Post-doctoral scientist, AI and Deep Learning for intrapartum CTG
- I am a machine learning scientist and my research focuses on developing innovative Deep Learning models to analyse big CTG data. The data consists of 100,000 births at term data available at Oxford that is collected to develop new generation data-driven cardiotocography (CTG) analysis and risk assessment.
- The aim of the project is to utilise Deep Learning algorithms in an individualised data-driven tool for clinical decision support, preventing brain injury of the baby during childbirth.
I studied electronics and communication engineering in my undergraduate education at Mekelle institute of technology, Tigray, Ethiopia. During this BSc degree, I had a great interest in digital signal processing and in my final year thesis, I worked on developing speaker recognition using Hidden Markov model. To pursue my academic interest further, I studied a Master's degree in vision and robotics (VIBOT), Erasmus Mundus joint master’s program conducted by Heriot-Watt University in Edinburgh, Scotland, Universitat de Girona in Girona, Spain and Université de Bourgogne in Le Creusot, France. The focus of my master’s thesis was on developing a deep convolution neural network to classify coral reefs.
In Jan 2017, I started my PhD at the Crabb lab at City university of London under the supervision of Prof. David Crabb and Dr. Pete Jones. The objective of my PhD work was to develop data analysis methods to detect glaucomatous visual field loss from natural eye movements. I conducted series of experiments and developed novel summary statistics measures to capture the spatiotemporal characteristics of eye movements. Furthermore, I developed a method to classify natural eye movements in order to separate a group of patients with glaucoma from age-matched healthy controls.
Currently, I am a postdoctoral scientist working in analysing cardiotocography (CTG) recordings using modern deep learning methods.