Deep Neural Network Classification method to Alzheimer?s Disease Detection

Abstract

Abstract-Early detection of Alzheimer disease (AD) is important for the management of disease. The human brainMagnetic resonance imaging (MRI) data have been used to detection of Alzheimer disease detection. The detection of AD is quite challenging and thus an automated tool to classify AD can be useful. Deep learning can make major advances in solving such problems. In this study, the longitudinal MRI data in non-demented and demented older adults data is utilized and the image processing technique was adopted for the data segmentation and attribute selection. Finally, deep neural network (DNN) classification is implemented for AD detection. The DNN 96.6% correctly identified AD and the minimum error rate obtained from a DNN. It shows the DNN will be useful for the development of improved computer aided diagnosis tool in MRI data.Keywords-Deep Neural Network, Alzheimer Disease, MRI Images, Classification I. INTRODUCTIONThe dementia is a global issue and that the effects of the future epidemic will be felt predominantly in low and middleincome countries. It was estimated that 46.8 million people worldwide were living with dementia in 2015 and this number will almost double every 20 years, to 87.88 million in 2035, 116.78 million in 2045 and 131.5 million in 2050. It has also a huge economic impact. The worldwide cost of dementia was estimated US$ 817.9 billion in the year of 2015 and it is projected the costs in 2030 will be around US$ 2 trillion [1]. Alzheimer"s disease (AD) is the most common cause of dementia associated with aging [2]; it accounts for 64 percent of all dementias [3]. AD is a growing public health problem among the elderly in developing countries, whose aging population is increasing rapidly [4]. AD is characterized by a progressive decline in cognitive function. AD is substantially increased among people aged 65 years or more, with a progressive decline in memory, thinking, language and learning capacity. AD should be differentiated from normal age-related decline in cognitive function, which is more gradual and associated with less disability. The disease often starts with mild symptoms and ends with severe brain damage. People with dementia lose their abilities at different rates [5]. The detection of AD in early and accurate is beneficial for the management of disease. Neuroimaging, such as magnetic resonance imaging (MRI) or computed tomography (CT) and with single photon emission computed tomography (SPECT) or positron emission tomography (PET) can be used to help exclude other cerebral pathology or subtypes of dementia. It may predict conversion from prodromal to Alzheimer"s disease.[6] [7]. Medical image processing and machine learning tools can help neurologists in assessing whether a subject is developing the Alzheimer disease. The image segmentation and classification is an important task in MRI data analysis for the AD detection. Image segmentation is intended to partition images into well-defined regions, where each region is a set of pixels that share the same range of intensitie...

Denouements of machine learning and multimodal diagnostic classification of Alzheimer’s disease

“…Gulhare et al [ 70 ] proposed a deep neural network (DNN) classification for studying the MR scans of AD, MCI, and HC. MR images are preprocessed and segmented to exclude the information that is not suitable for better understanding and select different features from the segmented images.…”

“…Islam and Zhang [ 69 ] implemented the deep CNN model for the classification of AD. This work could be more enhanced by transfer learning and exploring more hidden convolution layers, whereas the DNN proposed by Gulhare et al [ 70 ] using the Niblack thresholding algorithm proved to be more accurate, having an accuracy rate of 96.6%.…”

A Gender-Neutral Approach to Detect Early Alzheimer’s Disease Applying a Three-layer NN

“…Gulhare et al [13] proposed a Deep Neural Network (DNN) classification method to diagnose Alzheimer's from MRI. The resulting attributes were respectively the area of the extracted region, the perimeter, mean, standard deviation, 28 horizontal distances (D 1 , D 2 , ..., D 28 ), the height and the coordinates of the center of gravity of the region (G x , G y ).…”

Exploring of Classification Methods for Early Detection of Alzheimer’s Disease

“…It extracts specific features from pre processed image of different abnormal categories [ 18] , It's technique of decrease the original data by evaluating certain important characteristics of image. Different types of features used to diagnosis of AD like Structural features [8], wavelet based, Texture Features [9,12], volumetric feature [13], histogram based, color based, shape based, GLCM based from Brain MRI. In the classification stage it try to classify image based on their properties and assign a class label for the feature set.…”

“…The classifiers into two categorized one is Binary classifiers ,other is Multi-Class classifiers ,In Binary classifier Classification with only 2 distinct classes or with 2 pre defined possible outcomes either presence or absence of abnormalities of brain image, Multi-Class classifiers is Classification with more than two distinct classes. In the Study of AD various classification methods are use like Support Vector Machine [7,10,13], convolution neural network [11], DNN [12], ANN [14], Decision tree, Naive Bayes K-nearest neighbor, Fuzzy methods are so useful to early detection of Alzheimer's detection.…”