Sensor-based Solutions for Mental Healthcare: A Systematic Literature

Review

Nidal Drissi

, Soﬁa Ouhbi

, Jos

e Alberto Garc

ıa-Bern

, Mohammed Abdou Janati Idrissi

and Mounir Ghogho

ENSIAS, Mohammed V University in Rabat, Rabat, Morocco

Dept. of Computer Science & Software Engineering, CIT, UAE University, Al Ain, U.A.E.

Dept. of Informatics and Systems, University of Murcia, Murcia, Spain

TICLab, International University of Rabat, Rabat, Morocco

mounir.ghogho@uir.ac.ma

Keywords:

Sensors, e-Mental Health, Mental Health, Systematic Literature Review.

Abstract:

Mental well-being is a crucial aspect of the person’s general health, compromised mental health impairs the

person’s functioning, decreases the quality of life, and limits the person’s contribution to society. The mental

health industry is still facing some barriers to healthcare delivery such as costs, mental health illiteracy, and

stigma. Incorporating technological interventions in the treatment and the diagnosis processes might help

overcome these barriers. Sensors are devices that have been used for healthcare since the 1990s and have

been incorporated into mental healthcare in different forms. In this study, we conducted a systematic literature

review to identify and analyze sensor-based solutions for mental healthcare. 12 studies were identiﬁed and

analyzed. The majority of the selected studies presented methods and models and were empirically evaluated

and showed promising accuracy results. Different types of sensors were used to collect different types of data

about the patient such as physical and behavioral information. The selected studies mainly addressed the use

of sensors for common mental issues like stress and depression or the analysis of general mental status. Some

studies reported some limitations mainly related to technological issues and lack of standards.

1 INTRODUCTION

According to the World Health Organization (WHO),

mental health is “a state of well-being in which the

individual realizes his or her own abilities, can cope

with the normal stresses of life, can work produc-

tively and fruitfully, and is able to make a contribu-

tion to his or her community” (Galderisi et al., 2015).

Many mental disorders represent a life-long resource

of dysfunction and are consistently associated with in-

creased disability as they have a personal and socioe-

conomic impact (Ormel et al., 1994). Serious men-

tal illnesses have a marked impact on life expectancy

as well, which is generally higher than the impact of

other well-recognized exposures such as smoking, di-

abetes, and obesity (Chang et al., 2011). Unfortu-

nately, most people in need for mental care services,

in both developed and developing countries, receive

no services at all (Kazdin, 2019).

Many barriers to mental healthcare delivery exist,

such as cost and reimbursement; policy and legal con-

straints; limited number of providers; stigma and self-

stigma for seeking treatment; mental health illiter-

acy; and cultural and ethnic inﬂuences (Hinshaw and

Stier, 2008; Andrade et al., 2014). Another less fre-

quently discussed barrier, yet a very important one, is

the dominant model of delivering psychosocial care,

which is based on a one-to-one, in-person treatment,

with a trained mental health professional, provided in

clinical setting (Kazdin, 2019).

Incorporating information and communication

technologies (ICT) in the treatment and diagnosis pro-

cesses of mental disorders might help overcome some

of the aforementioned mental healthcare delivery bar-

riers. ICT provide several technology-enabled ther-

apies, including internet-based programs, mobile ap-

plications and informational websites (Stone and Wal-

dron, 2019; Drissi et al., 2019c; Drissi et al., 2019b).

Sensors have been used for health purposes since the

1990s, and have been successfully used for mental

healthcare purposes as well (Abdullah and Choud-

Drissi, N., Ouhbi, S., GarcÃ a-BernÃ ˛a, J., Idrissi, M. and Ghogho, M.

Sensor-based Solutions for Mental Healthcare: A Systematic Literature Review.

DOI: 10.5220/0009100805930600

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 5: HEALTHINF, pages 593-600

ISBN: 978-989-758-398-8; ISSN: 2184-4305

593

hury, 2018; Harari et al., 2017). Their use provides

different methods of data collection that can help

identify the person’s behaviors, physical indicators,

thoughts, feelings, and traits (Mohr et al., 2017), in

addition to several forms of patient’s monitoring and

treatment.

