Method for providing information about diagnosis of gallbladder polyp and device for providing information about diagnosis of gallbladder polyp using same

Patent metadata

Field	Value
Publication number	US-12511738-B2
Application number	US-202118017402-A
Country	US
Kind code	B2
Filing date	Jul 20, 2021
Priority date	Jul 22, 2020
Publication date	Dec 30, 2025
Grant date	Dec 30, 2025

Abstract

Provided are a method for providing information about the diagnosis of a gallbladder polyp and a device for providing information about the diagnosis of a gallbladder polyp using same. The method for providing information about the diagnosis of a gallbladder polyp being implemented by a processor includes the steps of receiving an ultrasound medical image including a gallbladder part of a subject, determining the pathogenesis of a gallbladder polyp in the subject using a first assessment model configured to determine the pathogenesis of a gallbladder polyp on the basis of the ultrasound medical image, and determining characteristics of the gallbladder polyp on the basis of a second assessment model configured to classify characteristics of the gallbladder polyp when the gallbladder polyp is determined in the ultrasound medical image.

First claim

What is claimed is:

1 . A method for providing information about the diagnosis of a gallbladder polyp implemented by a processor, the method comprising:
receiving an ultrasound medical image including a scale bar and a gallbladder part of a subject; determining a region of interest (ROI) with respect to the ultrasound medical image; cropping the ROI at a predetermined level based on the scale bar; determining pathogenesis of a gallbladder polyp (GB) in the subject using a first assessment model configured to determine the pathogenesis of the gallbladder polyp based on the ROI of the ultrasound medical image; and determining characteristics of the gallbladder polyp based on a second assessment model configured to classify characteristics of the gallbladder polyp, when the pathogenesis of the gallbladder polyp is determined in the ultrasound medical image.

2 . The method of claim 1 , wherein the determining of the pathogenesis of the gallbladder polyp of the subject comprises
determining gallbladder polyps or gallbladder stones using the first assessment model trained to classify the gallbladder polyps or the gallbladder stones based on the ultrasound medical image.

3 . The method of claim 2 , wherein the determining of the characteristics of the gallbladder polyp comprises
determining non-neoplastic GB polyps or neoplastic GB polyps using the second assessment model configured to classify the non-neoplastic GB polyps or the neoplastic GB polyps in a gallbladder polyp region.

4 . The method of claim 3 , further comprising:
after the determining the characteristics of the gallbladder polyp, determining adenocarcinoma GB polyps or adenomatous GB polyps using a third assessment model configured to classify the adenocarcinoma GB polyps or the adenomatous GB polyps in the neoplastic GB polyp region, when the neoplastic GB polyps is determined.

5 . The method of claim 1 , wherein the cropping at the predetermined level comprises:
calculating a size per pixel based on the scale bar; determining a target size based on the size per pixel; and cropping the ROI based on the target size.

6 . The method of claim 5 , wherein
the scale bar includes scales having a predetermined size, and the calculating of the size per pixel comprises:
determining an x-axial coordinate value for each of two scales selected from the scale bar of the ultrasound medical image;
calculating a pixel distance based on the x-axial coordinate; and
calculating the size per pixel based on the pixel distance and the predetermined size.

7 . The method of claim 5 , wherein
the scale bar includes scales having a predetermined size, and the calculating of the size per pixel comprises:
cropping the scale bar in the ultrasound medical image;
determining a position of a scale of the ultrasound medical image based on the scales of the cropped scale bar using a connected component algorithm; and
calculating the size per pixel based on the position of the scale and the predetermined size.

8 . The method of claim 1 , wherein the ultrasound medical image is an endoscopic ultrasound (EUS) medical image.

9 . The method of claim 1 , further comprising:
displaying and providing an ROI for the first assessment model on the ultrasound medical image while determining the pathogenesis of the gallbladder polyp, or displaying and providing an ROI for the second assessment model on the ultrasound medical image while classifying the characteristics of the gallbladder polyp.

10 . A device for providing information about the diagnosis of a gallbladder polyp, the device comprising:
a transceiver configured to receive an ultrasound medical image including a scale bar and a gallbladder part of a subject; and a processor coupled with the transceiver and configured to
determine a region of interest (ROI) with respect to the ultrasound medical image,
crop the ROI at a predetermined level based on the scale bar,
determine the pathogenesis of a gallbladder polyp (GB) in the subject using a first assessment model configured to determine the pathogenesis of the gallbladder polyp based on the ROI of the ultrasound medical image, and
determine characteristics of the gallbladder polyp based on a second assessment model configured to classify the characteristics of the gallbladder polyp, when the gallbladder polyp is determined in the ultrasound medical image.

11 . The device of claim 10 , wherein the processor is further configured to determine gallbladder polyps or gallbladder stones, using the first assessment model further trained to classify the gallbladder polyps or the gallbladder stones based on the ultrasound medical image.

12 . The device of claim 11 , wherein the processor is further configured to determine non-neoplastic GB polyps or neoplastic GB polyps, using the second assessment model configured to classify the non-neoplastic GB polyps or the neoplastic GB polyps in a gallbladder polyp region.

13 . The device of claim 12 , wherein the processor is further configured to determine adenocarcinoma GB polyps or adenomatous GB polyps, using a third assessment model configured to classify the adenocarcinoma GB polyps or the adenomatous GB polyps in the neoplastic GB polyp region, when the neoplastic GB polyps are determined.

14 . The device of claim 10 , wherein the processor is further configured to
calculate a size per pixel based on the scale bar, determine a target size based on the size per pixel, and crop the ROI based on the target size.

15 . The device of claim 14 , wherein
the scale bar includes scales having a predetermined size, and the processor is further configured to
determine an x-axial coordinate value for each of two scales selected from the scale bar of the ultrasound medical image,
calculate a pixel distance based on the x-axial coordinate, and
calculate the size per pixel based on the pixel distance and the predetermined size.

16 . The device of claim 14 , wherein
the scale bar includes scales having a predetermined size, and the processor is further configured to
crop the scale bar in the ultrasound medical image,
determine a position of a scale of the ultrasound medical image based on the scales of the cropped scale bar, using a connected component algorithm, and
calculate the size per pixel based on the position of the scale and the predetermined size.

17 . The device of claim 10 , wherein the ultrasound medical image is an endoscopic ultrasound (EUS) medical image.

18 . The device of claim 12 , wherein the is further configured to
display and provide an ROI for the first assessment model on the ultrasound medical image while determining the pathogenesis of the gallbladder polyp, or display and provide an ROI for the second assessment model on the ultrasound medical image while classifying the characteristics of the gallbladder polyp.

Assignees

• Univ Yonsei Iacf
• Univ Gachon Ind Acad Coop Found
• Gil Medical Ct

Inventors

• Cho Jae Hee
• Jang Sung Ill
• Kim Kwang Gi
• Kim Young Jae
• Kim Eui Joo

Classifications

• G06T2207/30092

  Stomach; Gastric · CPC title

• G06T2207/20132

  Image cropping · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/10132

  Ultrasound image · CPC title

• G06T2207/10068

  Endoscopic image · CPC title