Home » Health » How AI is Changing the Colorectal Endoscopy Landscape for Early Detection of Colorectal Cancer

How AI is Changing the Colorectal Endoscopy Landscape for Early Detection of Colorectal Cancer

Under the colorectal endoscope in white light image mode, the AI ​​system can use the blue frame to detect and mark the location of abnormal mucosal protrusions suspected of polyps in real time, reminding the endoscopist for further evaluation. (Photo/Provided by Yadong Hospital)

Malignant tumors rank first among the top ten causes of death among Chinese people, and the mortality rate of colorectal cancer ranks third among cancer patients. Eating habits of colorectal cancer are closely related. With the westernization of diet, changes in living habits, metabolic syndrome and the increase of obese population, the incidence rate is also increasing year by year. In the development of colorectal cancer, colorectal polyps are considered as precursor lesions that develop into tumors.

A large intestinal polyp is a protruding lesion formed by abnormal proliferation of large intestinal mucosal cells. When a large intestinal polyp is less than one centimeter or histologically hyperplastic, it is a benign lesion and will not have obvious symptoms or malignant lesions; however, if the polyp exceeds one centimeter or The occurrence of cell glandular lesions may lead to the possibility of canceration. There will be no obvious symptoms in the early stage. When the polyps are large enough or the symptoms of cancer metastasis appear, it is usually too late.

According to clinical research, early detection and removal of polyps by colonoscopy can reduce the mortality rate of colorectal cancer by 53%. In addition, for every 1% increase in the detection rate of colorectal endoscopic polyps, the risk of cancer can be reduced by 3%. Early detection, early treatment and removal of polyps are crucial to preventing the occurrence of colorectal cancer.

In recent years, the development of artificial intelligence in gastrointestinal endoscopy is changing with each passing day. The Ultrasound and Endoscopy Center of Yadong Hospital has introduced “CAD EYE Artificial Intelligence Endoscopy Imaging System”. Instead of relying on experience to judge the status of colorectal polyps in the past, doctors can now “real-time detect” colorectal polyps during endoscopic examination, and “automatically interpret” to distinguish the types of polyps, greatly improving the accuracy of diagnosis, and Provide guidance on whether to resect the treatment. Utilizing more than 250,000 colonoscopy images for AI deep learning, the system is an innovative technology specially developed for colonoscopy.

Zhong Chengxuan, director of Ultrasonic Endoscopy Center of Yadong Hospital, said that the “CAD EYE Artificial Intelligence Endoscopy Imaging System” can improve and assist doctors in the accuracy of endoscopic examination and treatment. During the inspection process, the system uses white light imaging and linked color imaging modes for detection. When an abnormality is found, a blue frame will be displayed to remind the doctor of suspected polyps or tumors. The sensitivity can reach 94% to 96%, which is better than past experience. .

And when using the blue light image mode to observe, the AI ​​function can instantly diagnose and identify the surface structure of polyps, and classify them as hyperplastic polyps or neoplastic polyps. The probability of hyperplastic polyps turning into malignant tumors in the future is low. Polyps with a relatively high risk of becoming cancerous in the future are recommended to be removed.

Through the intelligent assisted application of AI, CAD EYE not only becomes the right-hand man of endoscopists, but also provides real-time and sensitive detection information, so that doctors can make more accurate diagnosis and treatment options.

2023-06-01 09:05:36

#artificial #intelligence #endoscopic #imaging #system #detects #polyps #real #time #Precise #prevention #colorectal #cancer #Life #CTWANT

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.