Researchers are showing interest in the use of sen-

sors for mental healthcare. Among others. (Baig

et al., 2017) conducted a systematic review of wear-

able patient monitoring systems and described chal-

lenges and opportunities. (Mukhopadhyay, 2014)

reviewed sensors-based activity monitoring systems

and described challenges and issues. (Drissi et al.,

2019a) provided a survey of literature on the use of

sensors for mental healthcare, describing addressed

topics and weak areas of the research.

2 METHODOLOGY

The present study is a systematic literature review

(SLR), based on guidelines proposed by (Keele,

2007), conducted on 12 studies proposing sensors-

based solutions for mental healthcare. The studies se-

lection and analysis processes are presented below.

2.1 Research Questions

Table 1 presents the research questions (RQs) of this

study and their motivation.

2.2 Research Strategy

The search process started by consulting the follow-

ing sources: IEEE Digital Library, ACM Digital Li-

brary, ScienceDirect, Springer Link and PubMed.

The following search string was used for the auto-

matic search of publications in the mentioned digi-

tal sources: “Sensors” AND “Mental” AND “Health”

AND “Care”. The search string was formulated to

include a broad selection of literature. In order to

determine whether to include or exclude a study, the

ﬁrst author conducted a primary analysis by inspect-

ing each paper’s title, abstract and keywords.

2.3 Papers Selection

After the identiﬁcation of relevant publications for the

study by the ﬁrst author, information about each paper

from the outcome of search were ﬁlled in an excel ﬁle.

The second author revised the ﬁnal selection. The in-

clusion criterion (IC) was limited to:

IC: Publications that address the use of sensors for

mental healthcare.

The studies that met the following exclusion criterion

(EC) were eliminated:

EC: Publications that were not solution proposals

such as, opinion papers and reviews.

2.4 Data Extraction Strategy

The data extraction process was mainly centered

around providing answers to the RQs. Data extracted

from each selected study were:

• Type of the proposed solution, which can be clas-

siﬁed as: models, methods, tool-based techniques

or others.

• Types of sensors used in the proposed solution:

how they operate and what data they collect.

• Types of data collected by the sensors.

• Limitation of the proposed solution.

• Whether the proposed solution has been empir-

ically evaluated and how. An empirical evalua-

tion method can be classiﬁed as: experiment, case

study, survey, interview, or other.

• Mental disorders and mental health related issues

addressed by the selected solution proposals.

3 RESULTS

12 studies were selected as shown in Figure 1. Re-

sults from the papers selection and data extraction

processes are presented in Table 2.

Figure 1: Selection process.

3.1 RQ1. Types of the Proposed

Solutions (Contribution Types)

The contributions of the papers were mainly methods

and models with 41,67% each, and 16.66% tool-based

techniques. This section describes each solution and

how sensors were used in it.

(Yamaguchi et al., 1998) developed an indoor

monitoring system based on infrared positioning sen-

sors and magnetic sensors to observe the person’s be-

HEALTHINF 2020 - 13th International Conference on Health Informatics

594

Table 1: Research questions and their motivations.

ID Research Question Motivation

RQ1 What is the type of the proposed solution ? To identify the different contribution types of the selected pa-

pers

RQ2 What are the types of sensors used in the proposed solutions ? To identify the different types of sensors and the most used

ones for mental healthcare purposes

RQ3 What are the types of data collected by the sensors ? To identify the different types of data that could be collected

by sensors and used for mental health care

RQ4 What are the limitations of the proposed solutions ? To identify problems and limitations associated with the use

of sensors for mental healthcare

RQ5 Are the proposed solutions empirically validated or evaluated ? To discover whether the efﬁciency of a proposed solution has

been empirically validated or evaluated

RQ6 What are the most addressed mental disorders and mental health related issues

in publications proposing sensors-based solutions for mental illnesses ?

To identify mental issues and disorders that are likely to be

managed with sensors-based solutions

Table 2: Results. Acronyms: Electroencephalography (EEG), Electrooculography (EOG), Electrocardiogram (ECG), Gal-

vanic Skin Response (GSR), Global Positioning System (GPS), Heart Rate (HR), Short Message Service (SMS).

