Ultrasound image processing to render three-dimensional images from two-dimensional images

Patent metadata

Field	Value
Publication number	US-9697634-B2
Application number	US-201314095628-A
Country	US
Kind code	B2
Filing date	Dec 3, 2013
Priority date	Jun 30, 2007
Publication date	Jul 4, 2017
Grant date	Jul 4, 2017

Abstract

Methods for processing two-dimensional ultrasound images from an intracardiac ultrasound imaging catheter provide improved image quality and enable generating three-dimensional composite images of the heart. Two-dimensional ultrasound images are obtained and stored in conjunction with correlating information, such as time or an electrocardiogram. Images related to particular conditions or configurations of the heart can be processed in combination to reduce image noise and increase resolution. Images may be processed to recognize structure edges, and the location of structure edges used to generate cartoon rendered images of the structure. Structure locations may be averaged over several images to remove noise, distortions and blurring from movement.

First claim

I claim:

1. A method for processing a plurality of ultrasound image frames obtained from an intracardiac ultrasound imaging catheter having a phased array ultrasound transducer, comprising:
receiving the plurality of ultrasound image frames;
detecting edges of tissue structures within each of the plurality of ultrasound image frames;
determining a spatial location within a coordinate system of each of the detected edges in each of the plurality of image frames;
statistically analyzing the spatial locations of detected edges among the plurality of image frames to determine representative spatial locations for the tissue structures within the coordinate system;
generating a cartoon rendering of the tissue structures based upon the representative spatial locations; and
displaying the cartoon rendering of the tissue structures.

2. The method of claim 1 , wherein statistically analyzing spatial locations of detected edges among the plurality of ultrasound image frames comprises determining an average location of the detected edges among the plurality of ultrasound image frames.

3. The method of claim 1 , wherein statistically analyzing spatial locations of detected edges among the plurality of ultrasound image frames comprises determining a mean location of the detected edges among the plurality of ultrasound image frames.

4. The method of claim 1 , wherein statistically analyzing spatial locations of detected edges among the plurality of ultrasound image frames comprises determining a weighted average location of the detected edges among the plurality of ultrasound image frames.

5. The method of claim 1 , further comprising:
recognizing at least one of the tissue structures;
matching the recognized tissue structure to a model of heart tissue structures; and
back-calculating a location and orientation of the phased array ultrasound transducer within the coordinate system based upon the results of matching the recognized tissue structure to the model of heart tissue structures.

6. The method of claim 5 , further comprising registering the recognized tissue structure within an external coordinate reference frame.

7. The method of claim 1 , wherein statistically analyzing spatial locations of detected edges among the plurality of ultrasound image frames comprises calculating a moving average of spatial locations of detected edges among two or more ultrasound image frames selected from a stream of ultrasound image frames.

8. The method of claim 1 , wherein the steps of statistically analyzing the spatial locations of detected edges among the plurality of ultrasound image frames to determine representative spatial locations for the tissue structures and generating cartoon renderings of the tissue structures based upon the representative spatial locations are performed for each of a plurality of transducer rotational orientations to generate a plurality of cartoon rendered image frames corresponding to the plurality of transducer rotational orientations, and
further comprising combining the plurality of cartoon rendered image frames using their corresponding transducer orientations to generate a three-dimensional cartoon rendering of tissue structure;
wherein each of the plurality of ultrasound image frames is obtained at one or more of the plurality of transducer rotational orientations.

Assignees

• St Jude Medical Atrial Fibrillation Div Inc

Inventors

• Dala-Krishna Praveen

Classifications

• G06T7/12

  Edge-based segmentation · CPC title

• G06T2200/08

  involving all processing steps from image acquisition to 3D model generation · CPC title

• A61B8/4483

  characterised by features of the ultrasound transducer · CPC title

• G06T2207/10068

  Endoscopic image · CPC title

• G06T2207/30048

  Heart; Cardiac · CPC title