patent.fyi
Applications  /  US 2026/0354901
United States Patent Application Publication

US 2026/0354901 A1

Published
Nov 12, 2026
Filed
Mar 16, 2026
Earliest date on the face
Feb 26, 2018
Classification
A61B 34/10
First published here
Nov 19, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

1.2Registered two dimensional views

What ifWhat if the system uses two annotated anatomical slices without constructing a volumetric or surface model?
Claim language“a three dimensional model”

The processor extracts two intersecting slices from the anatomical scan and traces the target boundary on each slice. It projects the tracked catheter tip onto each plane and reports the perpendicular distance from the tip to that plane next to the marker.

1.3Electromagnetic catheter tip tracking

What ifWhat if electromagnetic sensors supply catheter position during the procedure without an image stream?
Claim language“receiving at least one intra-procedural image”

A field generator and a miniature coil near the catheter tip supply the tip position in tracker coordinates. Tracked contact points collected along the valve annulus relate those coordinates to the anatomical model, and subsequent coil readings update the displayed tip position.

1.4Registration with external fiducials

What ifWhat if the system derives image alignment from external fiducials on a patient-mounted frame?
Claim language“based on the anatomical reference structure”

A rigid frame carries six noncoplanar radiopaque beads that appear in the planning scan and the procedural radiograph. The processor fits the projection geometry by matching the scanned bead centers to their radiographic projections. Both acquisitions use the same respiratory phase to limit motion between the frame and the target.

1.5Direct localization in ultrasound

What ifWhat if the system locates the device relative to the target directly in a live ultrasound volume?
Claim language“using the registration transformation”

A volumetric ultrasound probe images the target boundary and an echogenic marker at the catheter tip in the same acquisition. The processor segments both in the volume's native coordinates and computes the tip-to-boundary distance from the voxel spacing.

1.6Spoken catheter position updates

What ifWhat if a speaker reports the catheter position relative to the target?
Claim language“generating an output visualization”

The processor resolves the tip offset from a selected target point along the patient's left–right, anterior–posterior, and superior–inferior axes. When the operator presses a footswitch, a speech synthesizer reads the three directional offsets in millimeters.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Borescope guidance inside a turbine casing

An industrial computed tomography scan supplies a three-dimensional model of a turbine blade assembly and the access port in its casing. During inspection, a calibrated stereo camera at the casing entrance images a ring of bolt heads around the port and a fiducial on the articulated borescope tip as it enters. The processor aligns the reconstructed bolt heads with their scan locations and applies the resulting transform to the tip position for display against the blade assembly.

Component substitution

Ultrasound volume registration for liver guidance

A tracked three-dimensional ultrasound probe replaces the intra-procedural projection imager. A baseline volume supplies a liver target model and the locations of portal vein bifurcations, while live ultrasound volumes show the same bifurcations and the echogenic needle tip. The processor fits a rigid transform to corresponding bifurcations and maps the needle tip into the baseline model for display.

Parameter or scale variation

Paired projections for catheter tip localization

Two calibrated projection imagers acquire paired frames at 7.5 pairs per second with their central rays separated by 60 degrees. The processor aligns projected landmarks from the three-dimensional model with the corresponding landmarks in both frames. It back-projects each observed catheter tip into a ray and displays the midpoint of the shortest segment between the rays in model coordinates.

Combination with existing systems

Registered overlay on a surgical microscope

A stereoscopic surgical microscope displays a registered three-dimensional model of a cortical target and the position of a probe tip. Its live camera pair images cortical vessel junctions and the tip, while a baseline image volume supplies the target geometry and matching junctions. The processor aligns the stereo reconstruction to the baseline junctions and uses the microscope calibration at the current zoom to project the model and tip into both eyepiece views.

Image corrected electromagnetic catheter tracking

An electromagnetic tracking system samples a coil near the catheter tip between image acquisitions. A rigid marker fixture carries both an electromagnetic reference sensor and radiopaque fiducials with known positions relative to that sensor. Each intra-procedural image shows the fixture fiducials, an anatomical reference structure and the catheter, allowing the processor to align the tracking frame with the imager and the imager with the anatomical model. The processor composes those transforms to display coil positions between image updates.

Automated or data-driven variant

Confidence weighted vessel landmark registration

A convolutional network trained on annotated procedure images returns pixel locations and confidence scores for vessel bifurcations and a catheter tip. The processor projects corresponding bifurcations from the three-dimensional model and updates the registration by minimizing confidence-weighted reprojection error with a robust loss. It maps tip detections from calibrated paired views into the model frame and flags the display when the landmark residual exceeds a configured threshold.

Specific version of a part

Live ultrasound volume acquisition

A matrix-array transesophageal probe acquires a volumetric sweep that includes the target chamber and a valve annulus. The guidance processor beamforms the probe's channel signals and scan-converts the samples into a volume for segmentation.

Gated contrast enhanced scan

The processor receives a contrast-enhanced computed tomography volume reconstructed at a selected phase of the cardiac cycle with 0.6 mm slice spacing. The volume includes the left atrial chamber as the anatomical target and the mitral annulus as the reference structure.

Triangulated left atrial surface

The processor segments the contrast-filled left atrial cavity and extracts a triangular mesh from the resulting voxel mask. It applies surface smoothing while holding the vertices at the pulmonary vein openings fixed.

Biplane landmark registration

Two calibrated fluoroscopic views capture corresponding landmarks along a calcified valve annulus. The processor fits a rigid transformation by minimizing the summed squared distances between the observed landmarks and the projections of their corresponding model points in both views.

