Classification for invasion depth of esophageal squamous cell carcinoma

Ishihara_headshot Post written by Ryu Ishihara, MD, from the Department of Gastrointestinal Oncology, Osaka International Cancer Institute, Osaka, Japan.

The aim of this study was to develop a computerized image analysis system using deep learning to diagnose the invasion depth of superficial esophageal squamous cell carcinoma (SCC). Diagnosing cancer invasion depth of esophageal squamous cell carcinoma is still difficult, and the findings are subjective and liable to inter-observer variability.

In this study, we developed the AI system to differentiate between EP-SM1 and SM2/3 cancers. Our AI system showed a favorable performance for diagnosing invasion depth of superficial esophageal SCCs, with a sensitivity, specificity, and accuracy of 90.1%, 95.8%, and 91.0%, respectively. The values were comparable to those obtained by experienced endoscopists.

Our AI system analyzed 30 images per second, which met the required speed for analyzing video images, indicating that further developments in technology will allow the real-time diagnosis of cancer invasion depth in the near future. Moreover, there are many innovations in the field of AI, including semantic segmentation, which provides pixel-level labeling for image classification. This represents a potentially important technique for image recognition because each pixel is labeled as belonging to a given semantic class. We therefore expect that the diagnostic accuracy could be further improved by implementing this method or by accumulating more training data.

Ishihara

Read the full article 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.

Leave a Comment

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

WordPress.com Logo

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

Google photo

You are commenting using your Google 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