Answers are grounded in this paper and cite their sources.
The technologies and services used in the healthcare industry are evolving quickly. Remote patient monitoring is a new advancement in this field that offers several benefits in an ageing global population with rising health issues. The technology has advanced to a level that patients may engage in their regular daily activities at home while still being observed using contemporary communication and sensor technologies, starting with very basic applications to monitor patients inside hospital rooms. A significant amount of important sensor data is produced by biosensors, which contribute to reducing the energy of these devices and thus decreasing their lifetime. Therefore, it is essential to remove redundant data before sending it to the fog gateway. In this paper, we propose an Energy‐efficient Two‐level Epileptic Seizure Detection Approach (ETESeDA) for remote patient monitoring in fog computing‐based IoMT networks. The ETESeDA approach works on two distinct levels. The first level operates at the edge gateway, where the ETESeDA approach applies a data reduction method based on a combination between Short‐Time Fourier Transform (STFT), Grey Level Co‐occurrence Matrix (GLCM), and Huffman encoding for extracting the useful information features from the EEG signals and compressing them before transferring them to fog gateway. The second level operates at the fog gateway, where the received features are decompressed and sent to machine learning model that has been trained to determine the condition of the patient and provide the suitable decision to medical staff. The proposed ETESeDA framework can alleviate both EEG data transmission and energy consumption with guaranteed precise and reliable epileptic seizure classification in IoMT environments. The performance evaluation using the Bonn EEG Dataset demonstrates that the proposed scheme achieves up to 100% accuracy for two‐class classification and 97%–98.3% for the three‐class classification, while maintaining superior sensitivity, specificity, precision, recall, and F1‐score performance. The scheme also attained more than 94% space and over 80% energy savings without impairing accuracy.
No takes yet. Share an insight, caveat, or question.
Alwash et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: