Health Outcome Prediction, CY P03 24 04

This research project focuses on the development of a retrieval-based machine learning model and a visual analytics interface for healthcare applications. In healthcare, while routinely collected data presents several opportunities for predicting adverse events and understanding the association between them and specific risk factors, the complex and incomplete nature of the data can lead to inaccurate and misinformed explanations. To overcome these limitations, our project aims to develop a machine learning (ML) model that incorporates relevant information from auxiliary data sources and memories to enhance outcome predictions. Additionally, we will create a visual analytics interface that supports feature engineering and decision provenance, enabling human decision-makers to access and utilise previously acquired knowledge. This research seeks to improve the explainability of ML models in healthcare and provide decision-makers in Public Health Management (PHM) with a robust tool for informed decision-making.

01/02/2024

4 years