EW BrightSparks 2022 profile: Dr Ejay Nsugbe

Author: EIS Release Date: Aug 24, 2022

Now in its fifth year of awards, EW BrightSparks sees Electronics Weekly partner with RS Grass Roots to highlight the brightest and most talented young engineers in the UK today.

 

Here, in our series on the latest EW BrightSparks of 2022, we highlight Dr Ejay Nsugbe, the Founder and Chief Engineer at Nsugbe Research Labs.

 

Achievement

Having an MSc from the University of Sheffield (Engineering Control Systems) and PhD from Cranfield University (Control and Instrumentation), Dr Ejay Nsugbe has shown scientific entrepreneurship by founding Nsugbe Research Labs where he has been applying AI technology to help innovate in various aspects of modern day medicine.

 

Particularly, he shared, he has been involved in areas where the limits of medical capability and knowledge constrain practitioners from meeting patient needs. And there’s also the aim of  improving affordability for treatments. For example, Dr Ejay has been working in various AI-related projects involving Data Science/AI in medicine over the previous year.

 

Firstly, there is Gesture Recognition for Prosthesis Control Using Wearable Sensing. Transhumeral amputees are the largest Upper-Limb amputee cohort in the UK, we were told. Despite this it remains challenging to design bionic upper-limb prosthesis for this group of amputees due to the poor signal quality from the residual stump.

 

He attempted to resolve this problem by using ergonomic and affordable wearable sensors alongside signal processing and Machine Learning (ML) methods for an intelligent control system to solve the gesture motion recognition to facilitate prosthesis control (this project also involved a Consultant Radiologist). The system allows for greater affordability of a transhumeral prosthesis control system while also solving the motion recognition problem.

 

This innovation received recognition from the Welcome Trust, with a seedcorn bursary, and the award of Best Post-Doctoral Research by the AI.Awards society at the 5th AI Annual Awards. There was also membership and an accompanying membership bursary by the International Society of Prosthesis and Orthotics, despite it not being a clinical prosthesis

 

Secondly, Dr Ejay has worked on Preterm Prediction using Uterine Contraction Signals and Machine Learning. We learned that Preterm is viewed by the WHO as a global epidemic and one of the leading causes of death in children under the age of 5.

 

Using acquired uterine contraction signals, this work looked into the design of an AI decision support model that can be incorporated into a clinical setup and capable of predicting preterm from uterine signals in the third trimester of pregnancy. The designed models were also extended towards a labour immanency forecast.

 

Papers

The work has also involved an international scale collaboration with researchers from China, Canada and Sheffield, with the UK at the centre of the innovation. Highlights of manuscripts produced and of some of the finding include the following;

 

There was a paper on a ‘Prediction Machine’ design which is intended for a Clinical Cybernetic framework for early identification of preterm followed by a recommended course of care intervention for the patient. The impacts of this would remove diagnosis subjectivity from the preterm obstetric process, and minimise False Positive diagnosis.

 

There was also a project involving uterine contraction signals acquired using Uterine EMG or Magnetomyogram (MMG), which is a highly expensive option due to large electrode channels to record signal across the womb of the pregnant patient. In this paper, a metaheuristic based optimisation and ML algorithm were combined to obtain low channel MMG representation capable of predicting pregnancy labour immanency. This was showcasing a combination of AI and ML for obstetric interventions, alongside clinical cost saving benefits increasing its appeal to low income countries:

 

Finally, Dr Ejay has also studied that gestation lengths vary amongst ethnicities: as part of strides towards personalised care strategies in obstetric medicine, it is necessary to take into consideration ethnicity when ruling a pregnancy as Term/Preterm. This paper, which he co-authored, investigated the use of ML to form a multi-tier based preterm prediction model which takes an ethnic-prior into account before making a classification. The ethnic specific model also showed a superior performance when compared with a generalised prediction model.

 

Community / STEM

In terms of involvement with wider society, Dr Ejay has also been an active ambassador and promoter of the benefits and impacts of AI to non-experts.

 

For example, he has been active to help disseminate his work to a wider audience, at general public and school level, to promote the mainstream appeal and impacts of AI technology. This has involved giving talks at schools on his on-going research work and writing articles geared towards secondary school readers and non-experts where he distilled the framework and ideas behind his AI-driven innovations.

 

He regularly takes part in STEM events and more recently he was invited to a school in the Bristol area where he talked to kids about how AI is allowing for the design of bionic arms, as can be seen in this post.