Methods and systems for x-ray imaging and labeling
Inventors
Bailey, SR., Michael Q. • Bender, Scott • Doyle, Patrick Raymond • Durgempudi, Pavan • Litchfield, Benjamin • Steines, David • Schestopol, Benjamin
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12186120-B2
Publication Date
2025-01-07
Expiration Date
Abstract
An example method includes capturing, via an x-ray machine, a plurality of x-ray images of a patient covering a number of different anatomy of the patient in any order, using a machine learning algorithm to process the plurality of x-ray images for identification of an anatomy in respective x-ray images of the plurality of x-ray images, associating a label with each of the plurality of x-ray images based on the identification of the anatomy, positioning each of the plurality of x-ray images upright based on a preset coordinate scheme for the anatomy, arranging the plurality of x-ray images into a predetermined order based on the species of the patient, and generating and outputting a data file including the plurality of x-ray images in the predetermined order, positioned upright, and labeled.
Core Innovation
The disclosed invention relates to x-ray imaging and labeling in which a plurality of x-ray images of a patient covering a number of different anatomy are captured. The images are captured in a predetermined shot order structure, and a preset label is applied to each x-ray image according to the predetermined shot order structure independent of content of the plurality of x-ray images. The method then ceases use of the predetermined shot order structure and enables use of a free-form shot order structure.
After enabling the free-form shot order structure, a machine learning algorithm executed by a computing device processes the plurality of x-ray images to identify an anatomy in respective x-ray images. Based on the identified anatomy, the computing device associates a label with each of the plurality of x-ray images, where the label is selected from among a preset labeling scheme for anatomy based on a species of the patient.
The document also describes correcting labeling when shot order assumptions fail, including mislabeling caused by out-of-order or retaken/duplicate images, and reorienting images upright using a preset coordinate scheme, reordering images into a predetermined order based on patient species, and generating/outputting labeled data files such as DICOM for downstream processing. The document further describes optional real-time feedback related to procedure completeness checks and exposure/quality-related safety feedback, and replacing predetermined shot-order software with updated radiology image capture software that enables free-form shot order capture.
Claims Coverage
The partial content provides two independent claims (a method claim and a system claim). Across these independent claims, the core inventive features include stopping a predetermined shot order structure and enabling a free-form shot order structure, using a machine learning algorithm to identify anatomy content, and associating preset anatomy labels selected according to patient species; and updating veterinary radiology image capture software to enable free-form shot order structure.
Shot-order structure transition with preset order-independent labeling
Capturing a plurality of x-ray images in a predetermined shot order structure and applying a preset label according to the predetermined shot order structure independent of content; ceasing use of the predetermined shot order structure and enabling use of a free-form shot order structure.
Machine learning anatomy identification driving preset species-based labeling
Using a machine learning algorithm executed by a computing device to process the plurality of x-ray images for identification of an anatomy in respective x-ray images; associating a label with each x-ray image based on the identification of the anatomy, where the label is selected from among a preset labeling scheme for anatomy based on a species of the patient.
Updating veterinary radiology capture software to enable free-form shot order structure
Updating radiology image capture software to enable free-form shot order structure for application of the preset labels to a plurality of x-ray images based on content of the plurality of x-ray images and independent of the free-form shot order structure.
The independent claim set centers on switching from predetermined shot-order-based labeling to free-form shot-order capture, using machine learning to identify anatomy in the captured images, and selecting preset anatomy labels based on patient species; the system claim further requires updating veterinary radiology capture software to support free-form shot order structure while still applying preset labels.
Stated Advantages
Improved labeling accuracy when shot order assumptions fail, including mislabeling due to out-of-order or retaken/duplicate images.
Enables free-form shot order structure instead of requiring a predetermined shot order structure for capturing anatomy-covering x-ray images.
Generates labeled outputs/data files for downstream processing, including labeled data files in DICOM format.
Documented Applications
Veterinary radiology for capturing and labeling multiple x-ray images of different anatomy across a patient, including processing to identify anatomy and assign species-dependent preset anatomy labels.
Downstream processing using labeled data files (e.g., DICOM) generated from the labeled and processed x-ray images.
Use in a workflow that supports real-time feedback including procedure completeness checks and exposure/quality-related safety feedback.
Interested in licensing this patent?