Skip to content

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

Notifications You must be signed in to change notification settings

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

Repository files navigation

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.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published