US-20260127742-A1 - COLONOSCOPY AREA INDICATION SYSTEM AND METHOD

US20260127742A1US 20260127742 A1US20260127742 A1US 20260127742A1US-20260127742-A1

Abstract

Proposed is a colonoscopy area indication method that includes loading a colonoscopy image analysis model and setting an analysis condition of the analysis model by a model loading/condition setting part, and initializing an analysis screen and displaying a picture of a normal colon by a controller, and preprocessing, by an image preprocessing part, a colonoscopy image received through an image receiving part so that subsequent image analysis is smoothly performed, and analyzing the preprocessed colonoscopy image by an image analysis part by using the image analysis model based on AI, and detecting and indicating, on the basis of a result of analysis by the image analysis part, at least one selected from a group of an examination area, a diverticulum area, and a lesion area in the colonoscopy image, and providing, by the controller, the result of analysis performed by the image analysis part.

Inventors

Jisoo Keum
Kyung Nam Kim

Assignees

WAYCEN INC.

Dates

Publication Date: 20260507
Application Date: 20250702
Priority Date: 20241106

Claims (18)

1 . A colonoscopy area indication system, comprising: a model loading/condition setting part that loads a colonoscopy image analysis model, and set an analysis condition of the analysis model; an image receiving part that receives a colonoscopy image frame; an image preprocessing part that preprocesses a colonoscopy image received through the image receiving part, and makes a resultant preprocessed colonoscopy image ready for smooth subsequent image analysis; an image analysis part that analyzes the colonoscopy image preprocessed by the image preprocessing part by using the image analysis model based on artificial intelligence (AI), and detects and indicates, on the basis of a result of analysis, at least one selected from a group of an examination area, a diverticulum area, and a lesion area in the colonoscopy image; and a controller that checks states and controls operations of the model loading/condition setting part, the image receiving part, the image preprocessing part, and the image analysis part, and initializes an analysis screen and displays a picture of a normal colon when the model loading/condition setting part completes loading of the colonoscopy image analysis model and setting of the analysis condition of the analysis model, and provides the result of analysis performed by the image analysis part, wherein detecting and indicating at least one selected from the group of the examination area, the diverticulum area, and the lesion area in the colonoscopy image by the image analysis part comprises the controller transmitting, to the image analysis part, an indication condition change command based on a detection state of the diverticulum area, the command suppressing the indication of a diverticulum when the colonoscopy image corresponds to a normal colon and enabling diverticulum indication when a diverticulum is detected, and further indicating examination time and withdrawal time.
2 . The colonoscopy area indication system of claim 1 , wherein preprocessing of the colonoscopy image by the image preprocessing part includes analysis region cropping and input size adjustment.
3 . The colonoscopy area indication system of claim 1 , wherein the image analysis model of the image analysis part is configured as a single image analysis model for detecting the examination area and the diverticulum area.
4 . The colonoscopy area indication system of claim 1 , wherein the image analysis model of the image analysis part is configured to include an examination area detection model for detecting the examination area, and a diverticulum detection model for detecting the diverticulum area.
5 . The colonoscopy area indication system of claim 1 , wherein the image analysis model of the image analysis part is configured to include an examination area detection model for detecting the examination area, a diverticulum detection model for detecting the diverticulum area, and a lesion detection model for detecting the lesion area.
6 . The colonoscopy area indication system of claim 1 , wherein the image analysis model of the image analysis part is configured to include a cecum/diverticulum detection model for detecting a cecum and a diverticulum, and a lesion detection model for detecting the lesion area.
7 . The colonoscopy area indication system of claim 6 , wherein the lesion detection model has a lesion attribute identification function for determining whether a lesion is benign or malignant.
8 . The colonoscopy area indication system of claim 1 , wherein in detecting and indicating at least one selected from the group of the examination area, the diverticulum area, and the lesion area in the colonoscopy image by the image analysis part, the examination area includes an appendix, a cecum, an ascending colon, a transverse colon, a descending colon, a sigmoid colon, and a rectum.
