Saratani AI

Saratani AI Company Limited

Saratani AI

An AI-based decision-support algorithm for cervical cancer screening.

Dar es Salaam

TZA

For-profit, including B-Corp or similar models

—

Cervical cancer is the fourth most common cancer among women globally, with over 85% of cases and 90% of deaths occurring in low- and middle-income countries. In Africa, it is the second most frequent cancer and the second leading cause of cancer deaths. Tanzania bears one of the heaviest tolls as the disease is the most diagnosed cancer and the top killer of women in the country. This is particularly tragic because cervical cancer is one of the most preventable cancers with timely screening and early treatment. Early screening through Pap smears can prevent most cases, but with only 43 pathologists for over 63 million people in Tanzania, timely diagnosis is nearly impossible. Manual analysis of Pap smear slides is time-consuming, error-prone, labor-intensive, and overwhelmed by demand, leading to a screening turnaround time of up to three weeks. As a result, nearly 70% of cases are detected too late, condemning 6,832 women to preventable deaths each year. Without intervention, cases could double in a decade. The urgent need is clear: faster, innovative and scalable solutions to save lives.

To bridge the existing gap between the shortage of pathologists and the growing demand for cervical cancer screening, we have built Saratani AI. This is an AI-based decision support algorithm that helps pathologists analyze Pap smear slides 20 times faster while improving accuracy. Developed using locally available Pap smear slides, the algorithm has been trained to recognize cellular abnormalities specific to our context. At the core of our solution is deep convolutional neural networks, a form of machine learning that mimics human pattern recognition but with superhuman speed and consistency. Here’s how it works: A pathologist uploads a digitized Pap smear image to our web-based application. Within seconds, the algorithm analyzes the image, provides the results, and highlights suspicious areas through a heat map, guiding the pathologist’s attention to potential lesions. It also supports remote teleconsultation, enabling pathologists to collaborate in real time, without needing to be in the same location. The value? A turnaround time for analysis cut from 15–20 minutes to less than 60 seconds. Enhanced accuracy through AI-driven consistency. Visual heat map support that improves decision-making. And the power to collaborate across distances. Simply put, Saratani AI empowers pathologists to work faster, smarter, and save more lives.

Our solution serves two critical groups: Pathologists and women at risk of cervical cancer in Tanzania. For pathologists, who currently face an overwhelming workload and mental exhaustion, our decision-support tool is a lifeline. It reduces their screening time per case from 20 minutes to under 60 seconds. The built-in heat maps guide their focus, and the teleconsultation feature enables faster collaboration without needing to be in the same room—or even the same region. For women, this means the difference between life and a preventable tragedy. Today, 70% of cases are caught too late because results take weeks. Our solution slashes the wait, helping to catch pre-cancerous lesions earlier, when treatment is simplest and survival rates exceed 90%. It means faster, more accurate diagnosis and timely initiation of treatment. By combining these needs, we’re transforming cervical cancer from the leading cause of cancer death in the country to a manageable health challenge. Every two minutes saved for a pathologist is a life potentially saved for a mother, daughter, or sister.