AI Improves Cervical Cancer Screening in Low-Resource Settings

The lack of access to laboratories, pathologists, and reliable healthcare infrastructure hampers diagnostic capacity and timely intervention. Now, a new study has shown that artificial intelligence (AI) can analyze cervical cell samples at the primary healthcare level, improving screening accuracy and accessibility in resource-limited areas.

Researchers from Karolinska Institutet (Stockholm, Sweden), Uppsala University (Uppsala, Sweden), and collaborators conducted the study to evaluate an AI-based system for cervical cancer screening. The research was implemented through clinical projects in rural Kenya and Tanzania, in collaboration with Kinondo Kwetu Hospital and Muhimbili University of Health and Allied Sciences, involving 3,000 women. Cell samples were digitized on-site and analyzed using AI, followed by review by medical experts. Local healthcare workers were trained to operate the system, which was integrated into existing healthcare frameworks.

The study demonstrated that AI-assisted screening can significantly improve access to diagnostics where trained specialists and laboratory facilities are scarce. The findings, published in The BMJ, suggest that the system can enable earlier detection and treatment, reducing cervical cancer mortality rates. However, the technology’s effectiveness depends on the availability of supporting healthcare infrastructure, including electricity, medical supplies, and community trust.

According to the research team, AI-based diagnostics must be adapted to local conditions and integrated with national healthcare programs to ensure equitable outcomes. They also highlight the need to combine cell-based AI analysis with molecular HPV testing for more comprehensive screening approaches. Future development should focus on offline functionality and low-resource compatibility to ensure effectiveness in remote and underserved regions.

“Only a third of women worldwide have been screened for cervical cancer, even though serious disease is preventable,” says the study’s last author, Johan Lundin. “AI enables screening in low-resource settings where access to pathologists and laboratories is limited, and can shorten the time from sampling to diagnosis.”

Related Links:
Karolinska Institutet
Uppsala University