Skip to main content

Multimodal AI and Machine Learning for Diabetes-Related Complications: Integrating Clinical Prediction, Decision Support and Longitudinal Risk Forecasting for Diabetic Foot Disease and Lower-Limb Amputation

Primary supervisor

Jesmin Nahar

Research area

Machine Learning

Background

Diabetes can lead to serious complications, including peripheral neuropathy, peripheral arterial disease, diabetic foot ulcers and lower-limb amputation. Early identification of patients at increased risk is important for timely intervention, preventive care and improved clinical outcomes. The increasing availability of electronic health records, clinical measurements and longitudinal patient data provides opportunities for advanced Machine Learning (ML) approaches.

Research Aim

This PhD research aims to develop advanced Machine Learning methods for prediction, risk stratification, clinical decision support and longitudinal forecasting of diabetes-related complications, with a particular focus on diabetic foot disease and lower-limb amputation risk.

Research Approach

The research will investigate electronic health records, glycaemic measures, clinical history, demographic information and, where available, longitudinal patient data. Machine-learning and ensemble approaches will be developed to predict diabetic foot complications, peripheral neuropathy, peripheral arterial disease and amputation risk.

Prediction and Decision Support

The research will develop interpretable predictive models to identify high-risk patients and provide risk estimates that could support early clinical intervention, preventive care and personalised management.

Longitudinal Forecasting

Where suitable longitudinal data are available, temporal and predictive modelling techniques will be investigated to forecast changes in patient risk and future diabetic foot complications or lower-limb amputation outcomes.

Explainable and Responsible ML

Explainable Machine Learning techniques will be explored to identify important clinical predictors and improve the transparency of model predictions. The research will consider model robustness, generalisability, fairness, uncertainty and responsible use of ML in healthcare.

Expected Contribution

The project aims to develop a robust, interpretable and clinically relevant ML framework integrating prediction, decision support and longitudinal forecasting to improve early identification and management of patients at risk of diabetic foot disease and lower-limb amputation.

Required knowledge

Required Knowledge

Students should have a strong background in:

  • Machine Learning and Data Science
  • Python and/or R programming
  • Data preprocessing, feature engineering and statistical analysis
  • Supervised and unsupervised machine-learning techniques
  • Ensemble learning and predictive modelling
  • Risk stratification and prognostic modelling
  • Temporal and longitudinal data analysis and forecasting
  • Model development, validation and performance evaluation
  • Explainable Machine Learning and interpretable modelling
  • Healthcare, clinical or electronic health record (EHR) data analysis
  • Basic knowledge of diabetes and diabetic foot disease is desirable
  • Research methodology, critical thinking and scientific communication
  • Responsible and ethical use of Machine Learning in healthcare

Learn more about minimum entry requirements.