Performance of a Full-Coverage Cervical Cancer Screening Program Using on an Artificial Intelligence- and Cloud-Based...

Area of Science:

• Public Health
• Medical Technology
• Oncology

Background:

• The World Health Organization (WHO) aims to eliminate cervical cancer, requiring 70% screening coverage.
• China's action plan includes a provincial full-coverage screening program in Hubei, utilizing AI and cloud diagnostics.

Purpose of the Study:

• Evaluate the performance of an AI diagnostic system in a full-coverage cervical cancer screening program.
• Assess indicators including accessibility, efficiency, diagnostic quality, and cost.

Main Methods:

• Analysis of screening data from 1,704,461 individuals (July 2022-January 2023).
• External diagnostic quality control on a random sample of 220 individuals.
• Assessment of program costs across participating institutions.

Main Results:

• Screening services reached all districts, with high rural participation (67.54%).
• AI achieved 13.45% coverage in 6 months, projected for full coverage in ~1 year (87.5x faster than manual methods).
• High compliance rates (>96%) and a low cost of ¥49 per person were reported.

Conclusions:

• AI-assisted diagnosis is accessible, efficient, reliable, and cost-effective for cervical cancer screening.
• AI technology is crucial for increasing screening coverage rapidly, supporting WHO's elimination goals, especially in resource-limited areas.