Depression Diagnosis and Forecast based on Mobile Phone Sensor Data

Previous studies have shown the correlation between sensor data collected from mobile phones and human depression states. Compared to the traditional self-assessment questionnaires, the passive data collected from mobile phones is easier to access and less time-consuming. In particular, passive mobi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: He, Xiangheng, Triantafyllopoulos, Andreas, Kathan, Alexander, Milling, Manuel, Yan, Tianhao, Rajamani, Srividya Tirunellai, Küster, Ludwig, Harrer, Mathias, Heber, Elena, Grossmann, Inga, Ebert, David D, Schuller, Björn W
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator He, Xiangheng
Triantafyllopoulos, Andreas
Kathan, Alexander
Milling, Manuel
Yan, Tianhao
Rajamani, Srividya Tirunellai
Küster, Ludwig
Harrer, Mathias
Heber, Elena
Grossmann, Inga
Ebert, David D
Schuller, Björn W
description Previous studies have shown the correlation between sensor data collected from mobile phones and human depression states. Compared to the traditional self-assessment questionnaires, the passive data collected from mobile phones is easier to access and less time-consuming. In particular, passive mobile phone data can be collected on a flexible time interval, thus detecting moment-by-moment psychological changes and helping achieve earlier interventions. Moreover, while previous studies mainly focused on depression diagnosis using mobile phone data, depression forecasting has not received sufficient attention. In this work, we extract four types of passive features from mobile phone data, including phone call, phone usage, user activity, and GPS features. We implement a long short-term memory (LSTM) network in a subject-independent 10-fold cross-validation setup to model both a diagnostic and a forecasting tasks. Experimental results show that the forecasting task achieves comparable results with the diagnostic task, which indicates the possibility of forecasting depression from mobile phone sensor data. Our model achieves an accuracy of 77.0 % for major depression forecasting (binary), an accuracy of 53.7 % for depression severity forecasting (5 classes), and a best RMSE score of 4.094 (PHQ-9, range from 0 to 27).
doi_str_mv 10.48550/arxiv.2205.07861
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2205_07861</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2205_07861</sourcerecordid><originalsourceid>FETCH-LOGICAL-a671-ce3c003df22d3e94b4b34b3ad5f03a0e7bf0acfd8349dbc7b0f085aa605d09f03</originalsourceid><addsrcrecordid>eNotj81qAjEUhbPpQqwP4Mq8wEzvTCbzsywzaguWFup-uElubMAmkojo23dqCwfO4nwc-BhbFpBXrZTwhPHqLnlZgsyhaetixvqBTpFScsHzweHBh-QSR2_4JkTSmM5cYSLDp_0tKHck_vEVPPFP8ilEPuAZH9mDxWOixX_P2X6z3vcv2e59-9o_7zKsmyLTJDSAMLYsjaCuUpUSU9BICwKBGmUBtTWtqDqjdKPAQisRa5AGuomZs9Xf7d1iPEX3jfE2_tqMdxvxA-9IRSs</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Depression Diagnosis and Forecast based on Mobile Phone Sensor Data</title><source>arXiv.org</source><creator>He, Xiangheng ; Triantafyllopoulos, Andreas ; Kathan, Alexander ; Milling, Manuel ; Yan, Tianhao ; Rajamani, Srividya Tirunellai ; Küster, Ludwig ; Harrer, Mathias ; Heber, Elena ; Grossmann, Inga ; Ebert, David D ; Schuller, Björn W</creator><creatorcontrib>He, Xiangheng ; Triantafyllopoulos, Andreas ; Kathan, Alexander ; Milling, Manuel ; Yan, Tianhao ; Rajamani, Srividya Tirunellai ; Küster, Ludwig ; Harrer, Mathias ; Heber, Elena ; Grossmann, Inga ; Ebert, David D ; Schuller, Björn W</creatorcontrib><description>Previous studies have shown the correlation between sensor data collected from mobile phones and human depression states. Compared to the traditional self-assessment questionnaires, the passive data collected from mobile phones is easier to access and less time-consuming. In particular, passive mobile phone data can be collected on a flexible time interval, thus detecting