CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Early detection of Alzheimer's disease using structural MRI: A research idea

Siavash Esmaeili Fashtakeh (Institutionen för signaler och system)
Life Science Journal-Acta Zhengzhou University Overseas Edition (1097-8135). Vol. 9 (2012), 3, p. 1072-1079.
[Artikel, refereegranskad vetenskaplig]

Alzheimer's disease (AD) is a common progressive neurodegenerative disorder that is not currently diagnosed until a patient reaches the stage of dementia. There is an urgent need to identify AD at an earlier stage, so that treatment can begin early. Structural imaging based on magnetic resonance imaging (MRI) is an integral component of the clinical assessment of patients with suspected AD. Rates of brain atrophy could be assessed in specific regions such as the hippocampus, entorhinal cortex, temporal and parietal lobes, and ventricles. Structural brain MRI is becoming increasingly used in the early diagnostics of AD. Volumetry and pattern recognition techniques for measuring cortical thinning and automated classification approaches that assess the overall pattern of atrophy seem to show promise for the early diagnosis of AD. The study is aimed at developing new pattern recognition techniques and automatic classifiers to reliably detect AD in its early stages. Data used in the preparation of this proposal is supposed to be obtained from the Alzheimer's disease neuroimaging initiative (ADNI) database. Study will begin with pre-processing of MRI images which includes correction of inhomogeneities, de-noising, registration to the stereotaxic space e. g., using a linear transform and cross normalization of the MRI intensity followed by data modulation. Brain tissue will be segmented into white matter (WM), grey matter (GM) and cerebrospinal fluid (CSF) by the SMP software. Customized tissue probability maps (TMPs) have to be created for bias correction. For feature reduction and feature selection, datasets will be inserted into a linear support vector machine (SVM). After training a model by a sub-group, cross-validation by another subgroup will be used to achieve SVM parameter optimization. We also try to develop a better classifier e. g. Neural Network for automate classification. It is expected that using structural MRI to predict AD during early stages will allow for diagnosis and treatment before irreversible neurodegeneration and functional impairment have occurred. The aim is to improve the classification accuracy that can be achieved by combining features from different structural MRI analysis techniques.

Nyckelord: Alzheimer's disease (AD), Biomarker, Structural MRI, Classification, future

Denna post skapades 2013-05-31. Senast ändrad 2013-06-28.
CPL Pubid: 177732


Läs direkt!

Lokal fulltext (fritt tillgänglig)

Institutioner (Chalmers)

Institutionen för signaler och system


Medicinsk bioteknologi

Chalmers infrastruktur