Vision-based position and orientation determination for endovascular tools

Inventors

Bell, David James • Sganga, Jake Anthony • Kahn, Gregory

Assignees

Remedy Robotics Inc

Interested in licensing this patent?

MTEC can help explore whether this patent might be available for licensing for your application.

Publication Number

US-12121307-B2

Patent

Publication Date

2024-10-22

Expiration Date


Abstract

Systems and methods for vision-based position and orientation determination for endovascular tools are disclosed. In one example, a method includes receiving a two-dimensional medical image including a view of at least a distal portion of a medical instrument, the distal portion of the medical instrument including one or more fiducials positioned thereon, the one or more fiducials being radio-opaque and visible in the medical image. The method also includes detecting, within the medical image, a two-dimensional appearance of the one or more fiducials, and based on the two-dimensional appearance of the one or more fiducials, determining at least one of a roll angle of the distal portion of the medical instrument, and an incline of the distal portion of the medical instrument.

Core Innovation

The invention relates to vision-based pose determination for a medical instrument using a two-dimensional medical image comprising a fluoroscopic image that includes a view of at least a distal portion. The distal portion includes one or more radio-opaque fiducials, and the system detects the distal portion within the two-dimensional medical image. Based on the detected distal portion, the system determines a heading of the medical instrument within an image plane by determining a centerline of the distal portion, determining an endpoint for the centerline, and determining a vector extending from the endpoint that is colinear with a distal portion of the centerline.

The system detects a two-dimensional appearance of the one or more radio-opaque fiducials and determines a roll angle of the distal portion about a longitudinal axis of the medical instrument based on the two-dimensional appearance. Various fiducial arrangements are disclosed to uniquely map two-dimensional projections to roll and out-of-plane/incline-related values, including marker geometries such as ring-shaped markers with ellipse projection, semicircular markers, helical fiducials, non-circumferential ring configurations, and asymmetric ring offsets. The disclosure also includes configurations with multiple dots, stacked rings with alignment gaps, differently sized markers, and sinusoidal or helix fiducials to resolve roll/incline.

Based on the determined roll angle, the system generates one or more motor commands configured to cause a robotic system coupled to the medical instrument to move the medical instrument, and causes the robotic system to move the medical instrument based on the one or more motor commands. The disclosed approach enables automated tool tracking and compares expected vs actual distal motion using fiducial-tracked distal roll/rotation to provide safety safeguards. The disclosure highlights determining additional distal tip position and navigation-relevant quantities such as heading using centerline and endpoint-based geometry.

Claims Coverage

The document contains two independent claims, a system claim and a method claim. Both claims include the same core inventive features: receiving a fluoroscopic two-dimensional image with distal radio-opaque fiducials, detecting the instrument distal portion, determining heading within the image plane using a centerline and a colinear vector, detecting a two-dimensional appearance of the fiducials, determining a roll angle about a longitudinal axis based on that appearance, and generating motor commands to move a robotic system coupled to the instrument.

Fluoroscopic two-dimensional fiducial-based instrument heading and roll determination for robotic control

Receive a two-dimensional medical image comprising a fluoroscopic image that includes a view of at least a distal portion of a medical instrument, the distal portion comprising one or more radio-opaque fiducials; detect the distal portion within the two-dimensional medical image; based on the detected distal portion, determine a heading of the medical instrument within an image plane by determining a centerline of the distal portion, determining an endpoint for the centerline, and determining a vector extending from the endpoint that is colinear with a distal portion of the centerline; detect a two-dimensional appearance of the one or more fiducials; based on the two-dimensional appearance, determine a roll angle of the distal portion about a longitudinal axis; based on the roll angle, generate one or more motor commands configured to cause a robotic system coupled to the medical instrument to move the medical instrument; and cause the robotic system to move the medical instrument based on the one or more motor commands.

Neural network-based detection of fiducial appearance for roll angle determination

Detect a two-dimensional appearance of the one or more fiducials using a neural network, and determine the roll angle based on the detected two-dimensional appearance.

Machine learning-based detection of fiducial appearance and roll angle determination

Apply a machine learning algorithm to the two-dimensional image to detect its two-dimensional appearance and determine a roll angle.

Distal tip detection and image-plane positioning

Detect the distal tip of the medical instrument within the two-dimensional medical image and determine a two-dimensional position of the distal tip within the image plane based on that detection.

Centerline and endpoint-based distal tip detection

Detect the distal tip of a medical instrument by determining a centerline of the distal portion from a two-dimensional medical image and determining an endpoint for that centerline.

Robotic catheter movement using fiducial-based heading and roll

Perform the system or method using a medical instrument that includes a catheter.

Across the independent claims, the invention centers on determining instrument heading and roll from a single-plane fluoroscopic two-dimensional image using radio-opaque fiducials, and then generating motor commands to move a robotic system coupled to the instrument. Dependent claim refinements specify neural network or machine learning-based detection of fiducial two-dimensional appearance, optionally detect the distal tip and its two-dimensional position in the image plane using centerline/endpoint geometry, and narrow the medical instrument to a catheter.

Stated Advantages

Automated tool tracking.

Risk mitigation via comparing expected vs actual distal motion based on fiducial-tracked distal roll/rotation.

Enables safer and precise catheter control, including robotic control.

Documented Applications

Automated tracking and robotic movement control of an endoluminal medical instrument/catheter based on vision-derived heading and roll from fluoroscopic images with radio-opaque fiducials.

JOIN OUR MAILING LIST

Stay Connected with MTEC

Keep up with active and upcoming solicitations, MTEC news and other valuable information.