Multipopulation Whale Optimization-Based Feature Selection Algorithm and Its Application in Human Fall Detection Using Inertial Measurement Unit Sensors DOI Creative Commons
Hang Cao,

Bingshuo Yan,

Dong Lin

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(24), P. 7879 - 7879

Published: Dec. 10, 2024

Feature selection (FS) is a key process in many pattern-recognition tasks, which reduces dimensionality by eliminating redundant or irrelevant features. However, for complex high-dimensional issues, traditional FS methods cannot find the ideal feature combination. To overcome this disadvantage, paper presents multispiral whale optimization algorithm (MSWOA) selection. First, an Adaptive Multipopulation merging Strategy (AMS) presented, uses exponential variation and individual location information to divide population, thus avoiding premature aggregation of subpopulations increasing candidate subsets. Second, Double Spiral updating (DSS) devised break out search stagnations discovering new positions continuously. Last, facilitate convergence speed, Baleen neighborhood Exploitation (BES) mimics behavior tentacles proposed. The presented thoroughly compared with six state-of-the-art meta-heuristic promising WOA-based algorithms on 20 UCI datasets. Experimental results indicate that proposed method superior other well-known competitors most cases. In addition, utilized perform human fall-detection extensive real experimental further illustrate ability addressing practical problems.

Language: Английский

Application of Wearable Sensors in Parkinson’s Disease: State of the Art DOI Creative Commons
Anastasia Bougea

Journal of Sensor and Actuator Networks, Journal Year: 2025, Volume and Issue: 14(2), P. 23 - 23

Published: Feb. 20, 2025

(1) Background: Wearable sensors have emerged as a promising technology in the management of Parkinson’s disease (PD). These can provide continuous and real-time monitoring various motor non-motor symptoms PD, allowing for early detection intervention. In this paper, I review current research on application wearable focusing gait, tremor, bradykinesia, dyskinesia monitoring.(2) Methods: involved literature search that spanned 2000–2024 period included following keywords: “wearable sensors”, “Parkinson’s Disease”, “Inertial “accelerometers’’, ‘’gyroscopes’’, ‘’magnetometers”, “Smartphones”, “Smart homes”. (3) Results: Despite favorable outcomes from development inertial sensors, like gyroscopes accelerometers smartphones, is still restricted because there are no standards, harmonization, or consensus both clinical analytical validation. As result, several trials were created to compare effectiveness with conventional evaluation methods order track course enhance quality life results. (4) Conclusions: hold great promise PD likely play significant role future healthcare systems.

Language: Английский

Citations

1
Applying Wearable Sensors and Machine Learning to the Diagnostic Challenge of Distinguishing Parkinson’s Disease from Other Forms of Parkinsonism DOI Creative Commons
Rana Momtaz, Lisa Shulman, Ann L. Gruber‐Baldini

et al.

Biomedicines, Journal Year: 2025, Volume and Issue: 13(3), P. 572 - 572

Published: Feb. 25, 2025

Background/Objectives: Parkinson’s Disease (PD) and other forms of parkinsonism share motor symptoms, including tremor, bradykinesia, rigidity. The overlap in their clinical presentation creates a diagnostic challenge, as conventional methods rely heavily on expertise, which can be subjective inconsistent. This highlights the need for objective, data-driven approaches such machine learning (ML) this area. However, applying ML to datasets faces challenges imbalanced class distributions, small sample sizes non-PD parkinsonism, heterogeneity within group. Methods: study analyzed wearable sensor data from 260 PD participants 18 individuals with etiologically diverse were collected during mobility tasks using single placed lower back. We evaluated performance models distinguishing these two groups identified most informative classification. Additionally, we examined characteristics misclassified presented case studies common practice, uncertainty at patient’s initial visit changes diagnosis over time. also suggested potential steps address dataset limited models’ performance. Results: Feature importance analysis revealed Timed Up Go (TUG) task When TUG test alone, exceeded that combining all tasks, achieving balanced accuracy 78.2%, is 0.2% movement disorder experts. differences some scores between correctly falsely classified by our models. Conclusions: These findings demonstrate feasibility sensors differentiating parkinsonian disorders, addressing key its streamlining workflows.

