Apparatus and method for fetal intelligent navigation echocardiography
Inventors
Yeo, Lami • Romero, Roberto • Abella, Gustavo Eduardo • Gayoso, Ricardo Enrique
Assignees
Wayne State University • National Institutes of Health NIH • US Department of Health and Human Services
Publication Number
US-12268551-B2
Publication Date
2025-04-08
Expiration Date
2033-09-08
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A method and apparatus of examining fetal hearts includes obtaining ultrasound data of a fetal heart to generate fetal heart images, displaying different marked sample images of a fetal heart and at respective cardiac phases, each marking a different feature of a fetal heart at a respective marked location, and sequentially presenting one of the different marked sample images to a user along with one of the fetal heart images that is a similar view. As each of the different marked sample images is sequentially presented, a user is prompted to mark the different feature in a respective fetal heart image based on its marked location to create a plurality of marked fetal heart images, and a geometric model of the heart is generated based on the marked fetal heart images, from which fetal echocardiography views can be extracted. The fetal echocardiography views of the fetal heart are displayed.
Core Innovation
The invention relates to an apparatus and method for fetal intelligent navigation echocardiography (FINE), which enables the visualization of standard fetal echocardiography views from ultrasound dataset volumes obtained via spatiotemporal image correlation (STIC). The method involves obtaining ultrasound data of a fetal heart to generate multiple fetal heart images at various cardiac phases, displaying different marked sample images for distinct features of the fetal heart, and prompting the user to sequentially mark corresponding features in the fetal heart images. This marking process results in the creation of a plurality of marked fetal heart images.
Upon completion of the marking process, a geometric model of the fetal heart is generated based on the marked images, from which fetal echocardiography views can be extracted. The views are displayed to the user, and include nine key diagnostic planes: three-vessels and trachea, four-chamber, five-chamber, left ventricular outflow tract, short axis view of great vessels/right ventricular outflow tract, abdomen/stomach, ductal arch, aortic arch, and superior and inferior vena cava. The system can also provide a Virtual Intelligent Sonographer Assistance (VIS-Assistance) function for additional navigation and exploration around each diagnostic plane, assisting users in visualizing anatomy that may be challenging to capture manually.
The core problem addressed by this invention is the difficulty and operator dependency in the prenatal diagnosis of congenital heart anomalies, due to the complexity and small size of the fetal heart and the challenges in extracting all necessary cardiac planes from volume datasets using current methods. The FINE method aims to simplify and standardize the examination process, automatically generating essential views after the user marks specific anatomical landmarks, thereby reducing the dependence on operator skill and experience.
Claims Coverage
There are three main independent inventive features claimed, corresponding to a method for examining fetal hearts, a machine readable medium including an instruction set for the method, and an imaging system implementing the method.
Method for examining fetal hearts using marked sample images and model reconstruction
The method involves: 1. Obtaining spatiotemporal image correlation ultrasound data of a fetal heart to generate multiple fetal heart images at different cardiac phases. 2. Displaying different marked sample images of a sample fetal heart, each marking a different feature at a respective location. 3. Sequentially presenting one marked sample image to the user along with a corresponding fetal heart image for marking. 4. As each marked sample image is presented, prompting the user to mark the corresponding feature in the fetal heart image, creating a plurality of marked images. 5. Reconstructing a unique geometric model of the fetal heart from the marked images, from which fetal echocardiography views are extracted. 6. Displaying the fetal echocardiography views of the fetal heart.
Machine readable medium with instruction set for fetal heart examination and view generation
A machine readable medium includes an instruction set which, when executed by a machine, causes the following operations: - Obtaining spatiotemporal image correlation ultrasound data of a fetal heart to generate a plurality of images at different cardiac phases. - Showing different marked sample images, each marking a feature at a specific location. - Sequentially presenting one marked sample image with a corresponding fetal heart image for user marking. - Prompting the user to mark features on the fetal heart images to create a plurality of marked fetal heart images. - Reconstructing a unique geometric model of the fetal heart based on the marked images, from which echocardiography views can be extracted. - Displaying the fetal echocardiography views.
Imaging system configured for interactive marking, modeling, and display of fetal echocardiography views
An imaging system comprising: - A processor executing an instruction set causing the system to obtain spatiotemporal image correlation ultrasound data to generate fetal heart images at different cardiac phases. - The system displays marked sample images at respective cardiac phases, each marking a different feature of the sample fetal heart. - It sequentially presents a marked sample image with a fetal heart image, prompts the user to mark the feature in the respective fetal heart image, and creates marked images. - The system reconstructs a unique geometric model of the fetal heart from the marked images, extracts fetal echocardiography views, and displays these views.
The claims broadly encompass an interactive method, machine readable medium, and system for semi-automatic marking, geometric modeling, and display of fetal echocardiography views derived from STIC ultrasound data, where model generation and view extraction are based on user-marked anatomical features.
Stated Advantages
Simplifies the examination of the fetal heart and reduces operator dependency.
Allows for automatic and immediate generation and display of nine standard fetal echocardiography views from STIC datasets.
Provides consistent, standardized, and adaptive generation of diagnostic planes regardless of gestational age, fetal position, or anatomical variability.
Decreases the number of manual steps required to obtain cardiac views, making the process less operator dependent.
Enables simultaneous display of all nine diagnostic planes in a single template.
Allows operator-independent sonographic navigation and exploration of structures around each diagnostic plane via VIS-Assistance.
Supports automatic labeling of echocardiography views and anatomical structures that remains visible during zoom or resizing.
Can operate on conventional computers and is not dependent on specific ultrasound platforms or manual navigation software.
Facilitates telemedicine by allowing STIC datasets, diagnostic planes, and VIS-Assistance video clips to be transmitted for remote expert evaluation.
Substantially reduces the time required to generate standard fetal cardiac views compared to manual methods.
Documented Applications
Visualization and analysis of nine standard fetal echocardiography views from STIC ultrasound datasets for prenatal examination of the fetal heart.
Aid for examination of the fetal heart in the general pregnant population to help determine normality or abnormality.
Use in assisting detection and evaluation of certain congenital heart defects, including coarctation of the aorta, tetralogy of Fallot, transposition of great vessels, and pulmonary atresia with intact ventricular septum.
Facilitation of telemedicine and remote expert consultation through transmission of STIC datasets, diagnostic planes, and VIS-Assistance video clips.
Interested in licensing this patent?