Computer-aided diagnosis by using endocytoscopy

Dr. MoriYuichi Mori, MD, PhD from the Digestive Disease Center at Showa University Northern Yokohama Hospital in Yokohama, Japan discusses this Original Article “Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos).”

We developed the brand-new automated diagnostic system for colorectal polyps by using endocytoscopy which can visualize in vivo nuclei and gland lumens. This system automatically extracts nuclei spots from an original endocytoscopic image and analyzes the six features of the nuclei (eg, size, variation of size, etc.), then instantly outputs the predicted pathology on the basis of multivariate logistic regression.

Figure 2

Figure 2. Sequence of image processing for nuclear segmentation from an original endocytoscopic image to extracted nuclei.

Endocytoscopy has a high diagnostic ability for characterization of colorectal polyps, but it requires training and experience. With this automated diagnostic system, even novice endoscopists can derive benefits from the endocytoscopy’s high diagnostic ability.

The automated diagnostic system provided instant classification of colorectal polyps with the sensitivity of 92%, specificity of 79%, and accuracy of 89% with perfect objectivity. It can be a powerful support to facilitate appropriate decision making during colonoscopy.

Watch the accompanying videos on the GIE journal website that demonstrate the use of this system during clinical practice.

Find the article abstract online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.

One thought on “Computer-aided diagnosis by using endocytoscopy

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s