Target area determination method and medical imaging system
AssigneeGE Precision Healthcare LLC
InventorsFuqiang Chen, Xihui Ju, Hankun Xue
Life of this patent
Estimating the term…
Abstract
Provided in the present application are a target area determination method and a medical imaging system, specifically a target area determination method and device, a medical imaging system, and a non-transitory computer-readable storage medium. The target area determination method includes: pre-processing an original image to acquire a pre-processed medical image, acquiring a target point in the medical image, and acquiring a plurality of sectional images of the medical image according to the target point, and acquiring, on the basis of a first model, a target area corresponding to the target point in the plurality of sectional images.
Claims12
A target area determination method for a medical image, comprising:
- pre-processing an original volumetric image to generate a standardized medical image, wherein the pre-processing includes at least one of resizing, pixel padding, normalization, or denoising;
- acquiring a target point in the medical image, the target point corresponding to a mapped location center of the original volumetric image;
- acquiring a plurality of sectional images of the medical image based on the target point, wherein each of the sectional images corresponds to a sectional plane passing through the target point, wherein the sectional plane includes at least one of a transverse plane, a coronal plane, and a sagittal plane;
- correcting the target point in each sectional image using a first deep learning model trained on offset data to acquire a corrected target point;
- acquiring a plurality of corrected sectional images according to the corrected target point;
- and
- acquiring, based on a second deep learning model trained on distance data representing spatial offsets from the corrected target point to anatomical boundaries, a target area corresponding to the target point in the plurality of corrected sectional images, wherein the target area is determined based on output distances from the target point to respective edges of a reference target area.
A medical imaging system, comprising a target area determination device, the target area determination device comprising:
- a pre-processing module configured to pre-process an original volumetric image to generate a pre-processed medical image, wherein the pre-processing includes at least one of resizing, pixel padding, normalization, or denoising;
- a target point acquisition module configured to acquire a target point corresponding to a mapped location center of the original volumetric image, and to acquire a plurality of sectional images of the medical image based on the target point, wherein each of the sectional images corresponds to a sectional plane passing through the target point;
- a target point correction module configured to correct the target point in each sectional image using a first deep learning model trained on offset data representing spatial deviations from anatomical boundaries, to acquire a corrected target point, and to acquire a plurality of corrected sectional images according to the corrected target point;
- and
- an area determination module configured to acquire, based on a second deep learning model trained on distance data representing spatial offsets from the corrected target point of anatomical boundaries, a target area corresponding to the target point in the plurality of sectional images, wherein the target area is determined based on output distances from the target point to respective edges of a reference target area.
Dependencies are inferred from the claim text.