Parkinson’s Disease Detection Using Machine Learning

Abstract

Parkinson’s disease is progressive nervous system disorder that affects movement leading to shaking, stiffness, and difficulty with walking, balance, and coordination. Parkinson’s symptoms begin gradually and get worse over time. It has 5 stages to it and estimated seven to 10 million people worldwide have Parkinson disease. This is chronic and still has no cure. It is a neurodegenerative disease that affects the neurons in the brain that contain dopamine. There is a model, in which using machine learning techniques we can detect Parkinson’s disease depending upon certain medical procedures. We used support vector machine for this and use the sklearn library to prepare the dataset. This gives the accuracy of 88%. In our model we used the dataset which contains biomedical voice measurements from 31 people. From the whole data 20% is used for testing and 80% is used for training. The data of any person is entered to check whether the person has Parkinson or not. The status column is set to 0 for healthy person and 1 for person having Parkinson disease.

Introduction

I. INTRODUCTION

Parkinson’s disease is considered a neurodegenerative disease because it involves the degeneration and death of neurons. It is most frequently seen in adults over the age of 50. The most recognizable symptoms of Parkinson’s initially are movement-related and generally involve a tremor, slow movement and rigidity. The cause of Parkinson’s are not fully understood, but a combination of genetic and environmental factors are likely involved. In this model, we accurately detect the presence of Parkinson’s disease in an individual based on the huge data collected. By using machine learning algorithm, will predict the person is having Parkinson’s or not.in this model a huge amount of data is collected from the normal person and previously affected person by Parkinson’s disease.

II. RELATED WORKS

“Parkinson’s progression prediction using ml and serum cytokines” by Glenda-M Halliday and Niccholas, 25-july-2019. Serum samples from a clinic are tested to detect Parkinson’s disease, and the same samples are tested using ML algorithms for detecting Parkinson’s disease. Bhauwendraat, C., Bandres-Ciga, S., and Singleton, A.B. using voice to predict progression in Parkinson’s disease patients. Das R. “A comparison of multi classification methods for the diagnosis of Parkinson’s disease”. Methods used to test Parkinson’s disease include ML, DM neural, regression, and decision trees, with ML showing high performance.

III. PROPOSED METHODOLOGY

In this proposed methodology, we aim to develop a robust and accurate Support Vector Machine-based diagnostic tool for early detection and management of Parkinson's disease, thereby improving patient outcomes and facilitating personalized treatment strategies.

A. Data Flow Diagram

Data flow diagram show the flow of data between various elements of a system in graphical form. It also expresses the requirement of the system and shows the current system is implemented.

In the first step we will collect comprehensive dataset comprising clinical data, genetic information, and other relevant biomarkers from Parkinson’s disease patients.

 After that, preprocess the data to ensure quality and consistency, including data cleaning, missing values imputation, normalization, and standardization.

Then classify the data into Train and Test datasets, apply Support Vector Machine Classifier to train the model. The dataset will be used to predict whether the patient has Parkinson’s disease or not.

Conclusion

Parkinson’s disease is a progressive disorder that affects the nervous system and the parts of the body controlled by the nerves. Parkinson’s disease can’t be cured. Therefore it is important to diagnose it early. We used SVM classifier for early detection of Parkinson’s disease. To diagnose Parkinson disease in patients we use machine learning model. This model provides diagnostics on time, in this way it reduce treatment costs. The algorithm that is used in this ml model is better than other algorithms. This model is also beneficial for medical students like they can use this as a teaching tool. This ML model can predict Parkinson’s disease with high accuracy by the help of different data like spiral images, medical records etc.

References

[1] Arvind Kumar Tiwari, ‘”Machine Learning based Approaches for Prediction of Parkinson’s Disease,” Machine Learning and Applications- An International Journal(MLAU)vol.3, June 2016. [2] Z. A. Dastgheib, B. Lithgow, and Z. Moussavi, “Diagnosis of Parkinson’s disease using electrovestibulography,” Medical & Biological Engineering & Computing, vol. 50, no. 3, pp.365-373,2017. [3] Yatharth Nakul1 , Ankit Gupta2, Hritik Sachdeva3,,, “Parkinson Disease Detection Using Machine Learning Algorithims” International Journal of Science and Research (ISJR) ISSN: 2319-7064 SJIF (2020): 7.803 Volume 10 Issue 6, June 2021. [4] M. Abdar and M. Zomorodi-Moghadam, “impact of Patient’s Gender on Parkinson’s disease using Classification Algorithim” Journal of AI and Data Mining, vol. 6, 2018. [5] Dr. Anupam Bhatia and RaunakSulekh, “Predictive model for Parkinson’s Disease through Naïve Bayes Classification” International Journal of Computer Science & Communication vol. 9, March 2018.

Copyright

Copyright © 2024 Chandraprabha Sahu, Maya Yadav , Parth Singh Dhruwe, Arpita Singh, Samiksha Shukla. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.