The application of Machine Learning for the diagnosis and monitoring of cognitive decline in Alzheimer's diseaseT

Conventional clinical assessment of cognitive decline in conditions such as Alzheimer's and Parkinson's disease is based on subjective evaluation and can therefore be unreliable. An objective approach employing "white box" machine learning may provide a more reliable means of assessing the patient's condition and also establish the efficacy of new drugs to treat these conditions in clinical trials.
The aim of this research project is to investigate new approaches to evaluating cognitive decline through the application of white box machine learning to data obtained from patients' responses to figure drawing tasks that are commonly used in routine assessment. The objectives are:
- To understand fully the clinical need for an objective assessment of cognitive decline
- To determine suitable figure drawing tasks used for assessing cognitive decline
- Establish the best means of digitizing the patients' responses to these tasks
- Extracting features from the data in preparation for subsequent application to machine learning
- To determine the appropriate machine learning representation
- To evaluate the results in relation to conventional clinical assessment
- To obtain feedback from clinical specialists with regard to the results obtained
The research methodology will follow a problem-based approach in response to the clinical need that utilises white box machine learning capable to interpret data obtained from digitized dynamic drawing activities of patients and verifying the solutions obtained.
This work is directly aligned with a number of EPSRC's research areas including: Clinical technologies and Artificial intelligence technologies.
The work will be carried out in collaboration with a number of medical centres worldwide including Leeds General Infirmary (Dr Stuart Jamieson), Maastricht Medical Centre (Dr Albert Leentjens) and Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University (Dr Shengdi Chen). Assistance will also be provided by ClearSky Medical Diagnostics Ltd., a spinout medical devices company from the University of York.