Deep Learning for Precise Timing of Neonatal Hypoxic-Ischemic Encephalopathy via EEG and Seizure Pattern Analysis

This PhD project offers a unique opportunity to develop cutting-edge AI solutions to precisely time and evaluate hypoxic-ischemic (HI) brain injury in neonates using EEG data. By creating advanced deep-learning algorithms, this project aims to identify subtle EEG patterns and temporal changes post-birth, which are often undetectable to the human eye. These algorithms will enable accurate tracking of injury progression and support critical early interventions, such as therapeutic hypothermia, within the crucial treatment window.

In addition, this research will establish robust AI-based models for seizure detection, designed to analyse seizure morphology and assess treatment effectiveness across various gestational stages and therapeutic approaches. Drawing from both preclinical and clinical datasets, this project strives to enhance diagnostic accuracy and deliver real-time, actionable insights to clinicians at the bedside, ultimately advancing personalized therapy options for HI-affected neonates.

The outcomes of this research have the potential to significantly transform rapid diagnostic assessments and treatment planning, impacting neonatal care at a critical time in development.

• Deep learning
• Machine learning
• AI
• Brain physiology
• Electrophysiology signal analysis
• EEG
• Neonatal hypoxic ischemia
  

This project is eligible for funding but is subject to eligibility criteria & funding availability.

Page expires: 4 July 2025