising a first plurality of the image features in a tracking region of the field of view of the image stream;
- monitoring a relative motion among the first plurality of the image features in the image stream over time;
- identifying a plurality of tracking features as a second feature set comprising a second plurality of the image features by filtering at least one moving feature of the first plurality of the image features in response to the relative motion of the at least one moving feature;
- and
- tracking a location and orientation of the patient anatomy among feature locations of the plurality of tracking features.
A surgical visualization system comprising:
- at least one camera configured to capture an image stream, the image stream comprising a plurality of frames in a field of view demonstrating image features representative of the patient features of a patient anatomy;
- a display comprising a screen positioned between a user and the field of view captured by the camera;
- and
- at least one controller in communication with that at least one camera and the display, the at least one controller configured to:
- identify a first feature set comprising a first plurality of the image features in a tracking region of the field of view of the image stream;
- monitor a relative motion among the first plurality of the image features in the image stream over time;
- identify a plurality of tracking features as a second feature set comprising a second plurality of the image features by filtering at least one moving feature of the first plurality of the image features in response to the relative motion of the at least one moving feature;
- track a location and orientation of the patient anatomy among feature locations of the plurality of tracking features;
- and
- generate display information presented on the display in a predetermined relationship to the patient anatomy.
A method for tracking patient anatomy in a surgical field, the method comprising:
- capturing an image stream with a camera, the image stream comprising a plurality of frames in a field of view demonstrating image features representative of the patient anatomy;
- identifying a first feature set comprising a first plurality of the image features in a tracking region of the field of view of the image stream;
- identifying one or more of the image features defining a disruptive object, wherein the disruptive object is identified as a predetermined object or class of objects that change in position or form relative to the patient anatomy over time;
- identifying a plurality of tracking features as a second feature set comprising a second plurality of the image features by filtering the image features of the disruptive object from the first plurality of the image features;
- and
- tracking a location or orientation of the patient anatomy based on the plurality of tracking features.
Dependencies are inferred from the claim text.