Back to results
US 11,298,195 B2Granted
Anatomical feature identification and targeting
AssigneeAuris Health, Inc.
InventorsMenglong Ye, Yanwei Wang, Elif Ayvali, David Paul Noonan
Life of this patent
Estimating the term…
Abstract
Methods of positioning a surgical instrument can involve advancing a first medical instrument to a treatment site of a patient, the first medical instrument comprising a camera, recording a target position associated with a target anatomical feature at the treatment site, generating a first image of the treatment site using the camera of the first medical instrument, identifying the target anatomical feature in the first image using a pretrained neural network, and adjusting the target position based at least in part on a position of the identified target anatomical feature in the first image.
Claims28
A method of positioning a surgical instrument, the method comprising:
- advancing a first medical instrument to a treatment site of a patient, the first medical instrument comprising a camera;
- recording a target position associated with a target anatomical feature at the treatment site;
- generating a first image of the treatment site using the camera of the first medical instrument;
- identifying a first anatomical feature in the first image using a pretrained neural network;
- after said generating the first image, generating a second image of the treatment site using the camera of the first medical instrument;
- identifying a second anatomical feature in the second image using the pretrained neural network;
- determining that the first anatomical feature of the first image and the second anatomical feature of the second image represent a same feature, said determining involving tracking movement of one or more features between the first image and the second image at least in part by determining that an amount of overlap between at least a portion of the second anatomical feature of the second image and at least a portion of the first anatomical feature of the first image is greater than a predetermined threshold;
- and
- in response to said determining that the first anatomical feature and the second anatomical feature represent the same feature, adjusting the target position based at least in part on a position of the second anatomical feature in the second image.
6 more elements
A method of targeting an anatomical feature, the method comprising:
- receiving target position data indicating a target position associated with a target anatomical feature of a patient;
- storing the target position data in one or more data storage devices;
- receiving a first image of a surgical site within a patient;
- using an artificial neural network to identify a first form in the first image;
- receiving a second image of the surgical site;
- using the artificial neural network to identify a second form in the second image;
- determining an amount of overlap between at least a portion of the second form of the second image and at least a portion of the first form of the first image;
- determining that the amount of overlap is greater than a predetermined threshold;
- determining that the second form of the second image and the first form of the first image both represent the target anatomical feature based at least in part on the determination that the amount of overlap is greater than the predetermined threshold;
- and
- in response to said determining that the second form and the first form represent the target anatomical feature, storing updated target position data based at least in part on a position of the second form in the second image.
8 more elements
A medical system comprising:
- an endoscope having a camera and an electromagnetic position sensor associated with a distal end thereof;
- a robotic medical subsystem including a plurality of articulating arms;
- and
- control circuitry communicatively coupled to the endoscope and the robotic medical subsystem, the control circuitry being configured to:
- advance the endoscope to a treatment site of a patient in response to user input;
- record a position of the distal end of the endoscope within an electromagnetic field;
- generate user interface data indicating the position as a target anatomical position;
- receive a first image from the camera of the endoscope;
- identify a first anatomical feature in the first image using a pretrained neural network;
- receive a second image from the camera of the endoscope;
- identify a second anatomical feature in the second image using the pretrained neural network;
- determine that the second anatomical feature of the second image and the first anatomical feature of the first image represent a common feature;
- estimate a three-dimensional position of the second anatomical feature of the second image based at least in part on the first image and the second image;
- and
- adjust the target anatomical position based at least in part on a position of the second anatomical feature in the second image.
12 more elements
A computing device comprising:
- an endoscope interface;
- and
- control circuitry comprising one or more processors and one or more data storage devices, the control circuitry configured to:
- receive sensor data indicating a position of a distal end of an endoscope coupled to the control circuitry over the endoscope interface;
- generate first user interface data indicating the position of the distal end of the endoscope as a target position;
- receive a first image of a medical site over the endoscope interface;
- identify a first anatomical form in the first image;
- receive a second image of the medical site over the endoscope interface;
- identify a second anatomical form in the second image;
- determine that an amount of overlap of the first form and the second form in an overlay of the first and second images is greater than a predetermined threshold;
- determine that the first anatomical form and the second anatomical form represent a same target anatomical feature;
- and
- update the target position based on said determining that the first anatomical form and the second anatomical form represent the same target anatomical feature.
10 more elements
Dependencies are inferred from the claim text.