IP Library Granted Patent US 12694558
Granted Patent B2
US 12694558 · App. 17/887,637 · Granted Jul 28, 2026

System and method for identifying feature in an image of a subject

Inventors: Patrick A. Helm (Canton, MA); Jeffrey H. Siewerdsen (Baltimore, MD); Ali Uneri (Columbia, MD); Craig K. Jones (Reisterstown, MD); Yixuan Huang (Baltimore, MD); Xiaoxuan Zhang (Baltimore, MD)
Assignees: Medtronic Navigation, Inc.; The Johns Hopkins University
G06T7/73A61B90/37G06T5/50G06T7/0012G06T7/11G06T7/337G06V10/16G06V10/22G06V10/25G06V10/40G06V10/44G06V10/764G06V10/82A61B2090/376G06T2207/10116G06T2207/20081G06T2207/20084G06T2207/20212G06T2207/20221G06V2201/03G06V2201/07
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12694558
App. No.
17/887,637
Granted
Jul 28, 2026
Kind
B2
Abstract

A method and system is disclosed for analyzing image data of a subject. The image data can be collected with an imaging system in a selected manner and/or motion. The image data may include selected overlap and be acquired with an imaging system that generates a plurality of perspectives for more than one location. An automatic system and method may then define or identify various features and/or allow for registration for alternative image data.

Claims (63)

1 . A method of detecting and classifying a feature in an image based on a plurality of individual image projections generated with an imaging system, the method comprising:

acquiring the plurality of individual image projections, the plurality of individual image projections comprising:

a first set of individual image projections from a first view of a subject, and

a second set of individual image projections from a second view of the subject,

wherein the first view is different than the second view;

determining proposed regions for the feature in each of a first feature map and a second feature map, wherein the first feature map is associated with the first view and the second feature map is associated with the second view;

combining the proposed regions of the feature in each of the first feature map and the second feature map;

classifying the feature and determining a position of the feature in each of the first view and the second view based on combining the proposed regions; and

outputting the determination of the position of the feature in each of a first output view and a second output view, wherein:

the first output view includes a first label identifying the feature and the position of the feature in the first output view, and

the second output view includes a second label identifying the feature and the position of the feature in the second output view.

2 . The method of claim 1 , wherein the determination of the proposed regions is based on a machine learning algorithm operable to recognize the feature in the plurality of individual image projections.

3 . The method of claim 1 , wherein the first view is a lateral view and the second view is an anterior-to-posterior view.

4 . The method of claim 1 , wherein the first view is an anterior-to-posterior view and the second view is different than the anterior-to-posterior view.

5 . The method of claim 1 , wherein the feature is extracted in each of the first view and the second view.

6 . The method of claim 5 , further comprising:

determining region proposals for the feature in each of the first view and the second view; and

aligning the region proposals between the first view and the second view.

7 . The method of claim 6 , wherein combining the proposed regions of the feature in each of the first feature map and the second feature map comprises:

concatenating the aligned region proposals of the first view and the second view.

8 . The method of claim 7 , further comprising:

evaluating the concatenated aligned region proposals via a region proposal convolutional neural network, wherein determining the position of the feature is based on evaluating the concatenated aligned region proposals.

9 . The method of claim 8 , further comprising:

confirming the classification of the feature based at least in part on imposing an order on classification of a plurality of features.

10 . The method of claim 1 ,

wherein the plurality of individual image projections are acquired with a slot filter;

wherein acquiring the plurality of individual image projections includes acquiring a set of individual slot image projections of the subject with an imaging head at a selected position relative to the subject;

wherein the slot filter includes a plurality of slots configured to cause the generation of the set of individual slot image projections simultaneously; and

wherein each individual slot image projection of the set of individual slot image projections has a unique perspective of the subject relative to each other individual slot image projection; wherein the first view and the second view are both generated with a selected set of the plurality of individual image projections.

11 . The method of claim 1 , wherein the plurality of individual image projections are generated via an x-ray source configured to emit a beam of x-rays.

12 . A system to detect and classify a feature in an image based on a plurality of individual image projections generated with an imaging system, the system comprising:

a processor module configured to execute instructions to:

acquire the plurality of individual image projections, the plurality of individual image projections comprising:

a first set of individual image projections from a first view of a subject, and

a second set of individual image projections from a second view of the subject, wherein the first view is different than the second view;

determine proposed regions for the feature in each of a first feature map and a second feature map, wherein the first feature map is associated with the first view and the second feature map is associated with the second view;

combine the proposed regions of the feature in each of the first feature map and the second feature map;

classify the feature and determining a position of the feature in each of the first view and the second view based on combining the proposed regions; and

output the determination of the position of the feature in each of a first output view and a second output view, wherein:

the first output view includes a first label identifying the feature and the position of the feature in the first output view, and

the second output view includes a second label identifying the feature and the position of the feature in the second output view.

13 . The system of claim 12 , wherein the determination of the proposed regions is based on a machine learning algorithm operable to recognize the feature in the plurality of individual image projections.

14 . The system of claim 12 , wherein the acquired first view is a lateral view and the acquired second view is an anterior-to-posterior view.

15 . The system of claim 12 , wherein the feature is extracted in each of the acquired first view and the acquired second view.

16 . The system of claim 15 , wherein the processor module is further configured to execute instructions to:

determine region proposals for the feature in each of the first view and the second view; and

align the region proposals between the first view and the second view.

17 . The system of claim 16 , wherein, to combine the proposed regions of the feature in each of the first feature map and the second feature map, the processor module is configured to execute instructions to:

concatenate the aligned region proposals of the first view and the second view.

18 . The system of claim 17 , wherein the processor module is further configured to execute instructions to:

evaluate the concatenated aligned region proposals via a region proposal convolutional neural network, wherein determination of the position of the feature is based on evaluation of the concatenated aligned region proposals.

19 . The system of claim 18 , wherein the processor module is further configured to execute instructions to:

confirm the classification of the feature based at least in part on imposing an order on classification of a plurality of features.

20 . The system of claim 12 ,

wherein, to acquire the plurality of individual image projections, the processor module is configured to execute instructions to acquire the plurality of individual image projections with a slot filter;

wherein, to acquire the plurality of individual image projections, the processor module is configured to execute instructions to acquire a set of individual slot image projections of the subject with an imaging head at a selected position relative to the subject;

wherein the slot filter includes a plurality of slots configured to cause the generation of the set of individual slot image projections simultaneously; and

wherein each individual slot image projection of the set of individual slot image projections has a unique perspective of the subject relative to each other individual slot image projection.

21 . The system of claim 12 , further comprising:

an x-ray source configured to emit a beam of x-rays;

a slot filter configured to split the beam of x-rays into at least two slot beams of x-rays; and

a detector configured to detect the slot beams of x-rays.

22 . The system of claim 12 , wherein the first view is an anterior-to-posterior view and the second view is different than the anterior-to-posterior view.