학술논문

LightGBM Model based Parkinson’s Disease Detection by using Spiral Drawings
Document Type
Conference
Source
2022 Sixth International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), 2022 Sixth International Conference on. :1-5 Nov, 2022
Subject
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Robotics and Control Systems
Spirals
Shape
Writing
Prediction algorithms
Data models
Classification algorithms
Rigidity
Parkinson’s disease
Bradykinesia
Micrographia
Tremor
Postural Instability
LightGBM Classifier
Language
ISSN
2768-0673
Abstract
People with Parkinson’s disease are typically diagnosed with a long-term prognosis. It is caused by the lack of the chemical dopamine, which is responsible for regulating various activities in the body. Voice is one of the symptoms of the disease, and researchers have identified it as an underlying cause. The disease can also affect the movement of people. Some of the other symptoms of Parkinson’s include rigidity, tremor, and micrographia. This condition is often one of the first signs of the disease. People with this condition often have small, cramped handwriting. People with Parkinson’s disease may eventually start writing letters in their regular handwriting. However, they may also start writing in smaller font. This condition can also affect the hand movements of those with it. To help diagnose the disease, researchers use the drawings of patients to determine the deviance of the picture in its shape. A model that is used to diagnose Parkinson’s disease takes into account the data collected by patients and spiral drawings to come up with a diagnosis. It can then classify the person as either a healthy or a Parkinson’s patient. The LightGBM algorithm, which is a better boost compared to other methods, can help the model clearly identify the healthy and the disease-affected individuals. To use the LightGBM algorithm, you need to create a window that will allow you to paste in more than one hundred words. For the best predictions, you should create at least five paragraphs.