MENDETEKSI PENYAKIT PARKINSON DENGAN OPENCV, COMPUTER VISION, DAN SPIRAL / WAVE TEST

Petrus Sianggian Purba
Afrizal Zein

SAINSTECH: Jurnal Penelitian dan Pengkajian Sains dan Teknologi
Vol 32 No 2 (2022): Sainstech : Jurnal Penelitian dan Pengkajian Sains dan Teknologi

ABSTRACT Parkinson's disease is the second most common neurodegenerative disease in humans after Alzheimer's disease. The disorder causes patients to experience a variety of symptoms, including intellectual and behavioral disturbances, dementia, memory loss, muscle weakness, stiffness (slow and stiff movements), and tremors. This study describes how to detect Parkinson's disease using Open CV and how geometric images can be used to detect and predict Parkinson's. We will then examine our image dataset collected from both patients with and without Parkinson's. After reviewing the dataset, I will teach you how to use the HOG image descriptor to scale the input image and then how we can train the Random Forest classifier over the extracted features. The expected result is that the system can detect and predict Parkinson's disease from a patient with an accuracy rate above 90% Keywords: Parkinson's detection, HOG, OpenCV, Deep learning.