Assessing visual attributes of handwriting for prediction of neurological disorders—A case study on Parkinson’s disease
Introduction
Parkinson’s Disease (PD) is a neurodegenerative disorder that affects the coordinated movements of a person due to loss of dopamine producing neurons in substantia nigra [4]. According to studies [6], [71], it is one of the most prevalent neurological diseases after Alzheimer’s [69] with an average onset age of 60. Patients with PD experience symptoms like posture deformation, rigidity, tremors and vocal impairments, etc. Traditional diagnostic procedures for determination of the disease include costly, invasive methods like SPECT and CT scans, which are usually effective when the disease has already progressed to a mature stage. Clinical practitioners therefore, first opt for manual, non-invasive screening tests like Unified Parkinson’s Disease Rating Scale (UPDRS) [17], for early detection of the disease. While this process is quite established and has been modified over years of experience, it remains relatively subjective.
With the advent of technology, a number of computerized systems have been proposed to identify the early symptoms of Parkinson’s and similar neurological diseases. Some of these studies analyze voice or speech patterns to observe subtle but progressive changes which are indicative of PD [62], while others monitor muscular movements using wearable sensors [38]. Over the period of time, a substantial number of studies [10], [36], [63] have suggested that handwriting, a product of perceptive, cognitive and fine motor skills [3], [16], [61], can also be employed as an effective tool for early diagnosis of PD.
Currently most of the research [9], [25], [49], [65] focuses on analyzing the kinematic and pressure aspects of handwriting to determine PD. Although these dynamic features are effective, they mostly require additional temporal information for prediction which can only be acquired by utilizing special equipment like digitizer tablets and customized electronic pens [64]. Recent advancements in image analysis and pattern classification techniques have encouraged researchers to re-investigate static features from offline samples for improved detection of PD as well [70]. Hypothetically believing in the importance of visual features, we propose a novel method of assessing their contribution in characterization of PD. We designed an enhanced system that extracts useful visual features from handwriting and drawing samples of subjects and applies various fusion techniques to accurately discriminate between the control and PD groups. Feature learning based classification has previously been applied to sensor based handwriting movement signals [43] for detection of symptomatic signs of PD. Nevertheless, to the best of our knowledge, feature learning of visual attributes has not been explored to its full potential. The main contributions of the paper are listed in the following.
- •
Investigation of the effectiveness of visual attributes of handwriting by employing machine learning techniques, in characterizing PD, as opposed to the dynamic online attributes traditionally considered in the literature.
- •
Use of multiple representations of raw data as input to learn discriminative patterns from handwriting samples.
- •
Fusion of results from multiple samples acquired from various graphomotor (handwritten & hand drawn) tasks for improved overall classification.
The rest of the paper is organized as follows. Section 2 presents an overview of related works. Section 3 describes the proposed methodology and experimental setup. Section 4 summarizes the results and their analysis. Finally conclusion and future directions are discussed in Section 5.
Section snippets
Computerized analysis of handwriting for prediction of Parkinson’s disease
Computerized analysis of handwriting and hand drawn shapes has remained an active area of research in the pattern recognition community for over three decades now. Contrary to popular applications like handwriting recognition, forensic investigation and information retrieval etc., automatic analysis of handwriting and hand drawn shapes for assessment of the mental health of the subject or for prediction of different neurological disorders, still requires further exploration.
Proposed methodology and experimental setup
As mentioned in the introductory section, the objective of this study is to evaluate the effectiveness of visual attributes as discriminators between graphomotor impressions created by PD and control subjects. To achieve this objective, we propose a system consisting of multiple networks trained on samples of various graphomotor tasks. To enrich feature learning, multiple representations of input data are used to train these networks. The features learned by different networks are then combined
Analysis of results
In this section, we evaluate the performance of our proposed scheme in light of the results of the experiments conducted.
Conclusion
This study investigated the potential of visual attributes of handwriting to predict Parkinson’s disease. While the existing literature primarily targets kinematic, pressure and spatio-temporal features, we exploit the static visual attributes of handwriting extracted using Convolutional Neural Networks. The idea is not to deny the effectiveness of the rich online features but to manifest the fact that visual information in handwriting can still be effectively employed for this problem. Indeed,
References (71)
- et al.
Epidemiology of parkinson’s disease
Lancet Neurol.
(2006) - et al.
