Evaluation of handwriting kinematics and pressure for differential diagnosis of Parkinson's disease

Abstract: Objective: We present the PaHaW Parkinson's disease handwriting database, consisting of handwriting samples from Parkinson's disease (PD) patients and healthy controls. Our goal is to show that kinematic features and pressure features in handwriting can be used for the differential diagnosis of PD. Methods and Material: The database contains records from 37 PD patients and 38 healthy controls performing eight different handwriting tasks. The tasks include drawing an Archimedean spiral, repetitively writing orthographically simple syllables and words, and writing of a sentence. In addition to the conventional kinematic features related to the dynamics of handwriting, we investigated new pressure features based on the pressure exerted on the writing surface. To discriminate between PD patients and healthy subjects, three different classifiers were compared: K-nearest neighbors (K-NN), ensemble AdaBoost classifier, and support vector machines (SVM). Results: For predicting PD based on kinematic and pressure features of handwriting, the best performing model was SVM with classification accuracy of Pacc = 81.3% (sensitivity Psen = 87.4% and specificity of Pspe = 80.9%). When evaluated separately, pressure features proved to be relevant for PD diagnosis, yielding Pacc = 82.5% compared to Pacc = 75.4% using kinematic features. Conclusion: Experimental results showed that an analysis of kinematic and pressure features during handwriting can help assess subtle characteristics of handwriting and discriminate between PD patients and healthy controls.

• The paper evaluates handwriting kinematics and pressure features from the PaHaW database as biomarkers for Parkinson's Disease diagnosis.
• Pressure features alone showed 82.5% accuracy, outperforming kinematic features, with combined features reaching 81.3% accuracy using an SVM classifier.
• This research provides a foundation for developing non-invasive decision support systems for early diagnosis and monitoring of PD.

Evaluation of Handwriting Kinematics and Pressure for Differential Diagnosis of Parkinson's Disease

This paper introduces a novel approach for the differential diagnosis of Parkinson's Disease (PD) through the analysis of handwriting kinematics and pressure features. By leveraging the Parkinson's disease handwriting (PaHaW) database, the authors aim to demonstrate the potential of these features as biomarkers for PD. The study involved 37 PD patients and 38 healthy controls undertaking a series of specific handwriting tasks, including drawing an Archimedean spiral and writing simple words and sentences. The data was analyzed using three classifiers: Support Vector Machines (SVM), AdaBoost, and K-nearest neighbors (K-NN).

Methods and Results Overview

The PaHaW database comprises comprehensive trajectory and pressure data collected using a digitizing tablet. Key features extracted include the kinematic aspects of pen movements and the exerted pressure during handwriting. The analysis highlights new pressure-based features, such as the number of changes in pressure (NCP), alongside conventional kinematic features like stroke speed and velocity.

The experiment results indicate that pressure features alone offered a classification accuracy of 82.5%, outperforming kinematic features, which achieved 75.4%. The best performing model combined both feature types and employed an SVM classifier, resulting in an accuracy of 81.3%, with sensitivity and specificity of 87.4% and 80.9%, respectively.

Critical Analysis

This work substantiates the hypothesis that handwriting characteristics, specifically pressure and kinematic data, could provide a viable non-invasive diagnostic tool for PD. The robust performance of pressure features suggests an underexplored dimension of handwriting analysis that offers valuable discriminative power.

The paper makes categorical claims regarding the viability of handwriting-based diagnostics. Consequently, tasks that involve writing complete sentences or simple words exhibited greater discriminatory potential, likely due to their ability to elicit micrographic symptoms, as observed in tasks 2 and 8.

Implications and Future Directions

In practice, this research provides a foundation for developing decision support systems that could facilitate early PD diagnosis and monitoring, complementing traditionally clinical approaches. The marriage of handwriting analysis with other approaches, such as speech processing, could further enhance diagnostic accuracy.

Future research should address the study's limitations by expanding the database to include data from PD patients off-medication and those with other neurodegenerative diseases to evaluate the specificity of handwriting markers for PD. Additionally, examining the temporal evolution of handwriting features in longitudinal PD studies could illuminate their utility in tracking disease progression.

In conclusion, this paper adds to the corpus of computational neurology by proposing innovative use of digital handwriting analysis for neurological diagnostics, thereby fostering advancements in non-invasive diagnostic modalities.