Machine learning based ECG analysis with data augmentation for fatigue status detection in elderly individuals

Q: What is the clinical evidence from this study?

Population: Fatigue (n=100). Intervention: GRU-based machine learning ECG analysis vs. LSTM, RNN, Random Forest, and Logistic Regression models. Primary outcome: Fatigue detection accuracy.

Q: What does this research mean for the field?

A GRU-based machine learning model analyzing ECG signals can detect fatigue in elderly individuals with high performance, achieving 97.96% accuracy and 0.99 AUC. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Chokri Baccouch; Chaima Bahar; Ahmed Zouinkhi; Aymen Flah; Claude Ziad El-bayeh; Shaaban M. Shaaban; Muath Odeh

doi:10.1038/s41598-026-62370-0

August 12, 2026Scientific ReportsOpen Access

Machine learning based ECG analysis with data augmentation for fatigue status detection in elderly individuals

View Full Paper

Ask AI

Bookmark

Share

Key result

GRU-based ML model analyzing ECG signals detects fatigue in the elderly with ~98% accuracy.

n=100

Why the study?

Fatigue detection in the elderly prevents health complications, but few ECG-based solutions are lightweight, interpretable, and deployable in real-world settings.

Does a portable, machine learning-based ECG analysis system accurately detect fatigue status in elderly individuals?

Key Points

To develop and evaluate a low-cost, portable electrocardiogram system using machine learning and data augmentation for real-time fatigue detection in older adults.
Acquired 6304 ECG segments from 100 elderly participants using an extended AD8232 sensor module and an ESP32 microcontroller.
Applied SMOTE and Gaussian noise data augmentation techniques to balance and enrich the signal dataset.
Trained and evaluated GRU, LSTM, and RNN models against Random Forest and Logistic Regression baselines using subject-wise data splitting.
The GRU model achieved 97.96% accuracy, an AUC of 0.99, and an F1-score of 0.98.
The GRU architecture significantly outperformed LSTM, RNN, Random Forest, and Logistic Regression baselines in classification performance.

Structured PICO

Does a portable, machine learning-based ECG analysis system accurately detect fatigue status in elderly individuals?

P

Population

100 elderly individuals who provided 6,304 ECG segments for the development and validation of machine learning models to detect fatigue.

I

Intervention

Portable, low-cost ECG-based system for real-time fatigue detection using an extended AD8232 module and ESP32 microcontroller, with data augmentation (SMOTE and Gaussian noise) and a Gated Recurrent Unit (GRU) machine learning model.

C

Comparator

Baseline machine learning models (LSTM, RNN, Random Forest, and Logistic Regression).

O

Outcome

Fatigue detection performance measured by accuracy, Area Under the Curve (AUC), and F1-score.surrogate

A low-cost, embedded ECG-based system utilizing a GRU machine learning model can accurately detect fatigue in elderly individuals, offering a potential tool for non-invasive remote monitoring.

Limitations

Small-scale non-clinical experiment conducted in a personal and informal setting
Fatigue levels were inferred using key ECG parameters rather than directly self-reported
Data collected from extended family and close neighbors without formal clinical ethical approval
Lack of formal ethical approval
Requires further clinical validation

Implication

Portable ECG fatigue detection may aid elderly monitoring; leaves open prospective validation before clinical use.

Abstract

Fatigue detection in elderly individuals is critical for preventing health complications and enhancing quality of life. Electrocardiogram (ECG) signals offer a non-invasive way to capture fatigue-related physiological changes, but few solutions are lightweight, interpretable, and deployable in real-world settings. This study proposes a portable, low-cost ECG-based system for real-time fatigue detection using an extended AD8232 module and ESP32 microcontroller. A dataset of 6304 ECG segments was collected from 100 elderly participants. Data augmentation (SMOTE and Gaussian noise) was applied. GRU, LSTM, and RNN models were trained and validated with subject-wise splitting. The GRU model achieved 97.96% accuracy, 0.99 AUC, and 0.98 F1-score, significantly outperforming LSTM, RNN, Random Forest, and Logistic Regression baselines. This work demonstrates the feasibility of low-cost, embedded ECG-based fatigue detection and provides interpretable, generalizable results suitable for elderly care applications. Further clinical validation is planned.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Baccouch et al. (2026) studied Fatigue (n=100). GRU-based machine learning ECG analysis vs. LSTM, RNN, Random Forest, and Logistic Regression models was evaluated on Fatigue detection accuracy. A GRU-based machine learning model analyzing ECG signals achieved 97.96% accuracy, 0.99 AUC, and 0.98 F1-score for detecting fatigue in elderly individuals.

synapsesocial.com/papers/6a7d76d82b0e0cff3f640915 https://doi.org/10.1038/s41598-026-62370-0

View Full Paper

Ask AI

Bookmark

Share

Machine learning based ECG analysis with data augmentation for fatigue status detection in elderly individuals

Key Points

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Machine learning based ECG analysis with data augmentation for fatigue status detection in elderly individuals

Key Points

Structured PICO

Limitations

Implication

Abstract

Discussion

Cite This Study

Keep reading

More related papers

More from Scientific Reports

Keep reading

More related papers

More from Scientific Reports