Ref. Year RQ1 RQ2 RQ3 RQ5 RQ6

(McGinnis et al., 2018) 2018 Method Belt-worn sensor Motion: acceleration and angular

velocity

Experiment Depression,

anxiety

(Sahu and Sharma, 2016) 2016 Model EEG and EOG sensors Brainwaves and eye movement No General

mental status

(Farhan et al., 2016) 2016 Method Smartphone sensors Physical Activity, light informa-

tion, conversations, phone lock,

GPS location, audio

Experiment Depression

(Sano et al., 2015) 2015 Method Wrist sensor Skin temperature, skin conduc-

tance, activity, light exposure,

call, SMS, location, internet us-

age and “screen on” timing

Experiment General

mental sta-

tus, sleep

quality

(Sandulescu et al., 2015) 2015 Method Wristband sensor Electro-dermal activity, pulse

plethysmograph and blood

volume

Experiment Stress

(Palmius et al., 2014) 2014 Model Wrist sensor and smart-

phone sensors

Activity, sleep, ambient light lev-

els and social network activity

Experiment General

mental status

(Sano and Picard, 2013) 2013 Tool-based

technique

Wrist sensor Acceleration information and

skin conductance

Experiment Stress

(Burns et al., 2011) 2011 Model Smartphone sensors GPS location, ambient light, re-

cent calls, SMS, emails

Experiment Depression

(Wijsman et al., 2011) 2011 Method Wireless chest belt, piezo-

electric ﬁlm sensor, wire-

less hand sensor and com-

mercial gel electrodes

Electrocardiography, respira-

tion, skin conductance, and

electromyography signals

Experiment Stress

(Fletcher et al., 2010) 2010 Tool-based

technique

Electrodermal activity sen-

sors, mobile Photoplethys-

mography sensors

Skin conductance, blood volume

pulse, HR

No General

mental status

(Sun et al., 2010) 2010 Model ECG sensor, chest strap,

GSR sensor, accelerometer

HR variability, GSR and move-

ment

Experiment Stress

(Yamaguchi et al., 1998) 1998 Model Infrared positioning sen-

sors and magnetic sensors

Activity and behavior (use of

rooms, equipment, bed. etc)

Experiment Behavior

havior and activity. Results showed that it is possi-

ble to learn and measure a person’s behavior, which

can be used for behavioral prediction if analyzed cor-

rectly. (Sun et al., 2010) presented an activity-aware

mental stress detection system, where physiological

measures like heart rate variability, galvanic skin re-

sponse (GSR) and movement data were collected dur-

ing three physical activities (sitting, standing, and

walking) using an Electrocardiogram (ECG) sensor,

a chest strap, a GSR sensor, and an accelerometer.

Collected data was analyzed and classiﬁed using ma-

chine learning features to detect stress. The system

achieved a classiﬁcation accuracy of 80%.

(Fletcher et al., 2010) described a list of differ-

ent types of sensors and wireless solutions, and pre-

sented a list of opportunities and challenges associ-

ated with the use of these technologies. (Wijsman

et al., 2011) presented an approach to detect stress

Sensor-based Solutions for Mental Healthcare: A Systematic Literature Review

595

Table 3: Categories of the used sensors. Acronyms: Electroencephalography (EEG), Electrooculography (EOG), Electrocar-

diogram (ECG), Galvanic Skin Response (GSR).

Ref Category Sensors

(McGinnis et al., 2018; Sahu and Sharma,

2016; Sano et al., 2015; Sandulescu et al.,

2015; Palmius et al., 2014; Sano and Picard,

2013; Wijsman et al., 2011; Fletcher et al.,

2010; Sun et al., 2010)

Sensors attached to the

users body

EEG, ECG and EOG sensors, belt-

worn sensors, wrist sensors, chest

straps, piezoelectric ﬁlm sensors, wire-

less hand sensors, commercial gel elec-

trodes, electrodermal Activity sensors,

accelerometers, and GSR sensors

(Farhan et al., 2016; Sano et al., 2015;

Palmius et al., 2014; Sano and Picard, 2013;

Burns et al., 2011; Yamaguchi et al., 1998)

Sensors attached to objects

used by the user or objects

surrounding the user

Built-in smartphone sensors, position-

ing sensors, and magnetic sensors

using physiological indicators. Different physiologi-

cal signals were measured using different sensors, an

analysis of the collected data resulted in a set of 7

principal components that could be used for stress de-

tection. The study showed a classiﬁcation accuracy of

80% between stress and non-stress conditions.