moment-by-moment psychological changes and helping achieve earlier interventions. Moreover, while previous studies mainly focused on depression diagnosis using mobile phone data, depression forecasting has not received sufficient attention. In this work, we extract four types of passive features from mobile phone data, including phone call, phone usage, user activity, and GPS features. We implement a long short-term memory (LSTM) network in a subject-independent 10-fold cross-validation setup to model both a diagnostic and a forecasting tasks. Experimental results show that the forecasting task achieves comparable results with the diagnostic task, which indicates the possibility of forecasting depression from mobile phone sensor data. Our model achieves an accuracy of 77.0 % for major depression forecasting (binary), an accuracy of 53.7 % for depression severity forecasting (5 classes), and a best RMSE score of 4.094 (PHQ-9, range from 0 to 27).</description><identifier>DOI: 10.48550/arxiv.2205.07861</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Learning</subject><creationdate>2022-05</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2205.07861$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2205.07861$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>He, Xiangheng</creatorcontrib><creatorcontrib>Triantafyllopoulos, Andreas</creatorcontrib><creatorcontrib>Kathan, Alexander</creatorcontrib><creatorcontrib>Milling, Manuel</creatorcontrib><creatorcontrib>Yan, Tianhao</creatorcontrib><creatorcontrib>Rajamani, Srividya Tirunellai</creatorcontrib><creatorcontrib>Küster, Ludwig</creatorcontrib><creatorcontrib>Harrer, Mathias</creatorcontrib><creatorcontrib>Heber, Elena</creatorcontrib><creatorcontrib>Grossmann, Inga</creatorcontrib><creatorcontrib>Ebert, David D</creatorcontrib><creatorcontrib>Schuller, Björn W</creatorcontrib><title>Depression Diagnosis and Forecast based on Mobile Phone Sensor Data</title><description>Previous studies have shown the correlation between sensor data collected from mobile phones and human depression states. Compared to the traditional self-assessment questionnaires, the passive data collected from mobile phones is easier to access and less time-consuming. In particular, passive mobile phone data can be collected on a flexible time interval, thus detecting moment-by-moment psychological changes and helping achieve earlier interventions. Moreover, while previous studies mainly focused on depression diagnosis using mobile phone data, depression forecasting has not received sufficient attention. In this work, we extract four types of passive features from mobile phone data, including phone call, phone usage, user activity, and GPS features. We implement a long short-term memory (LSTM) network in a subject-independent 10-fold cross-validation setup to model both a diagnostic and a forecasting tasks. Experimental results show that the forecasting task achieves comparable results with the diagnostic task, which indicates the possibility of forecasting depression from mobile phone sensor data. Our model achieves an accuracy of 77.0 % for major depression forecasting (binary), an accuracy of 53.7 % for depression severity forecasting (5 classes), and a best RMSE score of 4.094 (PHQ-9, range from 0 to 27).</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj81qAjEUhbPpQqwP4Mq8wEzvTCbzsywzaguWFup-uElubMAmkojo23dqCwfO4nwc-BhbFpBXrZTwhPHqLnlZgsyhaetixvqBTpFScsHzweHBh-QSR2_4JkTSmM5cYSLDp_0tKHck_vEVPPFP8ilEPuAZH9mDxWOixX_P2X6z3vcv2e59-9o_7zKsmyLTJDSAMLYsjaCuUpUSU9BICwKBGmUBtTWtqDqjdKPAQisRa5AGuomZs9Xf7d1iPEX3jfE2_tqMdxvxA-9IRSs</recordid><startdate>20220510</startdate><enddate>20220510</enddate><creator>He, Xiangheng</creator><creator>Triantafyllopoulos, Andreas</creator><creator>Kathan, Alexander</creator><creator>Milling, Manuel</creator><creator>Yan, Tianhao</creator><creator>Rajamani, Srividya