Scholarlink

<p>The integration of machine learning (ML) and Internet of Things (IoT) technologies has significantly advanced healthcare practices, offering innovative solutions for patient monitoring and treatment. In IoT-ML healthcare systems, the sensing layer collects patient data and transmits it via communication networks to storage layers, where ML algorithms analyze the data to support intelligent decision-making. This study proposes an automated method for detecting abrupt changes in electro-oculogram (EOG) signals within the wavelet domain. The approach integrates IoT frameworks with ML classification models, utilizing two wavelet families—Daubechies (db4) and Symlets (Sym4)—for signal processing. Key statistical parameters, combining maximum, minimum, mean, median, root mean square (RMS), standard deviation (Std), zero-crossing rate (ZCR), mean curve length (MCL), kurtosis, and skewness, are extracted to characterize the signals. Three ML models—Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Ensemble Trees (ET)—are utilized to classify EOG signals into different categories of eye movement, including up, down, right, left, no movement, and blinking. Experimental results demonstrate that the proposed method achieves superior performance compared to existing approaches, with the SVM model attaining the highest average classification accuracy of 96.2%.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

IoT-ML synergy for enhanced accessibility of the disabled using statistical Electrooculogram (EOG) signals processing

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Saly Abd-Elateif El-Gindy,
  • Ali Ahmed Khalil,
  • Fathi E. Abd El-Samie

摘要

The integration of machine learning (ML) and Internet of Things (IoT) technologies has significantly advanced healthcare practices, offering innovative solutions for patient monitoring and treatment. In IoT-ML healthcare systems, the sensing layer collects patient data and transmits it via communication networks to storage layers, where ML algorithms analyze the data to support intelligent decision-making. This study proposes an automated method for detecting abrupt changes in electro-oculogram (EOG) signals within the wavelet domain. The approach integrates IoT frameworks with ML classification models, utilizing two wavelet families—Daubechies (db4) and Symlets (Sym4)—for signal processing. Key statistical parameters, combining maximum, minimum, mean, median, root mean square (RMS), standard deviation (Std), zero-crossing rate (ZCR), mean curve length (MCL), kurtosis, and skewness, are extracted to characterize the signals. Three ML models—Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Ensemble Trees (ET)—are utilized to classify EOG signals into different categories of eye movement, including up, down, right, left, no movement, and blinking. Experimental results demonstrate that the proposed method achieves superior performance compared to existing approaches, with the SVM model attaining the highest average classification accuracy of 96.2%.