Artifact detection via a medical scan quality assurance system

Patent metadata

Field	Value
Publication number	US-12198085-B2
Application number	US-202217656140-A
Country	US
Kind code	B2
Filing date	Mar 23, 2022
Priority date	Nov 21, 2018
Publication date	Jan 14, 2025
Grant date	Jan 14, 2025

Abstract

A medical scan quality assurance system is operable to utilize artificial intelligence to train at least one computer vision model based on a training set of medical scans. A set of medical scans are received. Quality assurance data is generated for the set of medical scans utilizing artificial intelligence by performing at least one quality assurance function on the set of medical scans by utilizing the at least one computer vision model. A first medical scan is identified in the set of medical scans to include an artifact, detected by performing the at least one quality assurance function, that is determined to obscure at least a threshold percentage of a key anatomical part based on the quality assurance data. An artifact obstruction notification indicating the first medical scan is generated for transmission to a client device for display.

First claim

What is claimed is:

1. A medical scan quality assurance system, comprising:
at least one processor; and
a memory that stores operational instructions that, when executed by the at least one processor, cause the medical scan quality assurance system to:
utilize artificial intelligence to train at least one computer vision model based on a training set of medical scans;
receive a set of medical scans;
utilize artificial intelligence to generate quality assurance data for the set of medical scans by performing at least one quality assurance function on the set of medical scans by utilizing the at least one computer vision model;
identify a first medical scan in the set of medical scans that includes an artifact, detected by performing the at least one quality assurance function, that is determined to obscure at least a threshold percentage of a key anatomical part based on the quality assurance data; and
generate an artifact obstruction notification indicating the first medical scan for transmission to a client device for display.

2. The medical scan quality assurance system of claim 1 , wherein the set of medical scans each include a plurality of image slices, and wherein the operational instructions, when executed by the at least one processor, further cause the medical scan quality assurance system to:
identify, based on the quality assurance data, a second medical scan that is at least one of: missing at least one image slice, or has one or more image slices out of order; and
generate at least one image slice error notification indicating the second medical scan for transmission.

3. The medical scan quality assurance system of claim 2 , wherein the quality assurance data indicates the second medical scan is missing the at least one image slice based on the second medical scan including a number of slices in its plurality of image slices that is less than a known number of slices intended to be in a corresponding medical scan.

4. The medical scan quality assurance system of claim 2 , wherein the second medical scan includes image data of a hard copy of a corresponding medical scan based on being later digitized, and wherein the quality assurance data indicates the second medical scan is at least one of: missing the at least one image slice, or has the one or more image slices out of order, based on at least one slice number detected in the image data of the hard copy of the corresponding medical scan utilizing artificial intelligence.

5. The medical scan quality assurance system of claim 4 , wherein the operational instructions, when executed by the at least one processor, further cause the medical scan quality assurance system to:
automatically reorder a plurality of images slices of the second medical scan based on the at least one slice number detected in the image data of the hard copy of the corresponding medical scan.

6. The medical scan quality assurance system of claim 2 , wherein performing the at least one quality assurance function upon the second medical scan includes comparing image data of neighboring ones of a set of slices of the second medical scan utilizing artificial intelligence, and wherein the second medical scan determined to be at least one of: missing the at least one image slice, or has the one or more image slices out of order based on the quality assurance data indicating at least one of: that anatomical information is missing in at least two neighboring ones of the set of slices of the second medical scan, or that anatomical features do not align in at least two neighboring ones of the set of slices of the second medical scan.

7. The medical scan quality assurance system of claim 2 , wherein the at least one quality assurance function includes an anatomical reconstruction function, and wherein the second medical scan determined to be at least one of: missing the at least one image slice, or has the one or more image slices out of order based on the quality assurance data indicating the anatomical reconstruction function cannot be properly applied to a set of image slices of the second medical scan to completely reconstruct anatomy.

8. The medical scan quality assurance system of claim 1 ,
wherein the at least one computer vision model includes a first computer vision model trained via artificial intelligence to detect artifacts in image data of the training set of medical scans; and
wherein generating the quality assurance data for the set of medical scans includes performing an artifact detection function upon the set of medical scans by utilizing the first computer vision model, wherein the artifact is detected in the first medical scan based on performing the artifact detection function.

9. The medical scan quality assurance system of claim 8 , wherein the at least one computer vision model is trained using medical scans with label data indicating locations of artifacts in the image data of the training set of medical scans.

10. The medical scan quality assurance system of claim 8 ,
wherein the at least one computer vision model further includes a second computer vision model via artificial intelligence to detect the key anatomical part in image data of medical scans; and
wherein generating the quality assurance data for the set of medical scans includes performing an anatomical part detection function upon the set of medical scans by utilizing the first computer vision model, and wherein the key anatomical part is detected in the first medical scan based on performing the anatomical part detection function.

11. The medical scan quality assurance system of claim 10 , first wherein the second computer vision model is trained using medical scans with label data indicating locations of a plurality of distinct anatomical parts in the image data of the training set of medial scans, wherein the plurality of distinct anatomical parts includes the key anatomical part.

12. The medical scan quality assurance system of claim 10 , wherein a first span of a first location of the artifact is determined based on performing the artifact detection function upon the first medical scan, wherein a second span of a second location of the key anatomical part is determined based on performing the anatomical part detection function upon the first medical scan, and wherein generating the quality assurance data includes determining whether the artifact obscures the at least the threshold percentage of the key anatomical part based on the first span and the second span.

13. The medical scan quality assurance system of claim 10 , wherein the quality assurance data indicates the first medical scan includes the artifact obscuring the at least the threshold percentage of the key anatomical part based on the key anatomical part not being detected in the first medical scan when performing the anatomical part detection function upon the first medical scan; and further based on the key anatomical part being expected to be included in the first medical scan based on at least one of: metadata of the first medical scan indicating an anatomical region captured by the first medical scan, a scan type of the first medical scan, or at least one other anatomical part detected in the first medical scan that indicates the anatomical region captured in the first medical scan.

14. The medical scan quality assurance system of claim 1 , wherein the key anatomical part is detected in image data of the first medical scan based on at least one of a known location of the key anatomical part, known density features of the key anatomical part, a known size of the key anatomical part, and/or a known shape of the key anatomical part.

15. T

Assignees

• Enlitic Inc

Inventors

• Poblenz Eric C
• Yao Li
• Lui Keith
• Lyman Kevin

Classifications

• G06T12/20

  Inverse problem, i.e. transformations from projection space into object space · CPC title

• G06T11/26

  Drawing of charts or graphs · CPC title

• G06T11/10

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/09

  Supervised learning · CPC title