The Nigerian Artificial Intelligence (AI) Health Start-up, Ubenwa Intelligence Solutions founded by Charles Onu has developed a machine learning system that can take as input the cry of a baby, analyse the amplitude and frequency patterns in the cry, to give instant diagnosis of birth asphyxia. This condition which according to World Health Organisation is one the three major causes deaths of new-born babies worldwide, accounting for 23% of deaths, can now be easily diagnosed without medical expertise or the use of clinical methods.
The diagnostic system is now deployed as an embedded model on an Android app called Ubenwa (which means baby’s cry). This app uses a technique that have been developed for speech recognition leverages the audio processing and computational capabilities of smart mobile devices to analyse the new-borns’ cry and give a qualitative assessment of whether or not the new-born has or is at risk of asphyxia.
This big Health-Tech innovation helps parents and caregivers detect asphyxia earlier without having to wait on doctors. Ubenwa is a response to the pressing need for useful, affordable and sustainable medical tools/devices that can help tackle the increase in neonatal morbidity and mortality.
According to Charles Onu, the AI solution has achieved over 95% prediction accuracy in trials with nearly 1,400 pre-recorded baby cries. and according to Ubenwa website, the test results from the diagnostic software have shown a Sensitivity of over 86% and Specificity of 89%. The app is non-invasive and can be over 95% cheaper than existing clinical alternative.
This innovation will be having the strongest impact on the continent of Africa where death caused by Birth Asphyxia is the highest globally as the continent still records high infant mortality rate from the condition even though there are already methods and equipment that can detect it. It has caused the death of 280,000 babies a year (on their first day of life) in sub-Saharan Africa alone.
This is because only few public hospitals in Africa have the diagnostic equipment due to its high cost, and those that have the equipment still have to deal with factors like poor electricity and an unrealistic routine application for every child.
Charles Onu, the Founder and Principal Innovator of Ubenwa Software is an AI researcher, and software engineer. He obtained a Bachelor degree in Computer Science from the Federal University of Technology, Owerri in Nigeria. He is currently studying for a Master degree also in Computer Science in McGill University in Canada. He is an Associate Fellow of the Royal Commonwealth Society and was one of 12 Fellows selected globally to take part in the Jeanne Sauve Public Leadership Program in 2015.
The start-up is now raising funds to acquire more data to improve accuracy and get clinical approval from health institutions with clinical validation exercises currently being conducted in Nigeria (at the University of Port Harcourt Teaching Hospital) and Canada (at the McGill University Health Center).
“We want to do the tests in the hospital, interact directly with the babies, and compare how Ubenwa performs given all the new environmental challenges that would come up. The reason we are able to pursue this real-time validation in the clinical setting is as a result of the success of our earlier work,” Onu said.
The Start-up is already garnering international attention and is in the second round for the global IBM Watson AI XPRIZE competition, which has a $5 million prize though it is yet to figure out a definitive monetization business strategy.
“We are still finalizing a hybrid model. But in the meantime, we are planning to make it free for individuals and paid for organizations such as hospitals, clinics, governments, and others),” Onu said.