Beyond stigma: The Kenyan doctor rethinking vaginal health with AI
For millions of African women, vaginal infections remain one of the most common yet least openly discussed health challenges.
Conditions such as bacterial vaginosis, candidiasis and sexually transmitted infections continue to affect women across all age groups, often causing pain, discomfort, infertility risks and emotional distress. Despite their prevalence, diagnosis remains slow, inaccessible and, in many cases, inaccurate.
You might also like
The World Health Organization (WHO) estimates that Africa carries nearly 40 per cent of the global sexually transmitted infection burden, translating to roughly 63 million new cases annually of major curable STIs, including gonorrhoea, syphilis, chlamydia and trichomoniasis.
Bacterial vaginosis, one of the most common vaginal infections among women of reproductive age, frequently goes undiagnosed because many women show no symptoms at all. Health experts warn that untreated infections increase the risk of HIV transmission, infertility and pelvic inflammatory disease.
A Kenyan biotech startup is trying to change that through artificial intelligence.
Dr Kadryn Kadasia, founder and CEO of Misala, is leading a project titled “Context-Specific AI Diagnostics for Rapid Vaginal Health Assessment in African Women” under the Leadership for Innovation and Excellence in Accelerating Research on Women’s Health Fellowship (LEA-WH) Programme, run under the auspices of the National Academy of Medicine in partnership with the Kenya Medical Research Institute.
The project aims to develop an AI-powered diagnostic tool that can rapidly analyse vaginal samples and deliver accurate results to clinicians within minutes, rather than hours or days.
Kenya and Africa still lack rapid and reliable methods for diagnosing vaginal infections
Dr Kadasia says the project was born out of two significant gaps in women’s healthcare, and “We aim to address two gaps with this project. One is practical and the second is structural,” she told Willow Health Media.
On the practical side, many health facilities in Kenya and across Africa still lack rapid and reliable methods for diagnosing vaginal infections. As a result, women are frequently treated based on symptoms alone rather than laboratory-confirmed diagnoses.
“What tends to happen is that when women visit a clinic, they would either get a syndromic diagnosis, which is a guess and based on the experience and expertise of the healthcare professional,” she said.
This approach often leads to mistreatment, delayed care or missed infections entirely. Women may wait several days for laboratory results, if they receive them at all. For a woman already carrying the weight of pain, discomfort and uncertainty, that wait can be isolating.
The structural challenge runs even deeper. Many AI diagnostic systems being developed globally are trained on datasets that rarely include African women, raising serious concerns about whether such technologies can accurately serve African populations.
Dr Kadasia is building an Africa-first women’s health technology platform layered with AI
“One big question that we ask is, how can a model that has not been trained on data from African women serve the women in this context?” Dr Kadasia said.
To address this, Misala is building what Dr Kadasia describes as an Africa-first women’s health technology platform, one designed from the ground up around local populations, local disease patterns and local healthcare systems.
The AI tool is built to integrate into existing clinical workflows rather than replace them. For bacterial vaginosis, clinicians already rely on microscopy as the primary diagnostic method. Under the current system, healthcare workers collect vaginal swabs, prepare slides and depend on trained microscopists to manually analyse bacteria under a microscope, a process that can take hours, particularly in busy facilities with limited personnel.
Misala’s innovation adds an AI layer onto this existing process.
“A healthcare professional will collect a vaginal sample swab from a woman who has come to visit. They would prepare it, stain it, load it onto the microscope, everything routinely as they would do,” Dr Kadasia explained. “However, instead of needing an expert microscopist to read, analyse and count the number of bacteria on the slide, our technology comes in and does that.”
