Researchers have demonstrated that artificial intelligence can detect endometrial cancer with remarkable accuracy by analyzing samples from a simple vaginal swab, according to a study that could transform screening for the most common gynecologic cancer in developed countries.

The diagnostic approach, which uses machine learning algorithms to identify cancer markers in vaginal fluid, showed near-perfect detection rates in clinical testing, as reported by Rare Cancer News. The finding represents a potential breakthrough for a disease that currently requires invasive tissue biopsies for definitive diagnosis.

Endometrial cancer—which develops in the lining of the uterus—affects more than 65,000 women in the United States each year. Current diagnostic pathways typically begin with ultrasound imaging followed by endometrial biopsy, an uncomfortable procedure that requires inserting instruments through the cervix to collect tissue samples.

The AI-based swab test could offer a less invasive first-line screening option, particularly valuable for women experiencing abnormal bleeding or other warning signs. Early detection significantly improves survival rates, with five-year survival exceeding 95% when caught at localized stages.

While the study results appear promising, the technology will require validation in larger, more diverse patient populations before potential clinical deployment. Researchers have not yet disclosed whether the test can distinguish between cancer stages or identify pre-cancerous conditions, both critical factors for clinical utility.

The development aligns with broader efforts to apply artificial intelligence to cancer detection, including AI-assisted analysis of mammograms, cervical screening samples, and pathology slides. These technologies aim to improve diagnostic accuracy while reducing healthcare costs and patient discomfort.

The research team has not announced a timeline for seeking regulatory approval or commercial development of the diagnostic test.