Dr Chris Duckworth

Research groups

Research interests

• Artificial Intelligence for Health and Wellbeing
• Explainable Machine Learning
• Human Centred Interactive Systems
• Chronic Disease Management

Current research

Chris' current research focusses on the following areas:

• Chronic Disease Management: This research involves the application of AI and machine learning to understand if it can help reduce the burden of managing a chronic condition. A variety of technology exists (self-management apps, sensing devices), however, engagement is critical for its success and the ability to train AI from this data. Ongoing projects tackling type-one diabetes and chronic obstructive pulmonary disease have directly included people with the conditions, family members, and clinical care teams to ensure needs are met.

  • Computational Notebooks as Co-Design Tools: Engaging Young Adults Living with Diabetes, Family Carers, and Clinicians with Machine Learning Models

  • Explainable Machine Learning for Real-Time Hypoglycemia and Hyperglycemia Prediction and Personalized Control Recommendations

  • Monitoring wellness in Chronic Obstructive Pulmonary Disease using the myCOPD app

• Anomaly Detection and Workforce Operations: This research conducted in the Financial Services sector with industry partner OnCorps involved detecting anomalies in transactional data, and fairly allocating the generated correction tasks across a multi-skilled workforce using AI. An end-to-end pipeline was developed to detect anomalies with machine learning and allocate tasks with a genetic algorithm. 
  • Publication(s) currently under review.
• Hospital Trajectories and Outcomes: This research area involves using machine learning techniques to develop decision support tools for clinicians. Utilising mixed data including Electronic Health Records and imaging, ongoing projects aim to identify patients at high risk and in-need of further clinical review.

  • Using explainable machine learning to characterise data drift and detect emergent health risks for emergency department admissions during COVID-19

Research projects