Method and device for operating a medical imaging device for the positionally correct representation of non-anatomical structures during an imaging examination
Inventors
Hörndler, Klaus • Kachelriess, Marc • KNAUP, Michael • König, Thomas • Vöth, Tim
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12357189-B2
Publication Date
2025-07-15
Expiration Date
Abstract
Methods for operating a medical imaging device for positionally correct representation of non-anatomical structures during an imaging examination may include providing a first 3D image containing at least one anatomical structure, extracting at least one anatomical model from the at least one anatomical structure, providing 2D update images recorded at different times, extracting non-anatomical and anatomical structures from subsets of the update images, calculating a non-anatomical 3D image from at least two partial reconstructions based on the extracted non-anatomical structures, reconstructing an anatomical 3D image based on the extracted anatomical structures, registering the anatomical 3D image with the first 3D image by determining a coordinate transformation, and creating a navigation volume from the anatomical model and the non-anatomical 3D image using the coordinate transformation.
Core Innovation
The invention relates to operating a medical imaging device for the positionally correct representation of non-anatomical structures during an imaging examination. The method provides a first 3D image including at least one anatomical structure, and an anatomical model is extracted from the anatomical structure of the first 3D image. At least two 2D update images are provided, captured simultaneously or at different times, and subsets of these 2D update images are used for separate extractions of non-anatomical structures and anatomical structures.
Non-anatomical structures are extracted from a first subset of the 2D update images, and anatomical structures are extracted from a second subset of the 2D update images. A non-anatomical 3D image is calculated from at least two 3D limited projection geometry reconstructions, and calculating the non-anatomical 3D image includes removing artifacts from the at least two 3D limited projection geometry reconstructions using a machine learning method. In parallel, an anatomical 3D image is reconstructed from the extraction of the anatomical structures.
The anatomical 3D image is registered with the first 3D image by determining a coordinate transformation. Based on the determined coordinate transformation, a navigation volume is created using the anatomical model and the non-anatomical 3D image. A gantry-based medical imaging system includes provisioning of the first 3D image, at least two image chains configured to record 2D update images and 3D images, a computing unit performing the anatomical model extraction, structure extraction, non-anatomical 3D calculation with machine learning artifact removal, anatomical 3D calculation, navigation volume creation, and a display configured to display the navigation volume.
Claims Coverage
The document contains two independent claims. The first claim covers a method that reconstructs and registers anatomical and non-anatomical 3D representations for a positionally correct navigation volume using coordinate transformation and machine learning artifact removal. The second claim covers a gantry-based system implementing the same pipeline with image chains, computing unit, and display.
Positionally correct representation using first 3D, anatomical model, and 2D updates
Providing a first 3D image containing at least one anatomical structure; extracting at least one anatomical model from the at least one anatomical structure of the first 3D image; providing at least two 2D update images captured simultaneously or at different times; extracting non-anatomical structures from a first subset of the 2D update images; and extracting anatomical structures from a second subset of the 2D update images.
Non-anatomical 3D from limited projection reconstructions with machine learning artifact removal
Calculating a non-anatomical 3D image from at least two 3D limited projection geometry reconstructions, wherein calculating the non-anatomical 3D image comprises removing artifacts from the at least two 3D limited projection geometry reconstructions using a machine learning method.
Registered anatomical 3D and navigation volume via coordinate transformation
Reconstructing an anatomical 3D image from the extraction of the anatomical structures; registering the anatomical 3D image with the first 3D image by determining a coordinate transformation; and creating a navigation volume based on the at least one anatomical model and the non-anatomical 3D image by using the determined coordinate transformation.
Gantry-based medical imaging system with image chains, computing unit, and display
Providing a first 3D image including at least one anatomical structure; at least two image chains configured to record at least one of 2D update images and 3D images; a computing unit configured to extract an anatomical model, extract anatomical and non-anatomical structures from the 2D update images, calculate 3D limited projection geometry reconstructions from the extraction of non-anatomical structures, calculate a non-anatomical 3D image from at least two of the 3D limited projection geometry reconstructions including removing artifacts using a machine learning method, calculate an anatomical 3D image from extraction of anatomical structures, and create a navigation volume from the anatomical model and the non-anatomical 3D image by determining a coordinate transformation; and a display configured to display the navigation volume.
Across the independent claims, the core coverage is the combined pipeline for extracting anatomical and non-anatomical structures from 2D update images, computing a non-anatomical 3D image from 3D limited projection geometry reconstructions with machine learning artifact removal, reconstructing and registering an anatomical 3D image to the first 3D image by determining a coordinate transformation, and creating a navigation volume that enables positionally correct representation. The second independent claim additionally specifies the gantry-based system architecture with image chains, a computing unit, and a display configured to display the navigation volume.
Stated Advantages
Positionally correct representation of non-anatomical structures during an imaging examination.
Documented Applications
Positionally correct representation of non-anatomical structures during an imaging examination using a medical imaging device and navigation volume display.
Interested in licensing this patent?