Evolutionary Data Mining of Parkinson's Disease Data
Research type
Research Study
Full title
Using Evolutionary Data Mining Techniques to Understand Parkinson's Disease
IRAS ID
172876
Contact name
Michael Lones
Contact email
Sponsor organisation
Heriot-Watt University
Duration of Study in the UK
1 years, 0 months, 0 days
Research summary
Parkinson's disease (PD) is the second most common neurodegenerative disorder, affecting around 1% of the population over the age of 60. Over the last decade, it has been increasingly recognised that PD is a neuropsychiatric disorder, leading to cognitive dysfunction in addition to motor dysfunction. Cognitive impairments have been detected at an early stage in the disease, making them an important marker for early diagnosis and treatment, and many PD patients go on to develop dementia.
However, at present cognitive aspects of the disease are poorly understood, and an accurate prognosis is unlikely. To address this, this work aims to develop computational techniques that can identify and discriminate different groups of PD patients based on the occurrence of both cognitive and motor symptoms. More generally, the work aims to better understand the cognitive and motor signals present within patient groups, providing insight into disease understanding, prognosis and differential diagnosis.
The work will make use of data mining approaches. These are a group of computational techniques that can be used to discover patterns within data. They also allow the construction of computational or mathematical models that can predict the future course of a disease. A particular focus will be on multiobjective evolutionary algorithms, a computational method (based upon a model of biological evolution) that can be used to explore and characterise multiple diverse patterns embedded within a data set.
REC name
South Central - Oxford C Research Ethics Committee
REC reference
15/SC/0365
Date of REC Opinion
8 Jun 2015
REC opinion
Favourable Opinion