Diagnostic app shortlisted for International Student Innovation Award

Eyenimi Ndiomu studied the MSc Public Health in Developing Countries in 2015/16. He has been shortlisted for the prestigious International Student Innovation Award,  which celebrates London’s most talented and innovative international students. In this blog, Eyenimi writes about Ubenwa, a cry-based diagnostic app for birth asphyxia.


I am a medical doctor from Nigeria and have practised clinical medicine in Nigeria for about 5 years post-graduation. I have also been involved in strengthening primary health care in rural communities. I studied the MSc Public Health in Developing Countries at the London School of Hygiene & Tropical Medicine.

London and Partners’ new International Student Innovation Award and showcase identifies and rewards remarkable ideas and contributions to medicine & science; technology; fashion; and art & design. The award publicly acknowledges talents from international students that graduate from London Universities each year. The award also shows how London Universities support their entrepreneurial students to develop their ideas which have social and/or commercial potential.

Our innovation – Ubenwa – addresses the need for a means of early diagnosis of birth asphyxia. Birth asphyxia causes the death of over 1 million newborns annually, making it one of the top 3 causes of neonatal mortality in the world. Early detection, which greatly increases survival rate, is simply not possible in many developing country settings due to shortage of skilled personnel, the high cost of medical equipment, and other logistics (such as non-existent power supply).

Ubenwa is a cry-based diagnostic mobile app for birth asphyxia. It harnesses the audio processing and computational capabilities of today’s mobile devices to analyse newborns’ cries and provide an accurate prediction of whether or not there is a risk of asphyxia. Using Ubenwa, health care workers and parents can quickly detect asphyxia in their babies, and promptly refer them for potentially life-saving treatment.

At the core of Ubenwa, is a machine learning-based algorithm that receives as input a baby’s cry signal, correlates it with known samples of normal and asphyxiating newborns, and outputs a classification for the current baby. The algorithm is integrated into a mobile application. It has been tested on pre-recorded cries of both normal and asphyxiating babies to give a sensitivity of over 95% and specificity of 88%.

Our next step is to conduct controlled clinical testing of Ubenwa on live babies over a period of time. On one hand, our goal is to evaluate the performance of Ubenwa in a real clinical setting and to feedback learned lessons into our design and deployment strategy. A second goal is to generate necessary data to begin to seek regulatory approval for distribution. Thus, before the competition began, my team and I already designed the entire protocol for the clinical testing and applied for ethical approval. The approval has now been issued by the ethics committee. If we receive funds from this competition, it will go into executing this clinical testing.

Personally, I intend to continue practising clinical medicine, carrying out research and learning. My hope is to keep drawing from my experiences to form collaborations that could help address some of the biggest health challenges facing the world today.

Comments are closed.