Stereoscopic catheter and chamber view

A head-mounted display renders the chamber wall as a translucent surface and the catheter shaft as an opaque tube. The renderer places a marker at the current catheter tip and updates both eye views from the same device position.

Section 3 · quoted from USPTO records

Other work

Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.

Document AUS 9,295,435 B2
Published
Mar 29, 2016
Filed
Jan 31, 2012
Class
A61B 6/12
Document BUS 9,761,005 B2
Published
Sep 12, 2017
Filed
May 8, 2014
Class
G06T 7/0089
Claim 1
Document A
Document B
1.1receiving a first image data set comprising an anatomical reference structure and an anatomical target;
The anatomy representation as acquired by the proposed medical imaging device may be or may be derived from a two-dimensional or three-dimensional data set acquired pre-interventional, i.e. prior to the vessel intervention procedure. For example, the anatomy representation may be derived from a computer tomography (CT) scan.
A · US 9,295,435 B2 · ¶ 19
In operation, the preoperative workstation extracts a preoperative anatomical mesh from a preoperative anatomical image (e.g., a CT liver image or a MR liver image) based on a base topology of the anatomical mesh template.
B · US 9,761,005 B2 · ¶ 6
1.2generating a three dimensional model of the anatomical target;
In one embodiment of stage S11 as shown in FIG. 2, a CT/MR segmentation model 11 semi-automatically fits a liver mesh template 40 of a known base topology to image features (e.g., surface boundaries) of a CT/MR liver image 50 for an extraction of a CT/MR liver mesh 51 from CT/MR liver image 50.
B · US 9,761,005 B2 · ¶ 23
1.3receiving at least one intra-procedural image comprising the anatomical reference structure and a medical device;
(b) processing (DP) the at least one angiogram X-ray image and the at least one live fluoroscopy X-ray image in order to identify in each of the X-ray images the radio-opaque device;
A · US 9,295,435 B2 · abstract
The intraoperative mesh extractor extracts an intraoperative anatomical mesh from an intraoperative anatomical image (e.g., an ultrasound liver image) based on a preoperative topology of the preoperative anatomical mesh derived from the base topology of the anatomical mesh template.
B · US 9,761,005 B2 · ¶ 7
1.4determining a registration transformation aligning the intra-procedural image with the model based on the anatomical reference structure;
(c) registering (R2) the at least one angiogram X-ray image to the at least one live fluoroscopy X-ray image based on the identified radio-opaque device in order to provide an angiogram-fluoroscopy-registration; and (d) combining the anatomy-angiogram-registration and the angiogram-fluoroscopy-registration in order to provide an anatomy-fluoroscopy-registration (GTC;
A · US 9,295,435 B2 · abstract
More particularly, the identical topologies of the preoperative anatomical mesh and the intraoperative anatomical mesh derived from the known base topology of the anatomical mesh template establishes a mapping correspondence between the preoperative anatomical mesh and the intraoperative anatomical mesh that facilitates a spatial transformation of the preoperative anatomical image to the intraoperative anatomical image or vice-versa.
B · US 9,761,005 B2 · ¶ 26
1.5registering the medical device with the anatomical target using the registration transformation; and
(c) registering (R2) the at least one angiogram X-ray image to the at least one live fluoroscopy X-ray image based on the identified radio-opaque device in order to provide an angiogram-fluoroscopy-registration; and (d) combining the anatomy-angiogram-registration and the angiogram-fluoroscopy-registration in order to provide an anatomy-fluoroscopy-registration (GTC;
A · US 9,295,435 B2 · abstract
1.6generating an output visualization indicating a position of the medical device relative to the anatomical target.
Finally, an image representation resulting from the anatomy-fluoroscopy-registration showing an overlay of live images with the anatomy representation may be output thereby helping a surgeon to accurately position for example a synthetic aortic valve within an aortic root of a heart.
A · US 9,295,435 B2 · abstract

An empty cell means only that no passage was reproduced there.

Statements in these documents
There may be a need for a medical imaging device implementing a robust method for a registration process of an anatomic representation to live data.
A · US 9,295,435 B2 · ¶ 6
According to an aspect of the present invention, a medical imaging device for providing an image representation supporting the accurate positioning of an intervention device in a vessel intervention procedure is proposed.
A · US 9,295,435 B2 · ¶ 8
While, many image registration techniques have been proposed for the fusion of two different modalities, a fusion of an intraoperative ultrasound with any preoperative modality (e.g., CT or MR) has proven to be challenging due to lack of a functional dependency between the intraoperative ultrasound and the preoperative modality.
B · US 9,761,005 B2 · ¶ 3
The present invention generally relates to image segmentation of a preoperative anatomical image (e.g., a computed tomography (“CT”) scan or a magnetic resonance (“MR”) imaging scan of an anatomy) and of an intraoperative anatomical image (e.g., ultrasound (“US”) image frames of an anatomy) to facilitate a reliable registration of the preoperative anatomical image and the intraoperative anatomical image.
B · US 9,761,005 B2 · ¶ 2
Cite aspatent.fyi, US 2026/0354901 A1, claim 1, first published 2026-11-19T05:29:29-05:00, https://patent.fyi/US20260354901/, SHA-256 300b3bfd95a47edecac60b5acb9f7d1b13a9e34d3b1c2b90ea1e8da2c011d7c5
First published
Permanent addresshttps://patent.fyi/US20260354901/
Content fingerprint (SHA-256)300b3bfd95a47edecac60b5acb9f7d1b13a9e34d3b1c2b90ea1e8da2c011d7c5
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi