Hippocampus segmentation in magnetic resonance images of Alzheimer's patients using computer deep learning method

Abstract

Abstract

Background: Alzheimer’s disease is a progressive neurodegenerative disorder and the main cause of dementia in aging. Hippocampus is prone to the changes in the early stages of Alzheimer's disease. Detection and observation of the hippocampus changes using magnetic resonance imaging (MRI) before the onset of Alzheimer's disease leads to the faster preventive and therapeutic measures. Objective: The aim of this study was the segmentation of the hippocampus in the magnetic resonance (MR) images of the Alzheimer's patients using a computer deep learning method. Methods: In this computerized study, U-Net architecture of convolutional neural network was proposed to segment the hippocampus in the real MRI data. The MR images of the 100 and 35 patients available in Alzheimer’s disease Neuroimaging Initiative (ADNI) database was used for the train and test of the model, respectively. The performance of the proposed method was compared with the manual segmentation by measuring the similarity metrics. The hippocampus volume was measured by calculation of its area in each slice. Results: The desired segmentation achieved after 10 iterations. A Dice similarity coefficient (DSC) of 0.923 was obtained which is acceptable. The mean hippocampus volume was measured 2.48 cm3. Conclusion: The proposed approach is promising and can be extended in the prognosis of Alzheimer’s disease by the prediction of the hippocampus volume changes in the early stage of the disease.

Keywords: Hippocampus; Alzheimer’s disease; Magnetic Resonance Imaging; Deep Learning; Segmentation