Use this URL to cite or link to this record in EThOS:
Title: Developing a neuroimaging model of rheumatoid arthritis related fatigue
Author: Goñi, María
ISNI:       0000 0004 7967 6563
Awarding Body: University of Aberdeen
Current Institution: University of Aberdeen
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Fatigue is a common and burdensome symptom in Rheumatoid Arthritis (RA). For the effective development of future therapies, it is of vital importance to identify brain biomarkers of fatigue. In this study, we identify brain differences between RA patients who improved and did not improve their levels of fatigue, and we compare the performance of different classifiers able to distinguish between these samples at baseline. Fifty four fatigued RA patients underwent a magnetic resonance (MR) scan at baseline and 6 months later. At 6 months we identified those whose fatigue levels improved and those for whom it did not. More than 13000 features across four data sets were assessed as potential predictors of fatigue improvement. These data sets included clinical, structural MRI (sMRI), diffusion tensor imaging (DTI) and functional connectivity MRI (fcMRI) data. A filter-wrapper approach was used for feature selection. Features selected were the input to the following classifiers: a Least Squares Linear Discriminant (LSLD), linear SVM, gaussian SVM and a polynomial SVM. The highest accuracy (86.7%) was achieved when the whole set of features from all datasets were used with a linear SVM classifier. The most meaningful brain features belonged to the fcMRI dataset. The feature with the greatest prediction power about fatigue improvement was the functional connectivity between the middle insula of the cingulo opercular network and the lateral cerebellum of the cerebellum network. The results presented in this study show that the algorithms were able to distinguish those patients likely to improve their levels of fatigue, using features from MR scans. Further applications of these methods may help precise diagnosis and could give the key to the development of new targeted therapies.
Supervisor: Basu, Neil ; Waiter, Gordon David ; Murray, Alison D. Sponsor: University of Aberdeen
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Fatigue ; Brain ; Brain mapping ; Rheumatoid arthritis ; Biochemical markers