Decision support framework for parkinsons disease based on novel handwriting markers
IEEE Trans. Neural Syst. Rehabil. Eng.
(2015) - et al.
Prediction potential of different handwriting tasks for diagnosis of parkinson’s
E-Health and Bioengineering Conference (EHB), 2013
(2013) - et al.
Do children with autism and asperger’s disorder have difficulty controlling handwriting size? a kinematic evaluation
Res. Autism Spectrum Disorders
(2015) - et al.
Handwritten digit recognition with a back-propagation network
Neural Information Processing Systems (NIPS)
(1989) - et al.
The spiral axis as a clinical tool to distinguish essential tremor from dystonia cases
Parkinsonism Related Disorders
(2014) - et al.
Relearning of writing skills in parkinson’s disease: a literature review on influential factors and optimal strategies
Neuroscience Biobehav. Rev.
(2013) - et al.
A step towards the automated diagnosis of parkinson’s disease: Analyzing handwriting movements
IEEE 28th International Symposium on Computer-Based Medical Systems (CBMS), 2015
(2015) - N. Renau-Ferrer, C. Remi, A method for visuo-spatial classification of freehand shapes freely sketched,...
- et al.
Consistency of handwriting movements in dementia of the alzheimer’s type: a comparison with huntington’s and parkinson’s diseases
J. Int. Neuropsychol. Soc.
(1999)
Digitized spiral analysis is a promising early motor marker for parkinson disease
Parkinsonism Related Disorders
Pedestrian detection with convolutional neural networks
IEEE Proceedings of Intelligent Vehicles Symposium, 2005.
Going deeper with convolutions
Proceedings of IEEE Conference on Computer Vision and Pattern Recognition
Classification of handwriting patterns in patients with parkinson’s disease, using a biometric sensor
Global Adv. Res. J. Med. Med. Sci.
A novel computer-based technique for the assessment of tremor in parkinson’s disease
Age Ageing
Pathophysiology of bradykinesia in Parkinson’s disease
Brain
The Neuroscience of Handwriting: Applications for Forensic Document Examination
Prevalence and incidence of parkinson’s disease in europe
Euro. Neuropsychopharmacol.
Diagnosis and treatment of hand tremor.
J. Korean Med. Assoc./Taehan Uisa Hyophoe Chi
Micrographia secondary to lenticular lesions
Movement disorders
Convolutional neural network with data augmentation for sar target recognition
IEEE Geosci. Remote Sens. Lett.
Novel automated morphometric and kinematic handwriting assessment: A validity study in children with asd and adhd
J. Occup. Therapy, Schools Early Interven.
A new modality for quantitative evaluation of parkinson’s disease: in-air movement
IEEE 13th International Conference on Bioinformatics and Bioengineering (BIBE), 2013
Analysis of in-air movement in handwriting: A novel marker for parkinson’s disease
Comput. Methods Programs Biomed.
Evaluation of handwriting kinematics and pressure for differential diagnosis of parkinson’s disease
Artif. Intell. Med.
Computational analysis of open loop handwriting movements in parkinson’s disease: a rapid method to detect dopamimetic effects
Movement Disorders
Handwriting development, competency, and intervention
Develop. Med. Child Neurol.
Unified Parkinson’s Disease Rating Scale
A comparison of alternative tests of significance for the problem of m rankings
Ann. Math. Stat.
Children with autism show specific handwriting impairments
Neurology
Parkdetect: Early diagnosing parkinson’s disease
IEEE International Symposium on Medical Measurements and Applications (MeMeA), 2014
Going deeper into action recognition: A survey
Image Vision Comput.
Spiral analysis in niemann-pick disease type c
Movement Disorders
Shape synthesis from sketches via procedural models and convolutional networks
IEEE Trans. Vis. Comput. Graph.
Approximations of the critical region of the fbietkan statistic
Commun. Stat. Theory Methods
Cited by (79)
Parkinson's disease diagnosis using deep learning: A bibliometric analysis and literature review
2024, Ageing Research ReviewsComparison of one- two- and three-dimensional CNN models for drawing-test-based diagnostics of the Parkinson's disease
2024, Biomedical Signal Processing and ControlAutomatic diagnosis of multi-task in essential tremor: Dynamic handwriting analysis using multi-modal fusion neural network
2023, Future Generation Computer SystemsInterpol questioned documents review 2019–2022
2023, Forensic Science International: Synergy