9 - 10 . (canceled)
11 . A colonoscopy area indication system, comprising: a model loading/condition setting part that loads a colonoscopy image analysis model and a speech keyword recognition model, and set an analysis condition of the analysis model; an image receiving part that receives a colonoscopy image frame; an image preprocessing part that preprocesses a colonoscopy image received through the image receiving part, and makes a resultant preprocessed colonoscopy image ready for smooth subsequent image analysis is smoothly performed; an image analysis part that analyzes the colonoscopy image preprocessed by the image preprocessing part by using the image analysis model based on artificial intelligence (AI), and detects and indicates, on the basis of a result of analysis, at least one selected from a group of an examination area, a diverticulum area, and a lesion area in the colonoscopy image; a speech recognition part that reads audio from a buffer storing the audio while the image analysis part performs the image analysis, and analyzes the audio using the speech keyword recognition model based on AI, and recognizes a speech keyword on the basis of the result of analysis and transmits the speech keyword to the image analysis part; and a controller that checks states and control operations of the model loading/condition setting part, the image receiving part, the image preprocessing part, the image analysis part, and the speech recognition part, and initializes an analysis screen and displays a picture of a normal colon when the model loading/condition setting part completes loading of the colonoscopy image analysis model and setting of the analysis condition of the analysis model, and provides the result of analysis performed by the image analysis part, wherein the result of analysis is provided by linking an analysis target detected by the image analysis model with a speech command (keyword) related to the analysis target spoken by an examiner, wherein detecting and indicating at least one selected from the group of the examination area, the diverticulum area, and the lesion area in the colonoscopy image by the image analysis part comprises the controller transmitting, to the image analysis part, an indication condition change command based on a detection state of the diverticulum area, the command suppressing the indication of a diverticulum when the colonoscopy image corresponds to a normal colon and enabling diverticulum indication when a diverticulum is detected, and further indicating examination time and withdrawal time, wherein the speech recognition part continuously analyzes buffered audio in parallel with the image analysis and the controller links a recognized speech keyword to a corresponding analysis target only when both occur within a common temporal window.
12 . The colonoscopy area indication system of claim 11 , wherein preprocessing of the colonoscopy image by the image preprocessing part includes analysis region cropping and input size adjustment.
13 . The colonoscopy area indication system of claim 11 , wherein the image analysis model of the image analysis part is configured as a single image analysis model for detecting the examination area and the diverticulum area.
14 . The colonoscopy area indication system of claim 11 , wherein the image analysis model of the image analysis part is configured to include an examination area detection model for detecting the examination area, and a diverticulum detection model for detecting the diverticulum area.
15 . The colonoscopy area indication system of claim 11 , wherein the image analysis model of the image analysis part is configured to include an examination area detection model for detecting the examination area, a diverticulum detection model for detecting the diverticulum area, and a lesion detection model for detecting the lesion area.
16 . The colonoscopy area indication system of claim 11 , wherein the image analysis model of the image analysis part is configured to include a cecum/diverticulum detection model for detecting a cecum and a diverticulum, and a lesion detection model for detecting the lesion area.
17 . The colonoscopy area indication system of claim 16 , wherein the lesion detection model has a lesion attribute identification function for determining whether a lesion is benign or malignant.
18 . The colonoscopy area indication system of claim 11 , wherein in detecting and indicating at least one selected from the group of the examination area, the diverticulum area, and the lesion area in the colonoscopy image by the image analysis part, the examination area includes an appendix, a cecum, an ascending colon, a transverse colon, a descending colon, a sigmoid colon, and a rectum.
19 - 20 . (canceled)

