Esophageal cancer screening with new AI-powered model

Serdar HocamAuthor & Editor

The developed artificial intelligence model detects esophageal cancer using non-contrast computed tomography scans.

◉ 2 views
Large-scale esophageal cancer screening through noncontrast computed tomography and artificial intelligence - Nature Medicine

Researchers have developed a new artificial intelligence model capable of detecting esophageal cancer and pre-cancerous lesions by analyzing non-contrast computed tomography scans.

Early Diagnosis in Esophageal Cancer

Esophageal cancer is known as a common digestive system cancer with high mortality rates due to late diagnosis and the insufficiency of standard screening programs.

The AI Model Named EAGLE

Introduced within the scope of the research, the new artificial intelligence model named EAGLE identifies cancer and malignant lesions such as high-grade intraepithelial neoplasia through non-contrast computed tomography scans.

Two-Stage Working Mechanism

In the first stage, the system locates the esophagus within the three-dimensional tomography scan, and in the second stage, it segments the lesions while simultaneously predicting the probability of cancer.

Success Rates in International Tests

Achieving high accuracy rates in internal and external tests involving various centers such as China, the Czech Republic, and Australia, the model holds the potential to improve screenings in hospitals.