Back to results
US 2020/0138382 A1Published application
Methods and systems for whole body imaging
AssigneeGeneral Electric Company
InventorsNan Cao, Yongchuan Lai, Shiyu Li, Ting Zhang
Life of this patent
Estimating the term…
Abstract
Various methods and systems are provided for scanning an image subject along a scan axis. The method includes stitching sectional datasets acquired from different anatomical sections of the image subject based on locations of a landmark in the sectional datasets.
Claims20
A method for medical imaging, comprising:
- performing a first sectional scan on a first anatomical section of an image subject to obtain a first sectional dataset;
- performing a second sectional scan on a second anatomical section of the image subject to obtain a second sectional dataset;
- determining, with a trained deep learning network, a first location of a landmark in the first sectional dataset and a second location of the landmark in the second sectional dataset;
- and
- stitching the first sectional dataset and the second sectional dataset based on the first location of the landmark and the second location of the landmark.
2 more elements
An imaging apparatus, comprising:
- a scanner configured to scan an image subject;
- a table for placing the image subject, the table movable relative to the scanner;
- a memory storing a trained deep learning network;
- a controller coupled to the scanner, the table, and the memory and configured to:
- instruct the scanner to perform a first sectional scan on a first anatomical section of the image subject to acquire a first sectional dataset;
- move the table relative to the scanner;
- instruct the scanner to perform a second sectional scan on a second anatomical section of the image subject to acquire a second sectional dataset;
- determine a first location of a landmark in the first sectional dataset and a second location of the landmark in the second sectional dataset with the trained deep learning network;
- and
- stitching the first sectional dataset and the second sectional dataset based on the first location of the landmark and the second location of the landmark.
7 more elements
A non-transitory computer-readable medium comprising instructions that, when executed, cause a processor to:
- perform a first sectional scan on a first anatomical section of an image subject to obtain a first sectional dataset;
- perform a second sectional scan on a second anatomical section of the image subject to obtain a second sectional dataset;
- determine, with a trained deep learning network, a first location of a landmark in the first sectional dataset and a second location of the landmark in the second sectional dataset;
- and
- stitch the first sectional dataset and the second sectional dataset based on the first location of the landmark and the second location of the landmark.
2 more elements
Dependencies are inferred from the claim text.