Quantitative analysis of brain white matter lesions in multiple sclerosis subjects: Preliminary findings
Date
2008Author


Seimenis, Ioannis
Eracleous, Eleni A.


ISBN
978-1-4244-2255-5Source
5th Int. Conference on Information Technology and Applications in Biomedicine, ITAB 2008 in conjunction with 2nd Int. Symposium and Summer School on Biomedical and Health Engineering, IS3BHE 20085th International Conference on Information Technology and Applications in Biomedicine, ITAB 2008 in conjunction with 2nd International Symposium and Summer School on Biomedical and Health Engineering, IS3BHE 2008
Pages
58-61Google Scholar check
Keyword(s):
Metadata
Show full item recordAbstract
In this study the value of magnetic resonance image (MRI) shape and texture analysis was assessed in multiple sclerosis (MS) subjects, both in differentiating between normal and abnormal tissue and in assessing disease progression. Shape and texture analysis was carried out in normal and diseased lesions in transverse sections of T2-weighted magnetic resonance (MR) images acquired from 10 symptomatic untreated subjects with clinically isolated syndrome (CIS) scanned twice, with an interval of 6-12 months. All detected brain lesions were manually segmented by an experienced neurologist and confirmed by a neuro-radiologist, whilst different shape and texture features were extracted from the segmented lesions. The results showed that there was no significance difference between shape features of 0 and 6-12 months. For some texture features there was significance difference between normal tissue and MS lesions at 0 and 6-12 months and between MS lesions at 0 and 6-12 months (i.e contrast, difference variance, difference entropy, and other). Further research with more subjects is required for computing shape and texture features that may provide information for differentiating between normal tissue and MS lesions as well as for longitudinal monitoring of these lesions. In addition the proposed methodology can be used for the assessment of subjects at risk of developing future neurological events. The extracted shape and features can also offer additional information of undiagnosed lesions. ©2008 IEEE.