A Nigerian startup has developed an app, Ubenwa which is a learning system with the capacity to detect child birth asphyxia. The founders hope to save the lives of thousands of babies when the technology is fully brought into use.
The young startup is gaining a lot of international attention. It has already enrolled in the IBM Watson AI XPRIZE competition, with a prize worth of $5 million.
— TheLensNaija (@Thelensofficial) October 2, 2017
Developers of Ubenwa say the AI solution has achieved more than 95% prediction accuracy with almost 1,400 recorded baby cries. Currently, the startup is raising funds to enable them acquire more resources. The aim is to improve the machine’s accuracy and also, obtain approvals from clinical institutions.
According to WHO, birth asphyxia ranks the third highest under-five deaths and about a million neonatal deaths annually. It has also been linked to 1.1 million intrapartum stillbirths, neurological disabilities or impairment. Dr. Victoria Feyikemi, a physician at Babcock University Teaching Hospital in Ogun state, Nigeria, says the condition could be a manifestation of the immaturity of a baby’s respiratory system.
It’s a challenge detecting the condition in Nigeria and most parts of Africa. Thus, Feyikemi says “medical expertise is needed to positively establish diagnoses using blood gas analysis, and administer oxygen support to treat the underlying cause.”
How Ubenwa does it
Ubenwa, which means baby’s cry, helps parents detect asphyxia earlier without having to wait on doctors. Ubenwa’s founder, Charles Onu explained that the machine uses an infant cry as input, analyses the amplitude and frequency patterns of the cry and provides instant diagnosis of birth asphyxia. Despite the fact that the condition is detectable, just few public hospitals have been able to have the machine. Frequent electricity failure, high cost and unrealistic medical routines for most children, have hastened the need for Ubenwa.
Udeogu Innocent, co-founder and engineering lead of Ubenwa, said after achieving a certain level of success, they deployed a mobile app to easily diagnose birth asphyxia with speech recognition techniques. Ubenwa’s team is currently conducting clinical validation exercises in Nigeria and Canada.