Description

CROSS REFERENCE TO RELATED APPLICATION The present application claims priority to Korean Patent Application No. 10-2024-0156341, filed Nov. 6, 2024, and Korean Patent Application No. 10-2025-0001391, filed Jan. 6, 2025, the entire contents of which are incorporated herein for all purposes by this reference. BACKGROUND OF THE INVENTION Field of the Invention The present disclosure relates to a colonoscopy area indication system and method. More particularly, the present disclosure relates to a colonoscopy area indication system and method that inform an examiner of conditions, such as a diverticulum protruding from the colon wall, or indicate a diverticulum on an examination screen differently from a normal colon, in indicating a main examination area by applying an image recognition technology to a colonoscopy process. A Korean national project supported by Korean government associated with this invention is described below. Project Unique NumberNot AssignedProject Serial NumberRS-2024-00510314Government DepartmentMinistry of SMEs and StartupsSpecialized InstitutionKorea Technology and Informationfor Project ManagementPromotion Agency for SMEsTitle of Research BusinessStartup Growth Technology Development (R&D)Title of ProjectDevelopment of Artificial Intelligence-Based ColonoscopyQuality Enhancement Technology - Cecum Detection,Endoscopy Speed, and ExaminationTime Measurement TechnologySupervising InstituteWaycen Inc.Research Period1 Oct. 2024-30 Sep. 2025 Description of the Related Art Today, the incidence of colorectal cancer has been rapidly increasing due to the modernization of dietary habits and advancements in diagnostic technology. It is known that 80˜90% of colorectal cancers begin as small polyps (adenomas) in the colon. If such polyps are detected and removed early through colonoscopy, the mortality rate from colorectal cancer can be significantly reduced. The purpose of colorectal cancer screening is to detect colorectal cancer at an early stage in order to reduce mortality related to the colorectal cancer. According to previous studies, the effectiveness of colorectal cancer screening in reducing cancer mortality varies depending on the examination method. Cancer screening methods should have high sensitivity and specificity, no risks or complications, and low cost. Currently suggested methods for colorectal cancer screening include fecal occult blood testing, sigmoid colonoscopy, colonoscopy, and double-contrast barium enema. Fecal occult blood test has been reported to reduce colorectal cancer mortality by 15˜33% in large-scale randomized clinical trials conducted in Europe. A fecal occult blood test has no complications caused by the test, is inexpensive, and is relatively simple to perform. However, it has been noted to have limitations such as low sensitivity and positive predictive value in a single test, as well as a high false-positive rate that leads to the need for additional examinations. Therefore, screening using colonoscopy has been recommended in recent years, but it is applied in limited ways (such as additional examinations for people with abnormal results of fecal occult blood test) in national health screening programs targeting the general public due to relatively high cost, rare but serious complications (such as colon perforations), the examinee's pain and inconvenience caused by preparation, and lack of skilled endoscopists. In the meantime, Korean Patent Application Publication No. 10-2022-0140924 discloses “DEEP-LEARNING BASED COLONOSCOPY IMAGE ANALYSIS METHOD AND IMAGE ANALYSIS SYSTEM USING THE SAME”. The deep-learning based colonoscopy image analysis system includes: an endoscopy computer for receiving an image obtained by a colonoscope; a server for obtaining the image transmitted to the endoscopy computer through an application downloaded to the endoscopy computer in a hooking manner, and having a diagnosis algorithm for correcting the obtained image and performing a deep-learning based medical examination on the basis of the corrected image, and transmitting a result of diagnosis derived through the corrected image and the diagnosis algorithm to the endoscopy computer; and a display device for receiving and outputting the corrected image and the result of diagnosis from the endoscopy computer. In the above patent document, an image is loaded and processed using a window hooking method, allowing operation without the manufacturer's application programming interface (API). In addition, an image affected by light reflection is corrected and the size of polyps is accurately measured, thereby improving the reliability of diagnosis. However, there is no separate means for informing an examiner of conditions such as diverticula during a colonoscopy process or for indicating (displaying) the conditions on an examination screen differently from a normal colon. This carries the potential risk that the examiner may cause complications such as perforations. SUMMARY OF THE INV