Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches

Faiza Guerrache; David J. Brown; Mufti Mahmud

doi:10.1007/978-3-031-68639-9_9

Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches

Faiza Guerrache
, David J. Brown
, Mufti Mahmud^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper presents a comprehensive study on classifying depressed and healthy individuals using the Depresjon dataset, which contains motor activity data collected from wearable devices. We prepared six different datasets, including raw data, normalised raw data, PCA-transformed data, and statistical features extracted from the raw data. We trained and evaluated six popular machine learning algorithms and their combinations using a 5-fold cross-validation technique. Our results demonstrate that most models achieved the highest accuracy with the normalised statistical feature dataset. Furthermore, we fine-tuned these algorithms using GridSearchCV and selected the best threshold using the ROC curve. Our findings provide valuable insights into the potential of wearable sensor data for detecting and predicting depressive episodes.

Original language	English
Title of host publication	Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers
Editors	Mufti Mahmud, Hanene Ben-Abdallah, M. Shamim Kaiser, Muhammad Raisuddin Ahmed, Ning Zhong
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	126-147
Number of pages	22
ISBN (Print)	9783031686382
DOIs	https://doi.org/10.1007/978-3-031-68639-9_9
State	Published - 2024
Externally published	Yes
Event	3rd International Conference on Applied Intelligence and Informatics, AII 2023 - Dubai, United Arab Emirates Duration: 29 Oct 2023 → 31 Oct 2023

Publication series

Name	Communications in Computer and Information Science
Volume	2065 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	3rd International Conference on Applied Intelligence and Informatics, AII 2023
Country/Territory	United Arab Emirates
City	Dubai
Period	29/10/23 → 31/10/23

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

Keywords

Depression
Depressive Episodes
Machine Learning
Motor Activity
Stress Prediction

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-3-031-68639-9_9

Cite this

Guerrache, F., Brown, D. J., & Mahmud, M. (2024). Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches. In M. Mahmud, H. Ben-Abdallah, M. S. Kaiser, M. R. Ahmed, & N. Zhong (Eds.), Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers (pp. 126-147). (Communications in Computer and Information Science; Vol. 2065 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-68639-9_9

Guerrache, Faiza ; Brown, David J. ; Mahmud, Mufti. / Classifying Depressed and Healthy Individuals Using Wearable Sensor Data : A Comparative Analysis of Classical Machine Learning Approaches. Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers. editor / Mufti Mahmud ; Hanene Ben-Abdallah ; M. Shamim Kaiser ; Muhammad Raisuddin Ahmed ; Ning Zhong. Springer Science and Business Media Deutschland GmbH, 2024. pp. 126-147 (Communications in Computer and Information Science).

@inproceedings{2e45421551234ef09df51778855626aa,

title = "Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches",

abstract = "This paper presents a comprehensive study on classifying depressed and healthy individuals using the Depresjon dataset, which contains motor activity data collected from wearable devices. We prepared six different datasets, including raw data, normalised raw data, PCA-transformed data, and statistical features extracted from the raw data. We trained and evaluated six popular machine learning algorithms and their combinations using a 5-fold cross-validation technique. Our results demonstrate that most models achieved the highest accuracy with the normalised statistical feature dataset. Furthermore, we fine-tuned these algorithms using GridSearchCV and selected the best threshold using the ROC curve. Our findings provide valuable insights into the potential of wearable sensor data for detecting and predicting depressive episodes.",

keywords = "Depression, Depressive Episodes, Machine Learning, Motor Activity, Stress Prediction",

author = "Faiza Guerrache and Brown, \{David J.\} and Mufti Mahmud",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.; 3rd International Conference on Applied Intelligence and Informatics, AII 2023 ; Conference date: 29-10-2023 Through 31-10-2023",

year = "2024",

doi = "10.1007/978-3-031-68639-9\_9",

language = "English",

isbn = "9783031686382",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "126--147",

