EUVIMEDEUVIMED entdecken
Crossref

Multi-Model Ensemble Learning for Sleep-Based Depression Classification

M. Amudha, Dr.V. Kejalakshmi, L Raghul, M. Sriram, K. Rahulgandhi

International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)

DOI: 10.15662/ijeetr.2026.0802119 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von M. Amudha und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Mental health disorders such as depression have become a serious global health challenge affecting millions of people worldwide. Early identification of depressive symptoms is essential in order to provide timely intervention and reduce the risk of severe psychological and physical health complications. However, traditional methods of depression diagnosis mainly rely on clinical interviews, self-reported questionnaires, and periodic medical assessments, which may not always provide continuous monitoring of an individual’s mental health condition. With the advancement of wearable technology, Internet of Things (IoT) devices, and machine learning techniques, it has become possible to monitor physiological and behavioral patterns in real time and use these patterns to identify potential indicators of mental health disorders. In this work, a wearable-based depression monitoring system is proposed that utilizes multiple sensors and machine learning algorithms to detect possible depression-related patterns from physiological and activity data. The proposed system integrates an ESP32-C3 microcontroller with sensors such as the MPU6050 motion sensor for activity monitoring, the MAX30102 sensor for heart rate and blood oxygen measurement, and the DS3231 real-time clock module for time-based activity tracking. These sensors collect continuous physiological and movement- related data from the user while the device is worn on the wrist. The collected data is transmitted through Bluetooth Low Energy (BLE) to a mobile or web-based application where the data is processed and analyzed. Data preprocessing techniques including filtering, smoothing, and feature extraction are applied to improve signal quality and remove noise from the collected sensor data. Machine learning algorithms such as Logistic Regression, Decision Tree, Random Forest, and Support Vector Machine are used to classify the user’s mental health condition based on extracted features such as sleep duration, movement patterns, and heart rate variability. The system provides classification outputs that indicate possible depression levels such as normal, mild risk, or high risk. The proposed wearable system aims to provide a low- cost, portable, and real-time mental health monitoring solution that can support early detection of depression and promote proactive healthcare management

Zitieren

APA

Amudha, M., Kejalakshmi, D., Raghul, L., Sriram, M., & Rahulgandhi, K. (2026). Multi-Model Ensemble Learning for Sleep-Based Depression Classification. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802119

Vancouver

Amudha M, Kejalakshmi D, Raghul L, Sriram M, Rahulgandhi K. Multi-Model Ensemble Learning for Sleep-Based Depression Classification. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802119

BibTeX
@article{Amudha2026,
  author = {Amudha, M. and Kejalakshmi, Dr.V. and Raghul, L and Sriram, M. and Rahulgandhi, K.},
  title = {Multi-Model Ensemble Learning for Sleep-Based Depression Classification},
  journal = {International Journal of Engineering & Extended Technologies Research},
  year = {2026},
  volume = {08},
  number = {02},
  doi = {10.15662/ijeetr.2026.0802119}
}
Kostenlos starten →