@article{article_594977, title={USING DEEP LEARNING ALGORITHM TO DIAGNOSE PARKINSON DISEASE WITH HIGH ACCURACY}, journal={Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi}, volume={22}, pages={19–25}, year={2019}, DOI={10.17780/ksujes.594977}, url={https://izlik.org/JA95DA56DP}, author={Gemcı, Fahriye and Ibrıkcı, Turgay}, keywords={Binary classification,deep learning (DL),feed-forward neural network (FFNN),medical diagnosis,parkinson disease (PD)}, abstract={<p class="MsoNormal" style="margin-bottom:.0001pt;text-align:justify;line-height:normal;"> <span lang="en-us" style="font-family:’Times New Roman’, serif;" xml:lang="en-us">Early diagnosis of Parkinson’s disease, which causes
vital and permanent damage to both motor and non-motor symptoms, is very
important to prevent further deterioration of the patient condition. In the
present study, Parkinson’s Disease data set from UCI repository is classified
using deep learning architecture. The deep learning architecture in the study
is a feed-forward neural network (FFNN) which is builded by Keras of Python.
The architecture in the study composes of an input layer, two hidden layers and
softmax function with ReLu (Rectified Linear Units) as an output layer. The
deep learning architecture solves binary classification problem since PD data
set has two classes. In order to classify the PD data set, many tests were
performed by splitting the test and train data in different ratios. The PD data
set classification was succeeded with 100% accuracy using deep learning
algorithm splitting in %20 of the data as the test and the remaining as train
data in epoch 30. </span> </p> <p> </p>}