(Burns et al., 2011) studied the feasibility and ef-

fectiveness of a mobile phone- and Internet-based in-

tervention for depression sufferers (Mobilyze!). The

study yielded an accuracy of (60% to 91%) predict-

ing categorical contextual states (e.g., location) and

a poor predictive capability for states rated on scales

(e.g., mood). The study also showed the potential ef-

ﬁcacy of these types of interventions and described a

list of lessons learned and limitations of the approach.

(Sano and Picard, 2013) presented a tool-based

technique aiming to ﬁnd physiological or behavioral

markers for stress, the technique is based on data

about the patient’s acceleration information and skin

conductance collected from a wrist-worn sensor, sur-

veys, and mobile phone usage. Correlation analysis of

the collected data showed over 75% accuracy of low

and high perceived stress recognition.

(Palmius et al., 2014) proposed a smartphone-

based system that allows the remote realtime collec-

tion of data about psychiatric patient symptoms to en-

able efﬁcient allocation of psychiatric resources. The

data collection process is based on a wrist-worn ac-

tivity and sleep tracker and an Android-based mobile

application.

(Sano et al., 2015) proposed an approach to deter-

mine academic performance, sleep quality and mental

health conditions based on data collected from wear-

able wrist sensors like skin conductance, activity and

light exposure, as well as data on smartphone usage

such as calls, SMS and “screen on” timing. Collected

data was then analyzed using feature selection and

machine learning techniques. The evaluation process

yielded classiﬁcation accuracies that ranged from 67

to 92% between the method and self-ﬁled tests by par-

ticipants.

(Sandulescu et al., 2015) presented a machine

learning approach for stress detection using wearable

physiological sensors. Physiological data were col-

lected using a wristband sensor then analyzed using

machine learning techniques to classify the state of a

patient into “stressful” or “non-stressful” situations.

Classiﬁcation results were found to be 73% to 83%

accurate.

(Farhan et al., 2016) introduced an approach to

ﬁnd behavioral features that are correlated with de-

pression. The approach is based on smartphone data

collection to collect information about the patient

such as physical activity, light information, conver-

sation, phone lock and GPS location, with the help of

built-in smartphone sensors, then applies a multi-view

bi-clustering algorithm to identify homogeneous be-

havioral groups. A validation process of the approach

yielded an accuracy of 87%.

(Sahu and Sharma, 2016) presented a system that

collects certain parameters of brainwaves and eye

movement then interprets them using signal process-

ing modules to determine the mental status of the pa-

tient. Results can be displayed, used to initiate a mes-

sage alerts or feed the data to other systems.

(McGinnis et al., 2018) proposed a new approach

of diagnosing depression and anxiety in young chil-

dren, by the use of a 90-seconds fear induction task

during which the child movement is monitored us-

ing a commercially available belt-worn sensor cou-

pled with machine learning. The method has been

empirically evaluated by an experiment that showed

an accuracy that can reach 75%.

3.2 RQ2. Types of Sensors used in the

Proposed Solutions

Table 3 presents the different types of sensors used

in each solution. Two different categories of sen-

sors were identiﬁed: sensors that operate by being at-

tached to the user’s body and sensors that are either

attached to objects used by or surrounding the user.

Some studies used more than one type of sensors.

HEALTHINF 2020 - 13th International Conference on Health Informatics

596

Table 4: Categories of the collected data. Acronyms: Galvanic Skin Response (GSR), Global Positioning System (GPS), Heart

Rate (HR), Short Message Service (SMS).