Language: Английский

Citations

0
Sustainable AI Hardware for Advanced Healthcare Diagnostics DOI

N. Bharath,

Poonam Tiwari,

D. Lakshmi

et al.

Advances in computational intelligence and robotics book series, Journal Year: 2025, Volume and Issue: unknown, P. 267 - 310

Published: March 21, 2025

This paper focuses on making Artificial Intelligence (AI) sustainable, and particularly in the context of healthcare diagnostics. As AI revolutionizes with innovations predictive analytics, medical imaging, personalized treatment, rising energy demands these technologies emphasize need for sustainable hardware solutions. We explore evolution hardware, from early developments to modern, energy-efficient systems such as low-power chips Neural Processing Units (NPUs), which enables real-time, on-device data analysis. A major portion neuromorphic computing, an upcoming field inspired by brain's neural architecture. By leveraging Spiking Networks (SNNs), event-driven processing, synaptic plasticity, attain considerable efficiency, them well suited real-time applications like wearable monitors smart implants. The also delves into role cloud platforms parallelism minimizing carbon footprint technologies.

Language: Английский

Citations

0
Machine Learning and Statistical Analyses of Sensor Data Reveal Variability Between Repeated Trials in Parkinson’s Disease Mobility Assessments DOI Creative Commons
Rana Momtaz, Lisa Shulman, Ann L. Gruber‐Baldini

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(24), P. 8096 - 8096

Published: Dec. 19, 2024

Mobility tasks like the Timed Up and Go test (TUG), cognitive TUG (cogTUG), walking with turns provide insights into impact of Parkinson’s disease (PD) on motor control, balance, function. We assess test–retest reliability these in 262 PD participants 50 controls by evaluating machine learning models based wearable-sensor-derived measures statistical metrics. This evaluation examines total duration, subtask other quantitative across two trials. show that diagnostic accuracy for distinguishing from decreases a mean 1.8% between first second trial, suggesting task repetition may not be necessary accurate diagnosis. Although duration remains relatively consistent trials (intraclass correlation coefficient (ICC) = 0.62 to 0.95), greater variability is seen sensor-derived measures, reflected performance differences. Our findings also this differs only but among groups varying levels severity, indicating need consider population characteristics. Relying solely conventional metrics gauge mobility fail reveal nuanced variations movement.

Language: Английский

Citations

1
Multipopulation Whale Optimization-Based Feature Selection Algorithm and Its Application in Human Fall Detection Using Inertial Measurement Unit Sensors DOI Creative Commons
Hang Cao,

Bingshuo Yan,

Dong Lin

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(24), P. 7879 - 7879

Published: Dec. 10, 2024

Feature selection (FS) is a key process in many pattern-recognition tasks, which reduces dimensionality by eliminating redundant or irrelevant features. However, for complex high-dimensional issues, traditional FS methods cannot find the ideal feature combination. To overcome this disadvantage, paper presents multispiral whale optimization algorithm (MSWOA) selection. First, an Adaptive Multipopulation merging Strategy (AMS) presented, uses exponential variation and individual location information to divide population, thus avoiding premature aggregation of subpopulations increasing candidate subsets. Second, Double Spiral updating (DSS) devised break out search stagnations discovering new positions continuously. Last, facilitate convergence speed, Baleen neighborhood Exploitation (BES) mimics behavior tentacles proposed. The presented thoroughly compared with six state-of-the-art meta-heuristic promising WOA-based algorithms on 20 UCI datasets. Experimental results indicate that proposed method superior other well-known competitors most cases. In addition, utilized perform human fall-detection extensive real experimental further illustrate ability addressing practical problems.

Language: Английский

Citations

0