editor = "Mufti Mahmud and Hanene Ben-Abdallah and Kaiser, \{M. Shamim\} and Ahmed, \{Muhammad Raisuddin\} and Ning Zhong",

booktitle = "Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers",

address = "Germany",

}

Guerrache, F, Brown, DJ & Mahmud, M 2024, Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches. in M Mahmud, H Ben-Abdallah, MS Kaiser, MR Ahmed & N Zhong (eds), Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers. Communications in Computer and Information Science, vol. 2065 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 126-147, 3rd International Conference on Applied Intelligence and Informatics, AII 2023, Dubai, United Arab Emirates, 29/10/23. https://doi.org/10.1007/978-3-031-68639-9_9

Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches. / Guerrache, Faiza; Brown, David J.; Mahmud, Mufti.
Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers. ed. / Mufti Mahmud; Hanene Ben-Abdallah; M. Shamim Kaiser; Muhammad Raisuddin Ahmed; Ning Zhong. Springer Science and Business Media Deutschland GmbH, 2024. p. 126-147 (Communications in Computer and Information Science; Vol. 2065 CCIS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Classifying Depressed and Healthy Individuals Using Wearable Sensor Data

T2 - 3rd International Conference on Applied Intelligence and Informatics, AII 2023

AU - Guerrache, Faiza

AU - Brown, David J.

AU - Mahmud, Mufti

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

PY - 2024

Y1 - 2024

N2 - This paper presents a comprehensive study on classifying depressed and healthy individuals using the Depresjon dataset, which contains motor activity data collected from wearable devices. We prepared six different datasets, including raw data, normalised raw data, PCA-transformed data, and statistical features extracted from the raw data. We trained and evaluated six popular machine learning algorithms and their combinations using a 5-fold cross-validation technique. Our results demonstrate that most models achieved the highest accuracy with the normalised statistical feature dataset. Furthermore, we fine-tuned these algorithms using GridSearchCV and selected the best threshold using the ROC curve. Our findings provide valuable insights into the potential of wearable sensor data for detecting and predicting depressive episodes.

AB - This paper presents a comprehensive study on classifying depressed and healthy individuals using the Depresjon dataset, which contains motor activity data collected from wearable devices. We prepared six different datasets, including raw data, normalised raw data, PCA-transformed data, and statistical features extracted from the raw data. We trained and evaluated six popular machine learning algorithms and their combinations using a 5-fold cross-validation technique. Our results demonstrate that most models achieved the highest accuracy with the normalised statistical feature dataset. Furthermore, we fine-tuned these algorithms using GridSearchCV and selected the best threshold using the ROC curve. Our findings provide valuable insights into the potential of wearable sensor data for detecting and predicting depressive episodes.

KW - Depression

KW - Depressive Episodes

KW - Machine Learning

KW - Motor Activity

KW - Stress Prediction

UR - https://www.scopus.com/pages/publications/85202301052

U2 - 10.1007/978-3-031-68639-9_9

DO - 10.1007/978-3-031-68639-9_9

M3 - Conference contribution

AN - SCOPUS:85202301052

SN - 9783031686382

T3 - Communications in Computer and Information Science

SP - 126

EP - 147

BT - Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers

A2 - Mahmud, Mufti

A2 - Ben-Abdallah, Hanene

A2 - Kaiser, M. Shamim

A2 - Ahmed, Muhammad Raisuddin

A2 - Zhong, Ning

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 29 October 2023 through 31 October 2023

ER -

Guerrache F, Brown DJ, Mahmud M. Classifying Depressed and Healthy Individuals Using Wearable Sensor Data: A Comparative Analysis of Classical Machine Learning Approaches. In Mahmud M, Ben-Abdallah H, Kaiser MS, Ahmed MR, Zhong N, editors, Applied Intelligence and Informatics - 3rd International Conference, AII 2023, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2024. p. 126-147. (Communications in Computer and Information Science). doi: 10.1007/978-3-031-68639-9_9