Assessing sensor-derived features from a wrist-worn wearable device as indicators of upper extremity function in individuals with cervical spinal cord injury

Archives of Physical Medicine and Rehabilitation, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Language: Английский

---

Machine learning and deep learning approach to Parkinson’s disease detection: present state-of-the-art and a bibliometric review

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: 83(29), P. 72997 - 73030

Published: Feb. 10, 2024

Language: Английский

---

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

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 73095 - 73112

Published: Jan. 1, 2024

The burgeoning field of neurodegenerative disease detection and management necessitates the development robust comprehensive diagnostic approaches. Existing methodologies often fall short in effectively capturing complex interplay brain signals genetic markers, which are crucial early progression tracking such diseases. This paper introduces a novel multimodal framework that leverages advanced signal processing machine learning techniques to address these limitations, providing more accurate holistic understanding Our proposed model integrates multiple modalities: EEG analysis using Time-Frequency Analysis Wavelet Transform, functional Magnetic Resonance Imaging (fMRI) analyzed through Independent Component (ICA) Correlation Analysis, Magnetoencephalography (MEG) employing Beamforming Source Localization Techniques, Genomic Data Graph Neural Network for Genetic Pattern Recognition process. integration is realized fusion modalities Gated Recurrent Units (GRU) classification into classes via an efficient 1D Convolutional (CNN). reasons selecting methods twofold: they non-stationary characteristics exploit spatial information activity, while also identifying networks patterns associated with neurodegeneration conditions. clinical impact this work profound. Tested on BioGPS BrainLat datasets, our demonstrated 10.4% increase precision, 8.5% accuracy, 8.3% recall, 9.4% Area Under Curve (AUC), 7.5% specificity, 2.9% reduction delay compared existing methods.

Language: Английский

---

The effect of gravity on hand spatio-temporal kinematic features during functional movements

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(12), P. e0310192 - e0310192

Published: Dec. 31, 2024

Understanding the impact of gravity on daily upper-limb movements is crucial for comprehending impairments. This study investigates relationship between gravitational force and mobility by analyzing hand trajectories from 24 healthy subjects performing nine pick-and-place tasks, captured using a motion capture system. The results reveal significant differences in motor behavior terms planning, smoothness, efficiency, accuracy when are performed against or with gravity. Analysis showed that upward ( g − ) resembled transversal ones 0 but differed significantly downward + ). Corrective began later than , indicating different planning models. Velocity profiles highlighted smoother compared to . Smoothness was lower less coordinated movements. Efficiency variability no specific trends due subjective task duration among subjects. highlights importance considering effects evaluating movements, especially individuals neurological Planning metrics, including Percent Time Peak Standard Deviation, supporting Fitts’ law trade-off speed accuracy. Two novel indications were also introduced: Target Position Error Minimum Required Tunnel. These new indicators provided insights into hand-eye coordination movement variability. findings suggest efficiency influenced gravity, emphasizing need differentiated approaches assessing rehabilitating Future research should explore these metrics impaired populations develop targeted rehabilitation strategies.

Language: Английский

---

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

Sensors, Journal Year: 2023, Volume and Issue: 23(24), P. 9859 - 9859

Published: Dec. 16, 2023

The use of wearable sensors for calculating gait parameters has become increasingly popular as an alternative to optoelectronic systems, currently recognized the gold standard. objective study was evaluate agreement between Opal system and BTS SMART DX assessing spatiotemporal parameters. Fifteen subjects with progressive supranuclear palsy walked at their self-selected speed on a straight path, six were compared two measurement systems. carried out through paired data test, Passing Bablok regression, Bland-Altman Analysis. results showed perfect speed, very close cadence cycle duration, while, in other cases, either under- or over-estimated system. Some suggestions about these misalignments are proposed paper, considering that is widely used clinical context.

Language: Английский

---

Feasibility of Tree-Based Machine Learning Models to Discriminate Safe and Unsafe Posture During Weight Lifting

2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE), Journal Year: 2023, Volume and Issue: unknown, P. 870 - 875

Published: Oct. 25, 2023

The weight lifting is defined as any activity requiring the use of human force to lift or move a load which can be potentially harmful onsetting work-related musculoskeletal disorders. purpose this study was explore feasibility four tree-based Machine Learning (ML) models - fed with time-domain features extracted from signals acquired by means one inertial measurement unit (IMU) classify safe and unsafe postures during lifting. Inertial -linear acceleration angular velocity sternum 4 healthy subjects were registered using Mobility Lab System. manually segmented in order extract for each region interest, corresponding lifting, several features. Four predictive namely Decision Tree, Random Forest, Rotation Forest AdaBoost Tree implemented their performances tested. Interesting results terms evaluation metrics binary safe/unsafe posture classification obtained accuracy values greater than 93%. In conclusion present indicated that ML specific able discriminate only IMU placed on sternum. Future investigation larger cohort could confirm potential proposed methodology.

