Emma Rocheteau
Scholar

Emma Rocheteau

Google Scholar ID: NZUdzyIAAAAJ
PhD Student, University of Cambridge
Machine LearningDeep LearningArtificial IntelligenceHealthcareElectronic Health Records
Citations & Impact
All-time
Citations
259
 
H-index
5
 
i10-index
5
 
Publications
15
 
Co-authors
9
list available
Resume (English only)
Academic Achievements
  • 1. Published 'Representation Learning for Patients in the Intensive Care Unit', focusing on improving patient outcomes and healthcare system efficiency, including predicting patient deaths and estimated discharge dates.
  • 2. Published 'Dynamic Outcomes-Based Clustering of Disease Trajectory in Mechanically Ventilated Patients', training different time series models to uncover hidden patient subtypes.
  • 3. Published 'Temporal Pointwise Convolutional Networks for Length of Stay Prediction in the Intensive Care Unit', proposing a new TPC model that significantly outperformed LSTM and Transformer.
Research Experience
  • 1. Worked on machine learning problems for healthcare as part of the AI group in the computer science department at the University of Cambridge, focusing on predicting patient outcomes in the Intensive Care Unit (ICU).
  • 2. Designed deep learning models inspired by clinical decision-making, using temporal and pointwise convolution to process time series data in Electronic Health Records (EHRs).
  • 3. Worked on using graph neural networks to link experiences of similar patients, especially for rare diseases.
Education
  • 1. MB BChir (in progress), 2023, University of Cambridge
  • 2. PhD in Machine Learning for Healthcare, 2022, University of Cambridge
  • 3. BA in Engineering/Preclinical Medicine, 2016, University of Cambridge
Background
  • ML for Healthcare Researcher, Final Year Medical Student at the University of Cambridge. Research interests include Deep Learning, Graph Neural Networks, Electronic Health Records, and Healthcare.
Miscellany
  • Interests include research in the healthcare domain. Follow her on Twitter for the latest updates.