Using Features Extracted From Upper Limb Reaching Tasks to Detect Parkinson’s Disease by Means of Machine Learning Models

Abstract

While in the literature there is much interest in investigating lower limbs gait of patients affected by neurological diseases, such as Parkinson's Disease (PD), fewer publications involving upper limbs movements are available. In previous studies, 24 motion signals (the socalled reaching tasks) of the upper limbs of PD patients and Healthy Controls (HCs) were used to extract several kinematic features through a custom-made software; conversely, the aim of our paper is to investigate the possibility to build models-using these features-for distinguishing PD patients from HCs. First, a binary logistic regression and, then, a Machine Learning (ML) analysis was performed by implementing five algorithms through the Knime Analytics Platform. The ML analysis was performed twice: first, a leave-one out-cross validation was applied; then, a wrapper feature selection method was implemented to identify the best subset of features that could maximize the accuracy. The binary logistic regression achieved an accuracy of 90.5%, demonstrating the importance of the maximum jerk during subjects upper limb motion; the Hosmer-Lemeshow test supported the validity of this model (p-value=0.408). The first ML analysis achieved high evaluation metrics by overcoming 95% of accuracy; the second ML analysis achieved a perfect classification with 100% of

IMNMAGN: Integrative Multimodal Approach for Enhanced Detection of Neurodegenerative Diseases Using Fusion of Multidomain Analysis With Graph Networks

“…Hao et al [9] proposed a multimodal self-paced learning approach for Alzheimer's Disease diagnosis, demonstrating the efficacy of multimodal approaches in handling complex diseases. Cesarelli et al [10] used ML models to detect Parkinson's Disease from upper limb reaching tasks, further showcasing the diversity of data that can be leveraged for disease detection. Pérez-Toro et al [11] explored user state modeling based on the arousal-valence plane, providing insights into applications in healthcare.…”

Agreement between Optoelectronic System and Wearable Sensors for the Evaluation of Gait Spatiotemporal Parameters in Progressive Supranuclear Palsy

“…In detail, the instrumented gait analysis can assess an individual’s gait patterns through the calculation of quantitative gait-related parameters, namely kinetic, kinematic, and spatiotemporal variables. In neurodegenerative diseases, gait analysis has been used for monitoring disease progression over time, quantifying parkinsonian symptoms [ 7 ], evaluating treatments’ outcomes, implementing algorithms for diagnosis through the recognition of soft signs [ 8 ], and predicting the risk of falls [ 9 , 10 , 11 , 12 ]. As for PSP, recent evidence suggests that quantitative evaluation of gait may provide further insights in both the diagnostic process and in the evaluation of disease progression [ 1 , 13 , 14 ].…”

A Viscoelastic Model to Evidence Reduced Upper-Limb-Swing Capabilities during Gait for Parkinson’s Disease-Affected Subjects

“…In a similar test, Di Biase et al [11] explored the optimal configuration of IMUs for symptom detection in upper-arm tasks, demonstrating that, to discriminate between healthy and pathological subjects and between the ON and OFF medicine conditions, a single inertial unit placed on the distal location of the upper limb is sufficient. Cesarelli et al [12] focused on motor tasks performed by the upper limbs to evidence kinematic features useful for classifying PD vs. healthy conditions through the so-called Knime Analytics Platform. Ricci et al [13][14][15] employed a network of wearable IMUs and analyzed the collected data by means of the k-nearest neighbor (kNN) and Support Vector Machine (SVM) algorithms for assessing motor impairments in a PD de novo subject, and they found that the early stages of PD are characterized by a set of features that are not visible to the naked eye.…”