Usefulness of Automated Analysis System for Microvascular Blood Flow Rate Using Magnifying Endoscopy in Early Gastric...

Area of Science:

• Gastroenterology
• Medical Imaging
• Artificial Intelligence in Medicine

Background:

• Differentiating early gastric cancer (EGC) from patchy redness (PR) after Helicobacter pylori eradication is challenging.
• Magnifying endoscopy is crucial for EGC diagnosis, but interpretation can be difficult.
• Automated analysis of microvascular blood flow offers a potential solution.

Purpose of the Study:

• To develop and evaluate a novel automated analysis system for microvascular blood flow rate using magnifying endoscopy.
• To compare the diagnostic performance of this system against experienced endoscopists in differentiating EGC from PR.
• To assess the system's utility as a new diagnostic tool in clinical practice.

Main Methods:

• Analysis of 76 magnifying endoscopy videos (36 EGC, 40 PR).
• Development of an automated system to diagnose lesions based on microvascular blood flow rate.
• Comparison of the system's diagnostic accuracy, sensitivity, and specificity with 30 endoscopists (15 experts, 15 trainees).

Main Results:

• The automated system achieved diagnostic accuracy of 84.2%, sensitivity of 86.1%, and specificity of 82.5%.
• Endoscopists achieved diagnostic accuracy of 84.2%, sensitivity of 80.2%, and specificity of 87.3%.
• The system's sensitivity and negative predictive value were higher than those of endoscopists, with accuracy comparable to all endoscopists and superior to trainees.

Conclusions:

• The automated microvascular blood flow analysis system demonstrates diagnostic performance equivalent to or superior to skilled endoscopists.
• This system shows promise as a valuable new tool for differentiating early gastric cancer from patchy redness.
• Further clinical implementation of this automated system could enhance EGC diagnosis.