Alzheimer’s and Parkinson’s disease are the most common forms of dementia that degenerate neurons in the brain cells.
This paper targets a comparative study on the performance of machine learning classifier and Neural Network techniques in neuro-degenerative data. The Neural Network algorithms gives classification accuracy ~92% with One hot Encoding Method
The dataset we are using for our project is the ADPD dataset also known as Alzheimer's and Parkinson's disease dataset.It has 1439 attributes and 199 entries.
The first attribute represents the gene names and the following attributes represent the gene experimental values of patients.
The dataset was made from a congregation of multiple gene based datasets including Common Genes Alzheimer Parkinson(CGAP), brain tumore and glioblastoma.
All these datasets were downloaded from Biolabs Data Set Repository which stores both experimental values and gene names.______________________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dense (Dense) (None, 64) 92032
leaky_re_lu (LeakyReLU) (None, 64) 0
dense_1 (Dense) (None, 128) 8320
leaky_re_lu_1 (LeakyReLU) (None, 128) 0
dense_2 (Dense) (None, 256) 33024
dropout (Dropout) (None, 256) 0
dense_3 (Dense) (None, 128) 32896
leaky_re_lu_2 (LeakyReLU) (None, 128) 0
dense_4 (Dense) (None, 64) 8256
dense_5 (Dense) (None, 3) 195
=================================================================
Total params: 174,723
Trainable params: 174,723
Non-trainable params: 0
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent