Acute Pain Recognition using an Ensemble Learning Methods: Evaluation of Performance and Comparison

International Research Journal of Multidisciplinary Technovation, Journal Year: 2025, Volume and Issue: unknown, P. 102 - 114

Published: Jan. 22, 2025

Accurate assessment and classification of acute pain are critical for optimal therapy, particularly in healthcare environments which early intervention might prevent chronic development. Conventional recognition approaches mostly depend on the self-reported information, can be subjective by psychological factors communication problems, especially nonverbal organizations. Recent advancements technology have provided new opportunities using facial images biomedical signals such as electromyography (EMG). In this work, we proposed an ensemble learning-based model that combines both face EMG data classification, CNN ShuffleNet V2 approach is used feature extraction. Our objective to correct intensity levels from T0 T4 (no vs. pain). We techniques like TabNet, LightGBM, Hidden Markov, Gaussian Process classification. many kinds improve prediction performance, created a comprehensive framework insights into physiological responses pain. analysis results also indicates definitely surpasses previous whereby TabNet accuracy came 97.8%. Also, has great F1 score 97.6%, well recall at 97.3%, while kappa score, it goes up 92.4%, indicating dependability. These present good optimism our learning technique could change procedures therefore patient care treatment.

Language: Английский

Pseudo-labeling based adaptations of pain domain classifiers

Frontiers in Pain Research, Journal Year: 2025, Volume and Issue: 6

Published: April 23, 2025

Each human being experiences pain differently. In addition to the highly subjective phenomenon, only limited labeled data, mostly based on short-term sequences recorded in a lab setting, is available. However, beings clinic might suffer from long painful time periods for which even smaller amount of comparison sequences, The characteristics and long-term are different with respect reactions body. an accurate assessment, representative data necessary. Although recognition techniques, reported literature, perform well sequences. collection challenging techniques assessment episodes still rare. To create systems domain knowledge transfer inevitable. this study, we adapt classifiers using pseudo-labeling techniques. We analyze recordings physiological signals combination electric thermal stimulation. results study show that it beneficial augment training set pseudo samples. For early fusion approach, improved classification performance by 2.4% 80.4% basic approach. 2.8% 70.0%

Language: Английский