Skip to content

Ms Liz Felton

Job: PhD student

Faculty: Computing, Engineering and Media

School/department: School of Computer Science and Informatics

Address: De Montfort University, The Gateway, Leicester, LE1 9BH

T: N/A

E: elizabeth.felton@dmu.ac.uk

 

Personal profile

Liz Felton graduated with a Bachelor's degree in Artificial Intelligence, before spending some time working in technology communication and for a financial technology company. Her love of research brought her to DMU, where her MSc project proved that biometric data could be mapped to future mood states using time series analysis. This work forms the basis for her PhD project, which focuses on improving patient care by working to predict mental health crisis states. Alongside the PhD, she is a part-time lecturer with the university's Games, Maths and Intelligent Systems department and acts as the staff liaison for the student-led robotics club.

Research group affiliations

Institute of Artificial Intelligence (IAI)

Key research outputs

[Add Key Research Ouputs information here]

Research interests/expertise

Affective computing; AI for healthcare

Areas of teaching

  • Introduction to CI and Control Systems
  • Mobile Robotics 
  • Staff Liaison at Robot Club

Qualifications

BSc Artificial Intelligence with Computer Science (University of Birmingham, 2014)
MSc Intelligent Systems and Robotics (De Montfort University, 2019)

Honours and awards

The PhD project has been awarded a full scholarship for 3 years from De Montfort University.

Key articles information

[Add key research outputs information here] 

e.g ‘Outputs, e.g. selected publications’ produced since January 2008

PhD project

Title

Early mental health crisis intervention using user-generated and biometric data, with artificial intelligence techniques

Abstract

Design a system that monitors mental health over time, to provide recommendation for when early intervention is required (before crisis state is reached). The system may use external data, user-generated data (e.g. self-reported mood states), biometric data, and mobile phone data. The system will use an appropriate prediction technique to predict potential crisis states ahead of time. There is an opportunity to use existing verified datasets of mood prediction data to build a generalised model of mood reactions and personalise it to each participant based on their individual data. The systems may predict crisis state for multiple conditions.

Name of supervisor(s)

Professor Yingjie Yang, Professor Parvez Haris, Dr Pamela Hardaker