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-11138790-B2
Publication Date
2021-10-05
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 disclosed invention provides a method and system for generating a 3D surface mesh model of a patient specific anatomical feature from 2D medical images. A server receives the 2D medical images and automatically processes the images using a segmentation algorithm that assigns a label for each pixel, where the label corresponds to a tissue type selected from bone, soft tissue, blood vessel, or organs. This produces segmentation data that defines the patient specific anatomical feature.
To refine pixel-level tissue interpretation, the server accesses an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features labelled using a medical imaging ontology. An anatomical feature identification algorithm probabilistically matches the assigned label for each pixel against the anatomical knowledge dataset to generate a score for each pixel, classify each pixel based on the score, and thereby generate segmentation data defining the patient specific anatomical feature.
Using the segmentation data, the server generates a 3D surface mesh model defining the surface of the anatomical feature. The approach integrates segmentation and ontology-based anatomical feature identification to produce a patient specific 3D surface mesh model from 2D medical images.
Claims Coverage
The partial content includes two independent claims: one directed to a method and one directed to a computer-implemented system. Both independent claims contain the same core inventive workflow: receiving 2D medical images, assigning per-pixel tissue labels via a segmentation algorithm, probabilistically matching pixel labels against an ontology-based anatomical knowledge dataset to generate pixel scores and classifications, producing segmentation data, and generating a 3D surface mesh model from that segmentation data.
Server-based patient-specific anatomical segmentation to label pixels by tissue type
A server receives the 2D medical images and automatically processes them using a segmentation algorithm to assign a label for each pixel, where the label corresponds to a tissue type selected from bone, soft tissue, blood vessel, or organs.
Ontology-labelled anatomical knowledge dataset and probabilistic anatomical feature identification
The server accesses an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features labelled using a medical imaging ontology, and uses 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 classify each pixel based on the score.
Segmentation data generation and 3D surface mesh model generation
The server generates segmentation data defining the patient specific anatomical feature based on the classification of each pixel of the 2D medical images, and generates a 3D surface mesh model defining the surface of the anatomical feature using the segmentation data.
Across both independent claims, the inventive coverage centers on a server workflow that combines segmentation-based per-pixel tissue labeling with ontology-based, probabilistic anatomical feature identification to produce patient specific segmentation data, followed by 3D surface mesh model generation from that segmentation.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?