Tirunellai</creator><creator>Küster, Ludwig</creator><creator>Harrer, Mathias</creator><creator>Heber, Elena</creator><creator>Grossmann, Inga</creator><creator>Ebert, David D</creator><creator>Schuller, Björn W</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220510</creationdate><title>Depression Diagnosis and Forecast based on Mobile Phone Sensor Data</title><author>He, Xiangheng ; Triantafyllopoulos, Andreas ; Kathan, Alexander ; Milling, Manuel ; Yan, Tianhao ; Rajamani, Srividya Tirunellai ; Küster, Ludwig ; Harrer, Mathias ; Heber, Elena ; Grossmann, Inga ; Ebert, David D ; Schuller, Björn W</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a671-ce3c003df22d3e94b4b34b3ad5f03a0e7bf0acfd8349dbc7b0f085aa605d09f03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>He, Xiangheng</creatorcontrib><creatorcontrib>Triantafyllopoulos, Andreas</creatorcontrib><creatorcontrib>Kathan, Alexander</creatorcontrib><creatorcontrib>Milling, Manuel</creatorcontrib><creatorcontrib>Yan, Tianhao</creatorcontrib><creatorcontrib>Rajamani, Srividya Tirunellai</creatorcontrib><creatorcontrib>Küster, Ludwig</creatorcontrib><creatorcontrib>Harrer, Mathias</creatorcontrib><creatorcontrib>Heber, Elena</creatorcontrib><creatorcontrib>Grossmann, Inga</creatorcontrib><creatorcontrib>Ebert, David D</creatorcontrib><creatorcontrib>Schuller, Björn W</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>He, Xiangheng</au><au>Triantafyllopoulos, Andreas</au><au>Kathan, Alexander</au><au>Milling, Manuel</au><au>Yan, Tianhao</au><au>Rajamani, Srividya Tirunellai</au><au>Küster, Ludwig</au><au>Harrer, Mathias</au><au>Heber, Elena</au><au>Grossmann, Inga</au><au>Ebert, David D</au><au>Schuller, Björn W</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Depression Diagnosis and Forecast based on Mobile Phone Sensor Data</atitle><date>2022-05-10</date><risdate>2022</risdate><abstract>Previous studies have shown the correlation between sensor data collected from mobile phones and human depression states. Compared to the traditional self-assessment questionnaires, the passive data collected from mobile phones is easier to access and less time-consuming. In particular, passive mobile phone data can be collected on a flexible time interval, thus detecting moment-by-moment psychological changes and helping achieve earlier interventions. Moreover, while previous studies mainly focused on depression diagnosis using mobile phone data, depression forecasting has not received sufficient attention. In this work, we extract four types of passive features from mobile phone data, including phone call, phone usage, user activity, and GPS features. We implement a long short-term memory (LSTM) network in a subject-independent 10-fold cross-validation setup to model both a diagnostic and a forecasting tasks. Experimental results show that the forecasting task achieves comparable results with the diagnostic task, which indicates the possibility of forecasting depression from mobile phone sensor data. Our model achieves an accuracy of 77.0 % for major depression forecasting (binary), an accuracy of 53.7 % for depression severity forecasting (5 classes), and a best RMSE score of 4.094 (PHQ-9, range from 0 to 27).</abstract><doi>10.48550/arxiv.2205.07861</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2205.07861
ispartof
issn
language eng
recordid cdi_arxiv_primary_2205_07861
source arXiv.org
subjects Computer Science - Artificial Intelligence
Computer Science - Learning
title Depression Diagnosis and Forecast based on Mobile Phone Sensor Data
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T19%3A46%3A35IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Depression%20Diagnosis%20and%20Forecast%20based%20on%20Mobile%20Phone%20Sensor%20Data&rft.au=He,%20Xiangheng&rft.date=2022-05-10&rft_id=info:doi/10.48550/arxiv.2205.07861&rft_dat=%3Carxiv_GOX%3E2205_07861%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true