TY - GEN
T1 - Detection of Parkinson's Disease at the Level of Motor Experiences of Daily Living Using Spiral Handwriting
AU - Arraziqi, Dwi
AU - Sardjono, Tri Arief
AU - Miawarni, Herti
AU - Purnomo, Mauridhi Hery
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11/17
Y1 - 2020/11/17
N2 - Parkinson's disease (PD) is a neurological disease that gradually worsens and affects the brain's part that functions to coordinate body movements. As a result, sufferers have difficulty regulating body movements, including when talking, walking, and writing. The diagnosis of PD patients can be analyzed through handwriting. Measurement of Parkinson's disease at the level of motor experiences of daily living uses handwriting tasks. This paper aims to evaluate various image feature extraction techniques from handwriting. Handwriting were collected from 102 subjects (51 PD and 51 healthy control (HC)) The proposed method uses feature extraction of Histogram of Gradient (HOG), Oriented FAST and Rotated BRIEF (ORB), Speed-Up Robust Feature (SURF), Scale Invariant Feature Transform (SIFT), Color Gradient Histogram (CGH) and KAZE. Classifiers based on Random Forest (RF). The analysis shows that the extraction features of HOG and RF classification have the best accuracy 0.8167.
AB - Parkinson's disease (PD) is a neurological disease that gradually worsens and affects the brain's part that functions to coordinate body movements. As a result, sufferers have difficulty regulating body movements, including when talking, walking, and writing. The diagnosis of PD patients can be analyzed through handwriting. Measurement of Parkinson's disease at the level of motor experiences of daily living uses handwriting tasks. This paper aims to evaluate various image feature extraction techniques from handwriting. Handwriting were collected from 102 subjects (51 PD and 51 healthy control (HC)) The proposed method uses feature extraction of Histogram of Gradient (HOG), Oriented FAST and Rotated BRIEF (ORB), Speed-Up Robust Feature (SURF), Scale Invariant Feature Transform (SIFT), Color Gradient Histogram (CGH) and KAZE. Classifiers based on Random Forest (RF). The analysis shows that the extraction features of HOG and RF classification have the best accuracy 0.8167.
KW - Parkinson's disease
KW - feature extraction
KW - handwriting
KW - machine learning
UR - https://www.scopus.com/pages/publications/85099650766
U2 - 10.1109/CENIM51130.2020.9297932
DO - 10.1109/CENIM51130.2020.9297932
M3 - Conference contribution
AN - SCOPUS:85099650766
T3 - CENIM 2020 - Proceeding: International Conference on Computer Engineering, Network, and Intelligent Multimedia 2020
SP - 39
EP - 46
BT - CENIM 2020 - Proceeding
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 International Conference on Computer Engineering, Network, and Intelligent Multimedia, CENIM 2020
Y2 - 17 November 2020 through 18 November 2020
ER -