We are making an artificial intelligence system that is fast, accurate and affordable for cervical cancer screening and diagnosis. Like most cancers, especially in low- and middle-income countries, most patients get diagnosed at a very late stage which makes it difficult to be managed and they often end up just receiving palliative care. On top of that, we have too few Human Resources and most of them are located at tertiary level hospitals. This includes pathologists, who are responsible for cancer diagnosis. This means it can take anywhere from one to three weeks to get a diagnosis for cervical cancer, which causes both anxiety for patients and unnecessary delays in treatment.
In order to overcome these challenges, we have come up with the Saratani AI system, which leverages AI technology to make cervical cancer screening fast, accurate, and affordable. And this will not only service tertiary hospitals, but can be implemented down to the primary healthcare facility level where most patients reside.
We have a lot to celebrate. Initially, we used publicly available data sets, but now we are collecting data from local hospitals and have established partnerships with two national facilities (Muhimbili National Hospital in Tanzania and Ocean Road Cancer Institute). We have also purchased a digital microscope to collect data. That was initially a major obstacle in our startup and now we're going to get a local data set that reflects the reality of the patients we serve.
A significant challenge we are facing is gaining acceptance from medical personnel themselves. There is the perception that AI is going to replace them, which is not true. Our tool is actually there to complement their work and improve patient care. We have, however, seen a significant improvement in the perception of AI since we initially started in 2020. In order to overcome this challenge, we take the time to sit down with health workers, explaining the process to them, acknowledging the potential downsides of AI in healthcare, but also helping them to see the positives and working with them to mitigate the impact of the potential negatives.
Ideas are useless on their own; execution is required to bring an idea to life. This means that you need a strong team that believes in the vision and the product or service that you are building. There will be challenges, so without a strong team it will be very hard to navigate them when they arise.
Another thing to consider when you're working on a startup is mapping your stakeholders early on. Who do you want to collaborate with? When you have these people and organisations in mind early on, reaching out to them and establishing relationships, it means you are not on your journey alone. Find organisations that are doing things that complement your work and work alongside them. For example, it would have been very difficult for us to collect data from the national health facilities if it wasn’t for the partnership that was built early on.
Lastly, building a startup is not a straightforward process. Things do not always go as planned or take the amount of time you expect it to. Be patient and understand that building a startup is not predictable or linear.
As I mentioned, we are celebrating starting to collect our data sets and purchasing the microscope. Both of these things were made possible because of the funds we received from Villgro. Villgro also connected us with the Africa Oxford Initiative, where I applied and was accepted into their program for innovation and entrepreneurship. Additionally, the mentorship we are receiving from Villgro is helping us become more focused; where we previously were just using trial and error, we now are going through specific steps that help us improve our product and service.