Citation: Mian, T.S. An
Unsupervised Neural Network
Feature Selection and 1D
Convolution Neural Network
Classification for Screening of
Parkinsonism. Diagnostics 2022, 12,
1796. https://doi.org/10.3390/
diagnostics12081796
Academic Editors: Muhammad
Fazal Ijaz and Marcin Wo´ zniak
Received: 5 June 2022
Accepted: 21 July 2022
Published: 25 July 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the author.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
diagnostics
Article
An Unsupervised Neural Network Feature Selection and 1D
Convolution Neural Network Classification for Screening
of Parkinsonism
Tariq Saeed Mian
Department of IS, College of Computer Science and Engineering, Taibah University,
Madinah Al Munawara 43353, Saudi Arabia; tmian@taibahu.edu.sa; Tel.: +966-563304330
Abstract: Parkinson’s disease (PD) is the second most common neurodegenerative disorder after
Alzheimer’s disease. It has a slow progressing neurodegenerative disorder rate. PD patients have
multiple motor and non-motor symptoms, including vocal impairment, which is one of the main
symptoms. The identification of PD based on vocal disorders is at the forefront of research. In this
paper, an experimental study is performed on an open source Kaggle PD speech dataset and novel
comparative techniques were employed to identify PD. We proposed an unsupervised autoencoder
feature selection technique, and passed the compressed features to supervised machine-learning (ML)
algorithms. We also investigated the state-of-the-art deep learning 1D convolutional neural network
(CNN-1D) for PD classification. In this study, the proposed algorithms are support vector machine,
logistic regression, random forest, naïve Bayes, and CNN-1D. The classifier performance is evaluated
in terms of accuracy score, precision, recall, and F1 score measure. The proposed 1D-CNN model
shows the highest result of 0.927%, and logistic regression shows 0.922% on the benchmark dataset in
terms of F1 measure. The major contribution of the proposed approach is that unsupervised neural
network feature selection has not previously been investigated in Parkinson’s detection. Clinicians
can use these techniques to analyze the symptoms presented by patients and, based on the results
of the above algorithms, can diagnose the disease at an early stage, which will allow for improved
future treatment and care.
Keywords: Parkinson’s disease; ML; linear discriminate analysis; dimensionality reduction; principal
component analysis; neural network; random forest; support vector machine; logistic regression
1. Introduction
Parkinson’s disease (PD) is a slow progressing neurodegenerative disorder, causing
impaired motor function with slow movements, tremors, and gait and balance disturbances.
Various non-motor symptoms are common, and include disturbed autonomic function
with orthostatic hypotension, constipation and urinary disturbances, sleep disorders, and
neuropsychiatric symptoms. The onset of PD is insidious, with peak age of onset being
55–65 years. The disease has drastic effects on the lives of millions of people all over the
world [1]. The progression rate of PD is slow; however, as it progresses, the affected person
loses control over their movement, resulting in serious issues. The main cause is the loss of
dopaminergic neurons in the substantia nigra and the decrease in the level of dopamine
in the striatum. PD clinical diagnosis is possible from the following four cardinal motor
symptoms: a tremor at rest, bradykinesia, postural instability, and rigidity. The diagnosis
of PD based on symptoms becomes possible when almost 60% of dopaminergic neurons
are already dead [2].
PD symptoms can vary from person to person due to its variety of symptoms, as
exemplified by the multiple types of possible PD tremors, such as limb rigidity, tremors
in the hands, and balance and gait problems. PD symptoms are generally classified
into two categories: motor (movement-related) and non-motor (unrelated to movement).
Diagnostics 2022, 12, 1796. https://doi.org/10.3390/diagnostics12081796 https://www.mdpi.com/journal/diagnostics