AI-supported cervical cancer screening in East Africa

Researchers at Uppsala University, Karolinska Institutet, and the University of Helsinki tested whether AI-supported diagnostics could help detect cervical cancer in rural hospitals in Kenya and Tanzania, where access to pathologists and laboratories is limited.

The study involved 3,000 women who otherwise would not have been offered cervical cancer screening. At local hospitals, cervical cell and HPV samples were collected, digitized, and analyzed using artificial intelligence. Pathologists also examined the samples to compare results.

“In our study, we showed how AI can be used to detect cervical cancer in areas where there is otherwise limited access to pathologists and laboratories. Using digital tools, samples can be analysed faster and with fewer experts involved, meaning that more women can get access to screening. But for the AI to really work, it takes more than just the technology – it needs investments in staff, equipment and trust in the healthcare system,” says Nina Linder, the study’s lead author.

Local nurses, laboratory staff and pathologists were trained to use the system, which was integrated into local healthcare routines. Women showing signs of disease received treatment according to national guidelines.

One challenge in applying AI was the consistency of the digitized images. Variations in staining reagents, reagent quality, and frequent power outages affected image accuracy and testing speed.
“The AI method worked well technically, but unreliability in the supply of reagents, variations in reagent quality and power outages all affected accuracy as well as the capacity to perform these tests rapidly, including HPV analyses,” explains Nina Linder.

Another difficulty was patient follow-up.
“In Tanzania, we had quite a few problems with follow-up. Some women did not come back, and when we later checked their samples, it turned out that they had changes that needed treatment,” says Linder.

Despite these obstacles, the study demonstrates the potential of AI to expand access to diagnostics in resource-limited healthcare systems.
“For decades, diagnostic methods that are proven to be effective for women’s health – such as cell-sample based cervical cancer screening – have been dependent on highly trained experts. With the latest advances in medical AI, we can now re-evaluate these methods and introduce them even in resource-limited settings, making life-saving diagnostics far more accessible,” says Professor Johan Lundin of Karolinska Institutet.

MEDICA-tradefair.com; Source: Uppsala Universitet