A Multiobjective Evolutionary Approach to Understanding Parkinson's Disease

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. Another problem, which this work also addresses, is the difficulty of performing differential diagnosis between neurodegenerative diseases. These often have overlapping symptoms, meaning that patients can be incorrectly diagnosed and go on to receive inappropriate treatments. For instance, PD can be confused with Alzheimer's, progressive supranuclear palsy, multi-system atrophy and corticobasal degeneration in its early stages. Whilst there is no cure for PD, the development of drugs that halt the progress of the disease is an important area of research. By contributing towards better understanding of the disease and better methods for monitoring disease progress, the proposed research will help address some of the obstacles currently faced by drug development.

The proposed research aims to develop computational tools that perform objective accurate diagnosis and prognosis of neurodegenerative diseases using novel yet inexpensive devices. The proposed work targets three application areas: early diagnosis, monitoring and prognosis, and disease understanding.

Early Diagnosis
Due to poor training in primary care, limited provision of expert secondary care and the complexity of presenting symptoms many sufferers of neurodegenerative diseases currently go undiagnosed or are misdiagnosed. Early diagnosis is important for both disease sufferers and their carers, since it enables treatment of symptoms and allows patients to plan their lives and make informed decisions about future treatment and care. With the development of drugs with the potential to halt the progress of neurodegenerative diseases, early diagnosis is likely to become even more important. Furthermore, the combination of high accuracy, the ability to discriminate between different neurological conditions, and the use of inexpensive devices would make our methods well-suited for use in a screening programme for neurodegenerative diseases.

Monitoring and Prognosis
Monitoring is an important part of managing a patient's treatment. However, it is also important for drug development, where there is a need for accurate and objective methods for characterising symptoms and their severity. Current clinical approaches to monitoring rely on human judgment, and display poor consistency both between clinicians and between patients monitored. Hence, the development of objective accurate computational tools for monitoring would have a considerable impact upon both a patient's existing treatment and the development of new treatments. Related to this is the issue of accurate prognosis, which involves predicting the likely course of a disease based upon current measurements of a patient's symptoms. This is particularly important for PD patients, where currently there is no accurate means of determining whether the disease will lead to dementia, or when this will occur.

Disease Understanding
In general, the symptoms of neurodegenerative diseases such as Parkinson's are not well understood, and this is an important factor underlying the difficulty of diagnosis, prognosis and drug development. In addition to creating tools to perform diagnosis and prognosis, the proposed research also aims to understand the basis of this decision process, both in terms of cognitive and motor symptoms and, significantly, through reference to its neural basis as captured by neuroimaging data. By sharing this information with the medical community, we aim to improve the ability of medical doctors to perform diagnosis and prognosis, increase understanding of the underlying disease processes, and inform drug development.

Economic Impact
Increased early diagnosis rates could lead to considerable economic savings, particularly from delayed admission to hospitals and care homes. This figure has been estimated at around £6000 per person for sufferers of dementia, a condition that affects 1 in 6 people over the age of 80. With the current rate at which the UK population is aging, by 2030 it is estimated that there will be 6 million people over the age of 80, suggesting savings in the region of £6 billion for dementia alone.