Ref Category Data

(Sahu and Sharma, 2016; Sano et al., 2015;

Sandulescu et al., 2015; Sano and Picard,

2013; Wijsman et al., 2011; Fletcher et al.,

2010; Sun et al., 2010)

Physical information Brainwaves, eye movement, skin tem-

perature, skin conductance, electro-

dermal activity, pulse plethysmography,

blood volume, sleep, electrocardiogra-

phy, respiration, HR, GSR and elec-

tromyography signals

(McGinnis et al., 2018; Farhan et al., 2016;

Sano et al., 2015; Palmius et al., 2014; Sano

and Picard, 2013; Sun et al., 2010; Yam-

aguchi et al., 1998)

Activity and behavior Acceleration, angular velocity, physical

Activity, movement, sleep and behavior

(Farhan et al., 2016; Sano et al., 2015;

Palmius et al., 2014; Sano and Picard, 2013;

Burns et al., 2011)

Smartphone usage Ambient light, phone Lock, GPS Loca-

tion, audio, calls, SMS, emails, internet

usage and “screen on” timing

3.3 RQ3. Data Collected by the Sensors

We identiﬁed three categories of the collected data:

physical information, activity and behavior, and

smartphone usage as presented in Table 4. Some so-

lutions used more than one category of data. Table 2

shows data used in each solution.

3.4 RQ4. Limitations of the Proposed

Solutions

Only 5 out of the 12 selected studies reported some

limitations to their solutions. A number of the se-

lected studies reported some similar limitations, like

connectivity problems, power consumption issues,

amount and accuracy of the collected data, and ac-

curate analysis and interpretation of the data (Burns

et al., 2011; Wijsman et al., 2011; Fletcher et al.,

2010; Sun et al., 2010). When using positioning and

magnetic sensors for activity and behavior sensing,

some limitations were reported such as the limited

functioning of the sensors under some external envi-

ronmental factors like hot air, and the inability to dis-

tinguish data in case of multiple users, which makes

the system helpful only in case of monitoring one pa-

tient (Yamaguchi et al., 1998). Adaptability limita-

tions and lack of standards were also reported (Burns

et al., 2011; Fletcher et al., 2010).

3.5 RQ5. Empirical Evaluation of the

Solutions

The majority of the selected solutions (83%) were

empirically evaluated by experiments, only 17%

(Sahu and Sharma, 2016; Fletcher et al., 2010) were

not empirically evaluated and classiﬁed as theories.

Figure 2: Addressed mental disorders and mental health re-

lated issues in the proposed solutions.

3.6 RQ6. Addressed Mental Disorders

and Mental Health Related Issues

Stress is the most addressed issue by the selected so-

lutions and depression is the most addressed disorder.

Many studies did not address a speciﬁc mental dis-

order and proposed solutions to predict the general

mental status of the user. Other studies used sensors

to observe indicators of the sanity and mental status

of the user such as behavior, academic performance,

and sleep quality. Figure 2 presents all the addressed

mental disorders and mental health related issues.

4 DISCUSSION

4.1 Types of the Proposed Solutions

The selected studies were not limited to providing IT-

related solutions on the use of sensors, but also pro-

vided new psychotherapy-related approaches for the

diagnosis, treatment, and monitoring of mental dis-

orders that are only possible by the use of sensors

such as associating behavioral traits with the mental

status (Farhan et al., 2016; Yamaguchi et al., 1998),

Sensor-based Solutions for Mental Healthcare: A Systematic Literature Review

597

predicting mental status by analyzing use of smart-

phones (Sano and Picard, 2013; Sano et al., 2015) and

detection and diagnosis of mental illnesses based on

physiological indicators (Sandulescu et al., 2015; Wi-

jsman et al., 2011; Sahu and Sharma, 2016). The use

of sensors in mental healthcare has opened a new win-

dow of opportunities not only in creating new systems

and technologies but also in creating new approaches

and methods for the diagnosis and treatment of mental

disorders. A need for publications proposing guide-

lines and recommendations on the use of sensors for

mental healthcare has been identiﬁed, which is neces-

sary to prevent errors and create effective solutions.

