Shail2512 / Early-Detection-of-PD-using-SPECT

Early Detection of Parkinson's Disease is done using SPECT Image. The binary classification of PD patient and Healthy Control is done using CNN.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Early-Detection-of-PD-using-SPECT

Early Detection of Parkinson's Disease is done using SPECT Image. The binary classification of PD patient and Healthy Control is implemented using CNN.

Dataset is taken from https://www.ppmi-info.org/

Summary:
-Pre-processing of data includes normalization and binarization.
-The most significant image is obtained at the 40th slice.
-The label is converted generated using one-hot encoding.
-Keras Sequential model is impemented which also includes BatchNormalization. MaxPooling and Dropout.

About

Early Detection of Parkinson's Disease is done using SPECT Image. The binary classification of PD patient and Healthy Control is done using CNN.


Languages

Language:Jupyter Notebook 100.0%