Language: Английский

---

Robotics-Based Characterization of Sensorimotor Integration in Parkinson’s Disease and the Effect of Medication

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2023, Volume and Issue: 31, P. 3201 - 3211

Published: Jan. 1, 2023

Integration of multi-modal sensory inputs and modulation motor outputs based on perceptual estimates is called Sensorimotor (SMI). Optimal functioning SMI essential for perceiving the environment, modulating outputs, learning or modifying skills to suit demands environment. Growing evidence suggests that patients diagnosed with Parkinson's Disease (PD) may suffer from an impairment in contributes deficits, leading abnormalities. However, exact nature still unclear. This study uses a robot-assisted assessment tool quantitatively characterize impairments PD how they affect voluntary movements. A set tasks was developed using robotic manipulandum equipped virtual-reality system. The conditions virtual environment were varied facilitate SMI. hundred (before after medication) forty-three control subjects completed under varying conditions. kinematic measures obtained device used evaluate findings reveal across all conditions, had 36% higher endpoint error, 38% direction error reaching tasks, 43% number violations tracing than due integrating inputs. retained ability modulate outputs. medication worsened deficits as patients, medication, performed worse before when encountering dynamic environments exhibited impaired ability.

Language: Английский

---

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

Electronics, Journal Year: 2023, Volume and Issue: 12(15), P. 3347 - 3347

Published: Aug. 4, 2023

Parkinson’s disease (PD) is a chronic neurodegenerative disorder with high worldwide prevalence that manifests muscle rigidity, tremor, postural instability, and slowness of movement. These motor symptoms are mainly evaluated by clinicians via direct observations patients and, as such, can potentially be influenced personal biases inter- intra-rater differences. In order to provide more objective assessments, researchers have been developing technology-based systems aimed at measurements symptoms, among which the reduced and/or trembling swings lower limbs during gait tests, resulting in data prone evaluations. Within this frame, although upper walking likewise important, no efforts made reveal their support significance. To fill lack, work concerns assessment forearm-swing capabilities PD respect healthy counterparts. This was obtained adopting viscoelastic model validated tests tackled an inverse dynamic problem determining torque forces acting on forearms. The results evidence differences forearm movements subjects different pathology levels, particular, we evidenced how worsening cause mechanical offered forearm’s swing process.

Language: Английский

---

A Cluster Analysis for Parkinson's Disease Phenotyping with Gait Parameters

2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE), Journal Year: 2023, Volume and Issue: unknown, P. 882 - 887

Published: Oct. 25, 2023

Gait impairment and postural instability can lead to dangerous conditions for Parkinson's disease (PD) patients. Analysis combined with current machine learning (ML) techniques may help the clinicians improve prediction of an outcome or response rehabilitation treatments. This study aims define whether a dataset gait parameters acquired in patients idiopathic PD be used identify homogeneous groups separated from each other corresponding different phenotypes. An optoelectronic motion analysis system was obtain spatial-temporal during single walking task. unsupervised ML technique, namely clustering, employed on extracted find motor-phenotypes In particular, $\boldsymbol{k}-\mathbf{means}$ clustering individuated two (Cluster 1 Custer 2) specific gait-pattern. Cluster 2 characterized by increase double support phase, stance phase duration decrease velocity, cadence, step mean cycle length. These findings suggest that abnormalities provide data-driven phenotyping, which worse motor non-motor phenotype.

Language: Английский

---

Automatic Parkinson’s Disease Diagnosis with Wearable Sensor Technology for Medical Robot

Electronics, Journal Year: 2024, Volume and Issue: 13(14), P. 2816 - 2816

Published: July 17, 2024

The clinical diagnosis of Parkinson’s disease (PD) has been the subject medical robotics research. Currently, a hot research topic is how to accurately assess severity patients and enable robots better assist in rehabilitation process. walking task on Unified Disease Rating Scale (UPDRS) well-established diagnostic criterion for PD patients. However, determined based experience neurologists, which subjective inaccurate. Therefore, this study, an automated method improved multiclass support vector machine (MCSVM) proposed wearable sensors are used. Kinematic analysis was performed extract gait features, both spatiotemporal kinematic, from installed IMU pressure sensors. Comparison experiments three different kernel functions linear trajectory were designed. experimental results show that accuracies MCSVM 92.43%, 93.45%, 95.35%. simulation trajectories closest real trajectories, shows technique performs PD.

Language: Английский

---