4.2 Sensors Types

Sensors were incorporated in different ways in the se-

lected solutions, depending on the type, sensors were

used to collect data from the patient’s body and envi-

ronment, and to detect habits and behaviors. Sensors

are convenient devices that can easily be attached to

the body, or objects from the environment, and that

are already integrated into devices like smartphones

and computers. The variety in the types of sensors

helped detect different indicators and gather different

types of information that are helpful in the diagnosis

and management of mental illnesses, and also encour-

aged researchers to ﬁnd associations between differ-

ent types of data and mental status.

4.3 Data Collected by the Sensors

The selected solutions used sensors mainly to collect

and detect three categories of data: physical informa-

tion, activity and behavior, and smartphone usage.

• Patterns of physical symptoms are associated with

certain mental illnesses, therefore can be used

to detect psychological disorders (Larson et al.,

2001).

• Activity and behavior were found to have a bidi-

rectional connection with mental disorders, as

regular physical activity have an inverse associa-

tion with depression and anxiety symptoms (King

et al., 2013), also physical activity is associated

with reduced risk of some mental disorders and

reduced degree of co-morbidity (Str

ohle et al.,

2007) and may be used as a safe and cost-effective

prevention and treatment tool (King et al., 2013),

analyzing activity and behavior can create an un-

derstanding of the mental status of the person.

On the other hand, some mental barriers like

fear were proved to have a negative impact on

the person’s physical activity and social behav-

ior (Stafford et al., 2007), changing bad behaviors

and improving physical activity can improve the

mental status.

• Analyzing smartphone usage habits can provide

an initial view of the mental status of the per-

son. Compulsive usage of smartphones is re-

lated to some psychological traits like anxiety

(Lee et al., 2014), self-control and communica-

tion issues, and general mental discomfort (Jang

and Kwag, 2015).

4.4 Limitations of the Proposed

Solutions

The use of sensors for mental healthcare offers many

opportunities, yet it faces some limitations that need

to be addressed to beneﬁt effectively from the sen-

sors technology. Most of the reported limitations were

related to technological issues which are relatively

straightforward to ﬁx once recognized, either by us-

ing a different technology or approach. Technologi-

cal issues are common and are usually overcome due

to the continuous technological development. Lack

of standards was also among the reported limitations,

which needs to be addressed by standardization or-

ganizations, as standards and guidelines are crucial

to the making and success of the solutions, more re-

search and solutions are needed in this area.

4.5 Empirical Evaluation

The only identiﬁed empirical evaluation method in the

selected studies was experiments, which was adopted

mainly for the controlled conditions and environment.

All empirically evaluated solutions followed similar

processes and had the main necessary elements of ex-

periments (Jedlitschka and Pfahl, 2005). The empir-

ical evaluation of the sensors-based solutions yielded

promising accuracy results, with a maximum reported

accuracy level of 92% (Sano et al., 2015). But the

controlled environment of experiments raises some

concerns, as most of the proposed solutions are sup-

posed to be used by the patients in real and not con-

trolled circumstances, that differ from a patient to

another and from an environment to another, which

might affect the reported accuracy.

The use of other empirical evaluation methods can

be beneﬁcial and help improve the proposed solu-

tions, for example, case studies can help detect possi-

ble issues and barriers of implementing the solutions

in real life, surveys can help gather needed data, and

interviews can be used to identify and understand pa-

tients’ and clinicians’ opinion on the solutions, which

is crucial and taking it into consideration may result

in more effective solutions.

HEALTHINF 2020 - 13th International Conference on Health Informatics

598

4.6 Addressed Mental Disorders and

Mental Health Related Issues

Depression and stress were proved to have an associa-

tion with physical indicators like heart rate and phys-

ical activity (King et al., 2013), which makes sensors

a helpful tool to manage these types of disorders and

issues, as sensors are mainly used to measure physi-

cal indicators and observe activity and behavior. The

selected solutions do not address complex disorders

such as schizophrenia, bipolar disorder or psychosis,

which might be due to the complex nature of these

disorders and the unwillingness or inability of patients

with these types of mental illnesses to use sensors-

based treatments. The use of sensors in mental health-

care creates new ways of diagnosis, treatment, and

monitoring. Future studies should address the use of

sensors for different types of mental disorders.

