Zim man invents disease-detecting vosho app
WHILE Zimbabwean artificial intelligence fundi and strategist Babusi Nyoni might have initially designed his Vosho Fosho as a way to measure who was better at executing the viral dance that swept through Southern Africa a few years ago, a collaboration with a local nursing home turned it into a potential life saving app after further development revealed it could help detect Parkinson’s disease.
Released at the end of 2018 just in time for that year’s festive season, Nyoni released the app as a joke because of his failure to master a dance style that was rampantly popular that year.
The Vosho is not an easy dance to master. Those that are not nimble-footed are unlikely to come back up after they have executed the dance’s trademark sudden downward movement with the twist of the hip and bend of the leg.
Initially, Nyoni was only interested in observing who the best dancers were by observing how quickly they went down and up without help. Here was an app that was about to prove who the best vosho dancers were, he thought. There was no need for bar debates on the subject anymore.
However, after consideration, Nyoni thought that the app’s ability to map the progress or decline of a person’s motor movements over time could help in the diagnosis of Parkinson’s disease. After partnering with Ekuphumuleni Geriatric Nursing Home in his hometown Bulawayo, South Africa-based Nyoni developed his vosho app into Patana Al, an app that can be used to test for an ailment that usually afflicts older people.
For Nyoni, the app is an example of useful experimentation and innovation.
“Tech has to serve a purpose. It has to be contextual to where it’s made. When we started Tripleblack Agency we wanted to make impactful things using (new) technology, but a lot of our clients didn’t understand the tech. So instead, we took a year to experiment with machinelearning and artificial intelligence on our own terms,” Nyoni told the Mail & Guardian.
A year after its creation, the prototype for the app has been showcased at the Tensorflow World Conference where Nyoni had been invited to speak about its uses beyond rating dance moves and possibly, in conjunction with guidance from healthcare practitioners, diagnosing Parkinson’s disease. The Silicon Valley trip was part of a year-long speaking tour for Nyoni, which saw him visiting France, Switzerland, Kenya and the Fak’ugesi African Digital Innovation Festival in Johannesburg at the end of 2019.
Despite gaining global acclaim, Nyoni still insists that his product is proudly African.
“We appreciate the work the TensorFlow.js team has put into making their machine learning framework work on low-end mobile devices to power apps such as this. It is a deliberately African solution to a global problem,” he said.
Always in search of new innovations, Nyoni has also made a robot that writes jokes similar to those usually made by fast food outlet Nandos and an app that can tell black women’s hairstyles.