Method for generating a 3D physical model of a patient specific anatomic feature from 2D medical images
Inventors
HASLAM, Niall • TROJAN, Lorenzo • CRAWFORD, Daniel
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11288865-B2
Publication Date
2022-03-29
Expiration Date
Abstract
There is provided a method for generating a 3D physical model of a patient specific anatomic feature from 2D medical images. The 2D medical images are uploaded by an end-user via a Web Application and sent to a server. The server processes the 2D medical images and automatically generates a 3D printable model of a patient specific anatomic feature from the 2D medical images using a segmentation technique. The 3D printable model is 3D printed as a 3D physical model such that it represents a 1:1 scale of the patient specific anatomic feature. The method includes the step of automatically identifying the patient specific anatomic feature.
Core Innovation
The invention defines a patient specific anatomical feature from 2D medical images by receiving, by a server, 2D medical images of a patient and automatically processing the 2D medical images using a segmentation algorithm. The segmentation algorithm assigns a label for each pixel of the 2D medical images, and the server accesses an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features.
Using an anatomical feature identification algorithm, the assigned label for each pixel of the 2D medical images is probabilistically matched against the anatomical knowledge dataset. The algorithm generates a score for each pixel based on the probabilistic matching, and the server generates a confidence dataset comprising the label and the score for each pixel of the 2D medical images for defining the patient specific anatomical feature.
The same pipeline is implemented as a computer implemented system with at least one processor configured to receive the 2D medical images, process them using a segmentation algorithm, access the anatomical knowledge dataset, use the anatomical feature identification algorithm to probabilistically match pixel labels and generate per-pixel scores, and generate the confidence dataset to define the patient specific anatomical feature.
Claims Coverage
The document provides three independent claims: a method, a computer implemented system, and a non-transitory computer readable media. Each independent claim centers on segmentation-based per-pixel labeling and probabilistic pixel-wise anatomical feature identification against an anatomical knowledge dataset to generate a confidence dataset.
Server-based per-pixel confidence dataset generation for patient-specific anatomical feature
Receiving 2D medical images of a patient, automatically processing the images using a segmentation algorithm to assign a label for each pixel, accessing an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features, probabilistically matching the assigned label for each pixel against the anatomical knowledge dataset to generate a score for each pixel, and generating a confidence dataset comprising the label and the score for each pixel for defining the patient specific anatomical feature.
Processor-based patient-specific anatomical feature confidence dataset generation
A computer implemented system with at least one processor configured to receive 2D medical images of a patient, automatically process the images using a segmentation algorithm to assign a label for each pixel, access an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features, use an anatomical feature identification algorithm to probabilistically match the assigned label for each pixel against the anatomical knowledge dataset to generate a score for each pixel, and generate a confidence dataset comprising the label and the score for each pixel for defining the patient specific anatomical feature.
Non-transitory media instructions for confidence dataset generation for patient-specific anatomical feature
A non-transitory computer readable media having instructions that, when executed, cause a processor to receive 2D medical images of a patient, automatically process the images using a segmentation algorithm to assign a label for each pixel, access an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features, use an anatomical feature identification algorithm to probabilistically match the assigned label for each pixel against the anatomical knowledge dataset to generate a score for each pixel, and generate a confidence dataset comprising the label and the score for each pixel for defining the patient specific anatomical feature.
Across the independent claims, the claims consistently require segmentation-based per-pixel labeling, probabilistic pixel-wise matching to an anatomical knowledge dataset to produce per-pixel scores, and generation of a confidence dataset comprising the label and the score for each pixel.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?