4.7 Limitations

This study may have some limitations, such as: (i)

missing terms from the search string might have im-

plications on the results; and (ii) other RQs might

have been relevant to extract further information from

the selected studies.

5 CONCLUSION

Sensors have been incorporated in mental healthcare

for several purposes, sensors can be used to moni-

tor the physical activity and behavior, measure phys-

ical indicators or collect daily habits and information

about the person. In this paper, we identiﬁed and an-

alyzed 12 sensors-based solutions for mental health-

care. Different types of sensors were used in the solu-

tions, to collect and manage different types of infor-

mation about the patients. The majority of the solu-

tions were designed to monitor stress, depression and

general mental status, and were empirically evaluated

and showed promising results.

ACKNOWLEDGEMENTS

This paper is part of the Startup project 1093 funded

by UAE University (2019-2021).

REFERENCES

Abdullah, S. and Choudhury, T. (2018). Sensing technolo-

gies for monitoring serious mental illnesses. IEEE

MultiMedia, 25(1):61–75.

Andrade, L., Alonso, J., Mneimneh, Z., Wells, J., Al-

Hamzawi, A., Borges, G., Bromet, E., Bruffaerts,

R., De Girolamo, G., De Graaf, R., Florescu, S.,

Gureje, O., HR, H., Hu, C., Huang, Y., Hwang, I.,

Jin, R., Karam, E., Kovess-Masfety, V., Levinson, D.,

Matschinger, H., O’Neill, S., Posada-Villa, J., Sagar,

R., Sampson, N., Sasu, C., Stein, D., Takeshima, T.,

Viana, M., Xavier, M., and Kessler, R. (2014). Barri-

ers to mental health treatment: results from the WHO

world mental health surveys. Psychological medicine,

44(6):1303–1317.

Baig, M. M., GholamHosseini, H., Moqeem, A. A., Mirza,

F., and Lind

en, M. (2017). A systematic review

of wearable patient monitoring systems–current chal-

lenges and opportunities for clinical adoption. Journal

of medical systems, 41(7):115.

Burns, M. N., Begale, M., Duffecy, J., Gergle, D., Karr,

C. J., Giangrande, E., and Mohr, D. C. (2011). Har-

nessing context sensing to develop a mobile interven-

tion for depression. Journal of Medical Internet Re-

search, 13(3):e55.

Chang, C.-K., Hayes, R. D., Perera, G., Broadbent, M. T.,

Fernandes, A. C., Lee, W. E., Hotopf, M., and Stewart,

R. (2011). Life expectancy at birth for people with

serious mental illness and other major disorders from

a secondary mental health care case register in london.

PloS one, 6(5):e19590.

Drissi, N., Ouhbi, S., Janati Idrissi, M. A., El Koutbi, M.,

and Ghogho, M. (2019a). On the use of sensors

in mental healthcare. Intelligent Environments 2019,

26:307–316.

Drissi, N., Ouhbi, S., Janati Idrissi, M. A., and Ghogho,

M. (2019b). Gamiﬁcation-based apps for PTSD:

an analysis of functionality and characteristics. In

16th ACS/IEEE International Conference on Com-

puter Systems and Applications (AICCSA), pages 1–6.

Drissi, N., Ouhbi, S., Janati Idrissi, M. A., and Ghogho, M.

(2019c). Mobile apps for Post Traumatic Stress Dis-

order. In 41st Annual International Conference of the

IEEE Engineering in Medicine and Biology Society

(EMBC), pages 4279–4282. IEEE.

Farhan, A. A., Lu, J., Bi, J., Russell, A., Wang, B., and

Bamis, A. (2016). Multi-view bi-clustering to identify

smartphone sensing features indicative of depression.

In IEEE First International Conference on Connected

Health: Applications, Systems and Engineering Tech-

nologies (CHASE), pages 264–273. IEEE.

Fletcher, R. R., Ming-Zher Poh, R., and Eydgahi, H. (2010).

Wearable sensors: opportunities and challenges for

low-cost health care. In Annual International Confer-

ence of the IEEE Engineering in Medicine and Biol-

ogy Society (EMBC). Institute of Electrical and Elec-

tronics Engineers (IEEE).

Galderisi, S., Heinz, A., Kastrup, M., Beezhold, J., and Sar-

torius, N. (2015). Toward a new deﬁnition of mental

health. World Psychiatry, 14(2):231–233.

Sensor-based Solutions for Mental Healthcare: A Systematic Literature Review

599

Harari, G. M., M

uller, S. R., Aung, M. S., and Rentfrow,

P. J. (2017). Smartphone sensing methods for study-

ing behavior in everyday life. Current opinion in be-

havioral sciences, 18:83–90.

Hinshaw, S. P. and Stier, A. (2008). Stigma as related to

mental disorders. Annu. Rev. Clin. Psychol., 4:367–

393.

Jang, H.-J. and Kwag, Y.-K. (2015). Comparison of factors

associated with smartphone over-usage: Focusing on

self-control, mental health and interrelationship. Jour-

nal of the Korea Academia-Industrial cooperation So-

ciety, 16(1):146–154.

Jedlitschka, A. and Pfahl, D. (2005). Reporting guidelines

for controlled experiments in software engineering. In

International Symposium on Empirical Software En-

gineering, 2005., pages 10–pp. IEEE.

Kazdin, A. E. (2019). Annual research review: Expand-

ing mental health services through novel models of in-

tervention delivery. Journal of Child Psychology and

Psychiatry, 60(4):455–472.

Keele, S. (2007). Guidelines for performing systematic lit-

erature reviews in software engineering. Technical re-

port, Technical report, Ver. 2.3 EBSE Technical Re-

port. EBSE.

King, W. C., Kalarchian, M. A., Steffen, K. J., Wolfe, B. M.,

Elder, K. A., and Mitchell, J. E. (2013). Associations

between physical activity and mental health among

bariatric surgical candidates. Journal of psychoso-

matic research, 74(2):161–169.

Larson, G. E., Booth-Kewley, S., Merrill, L. L., and

Stander, V. A. (2001). Physical symptoms as indi-

cators of depression and anxiety. Military medicine,

166(9):796–799.

Lee, Y.-K., Chang, C.-T., Lin, Y., and Cheng, Z.-H. (2014).

The dark side of smartphone usage: Psychological

traits, compulsive behavior and technostress. Com-

puters in human behavior, 31:373–383.

McGinnis, R. S., McGinnis, E. W., Hruschak, J., Lopez-

Duran, N. L., Fitzgerald, K., Rosenblum, K. L., and

Muzik, M. (2018). Wearable sensors and machine

learning diagnose anxiety and depression in young

children. In IEEE EMBS International Conference on

Biomedical & Health Informatics (BHI), pages 410–

413. IEEE.

Mohr, D. C., Zhang, M., and Schueller, S. M. (2017).

Personal sensing: understanding mental health using

ubiquitous sensors and machine learning. Annual re-

view of clinical psychology, 13:23–47.

Mukhopadhyay, S. C. (2014). Wearable sensors for human

activity monitoring: A review. IEEE sensors journal,

15(3):1321–1330.

Ormel, J., VonKorff, M., Ustun, T. B., Pini, S., Korten, A.,

and Oldehinkel, T. (1994). Common mental disorders

and disability across cultures: results from the who

collaborative study on psychological problems in gen-

eral health care. JAMA, 272(22):1741–1748.

Palmius, N., Osipov, M., Bilderbeck, A., Goodwin, G.,

Saunders, K., Tsanas, A., and Clifford, G. (2014). A

multi-sensor monitoring system for objective mental

health management in resource constrained environ-

ments. In Appropriate Healthcare Technologies for

Low Resource Settings (AHT). IET.

Sahu, S. and Sharma, A. (2016). Detecting brainwaves to

evaluate mental health using labview and applications.

In International Conference on Emerging Technologi-

cal Trends (ICETT), pages 1–4. IEEE.

Sandulescu, V., Andrews, S., Ellis, D., Bellotto, N., and

Mozos, O. M. (2015). Stress detection using wear-

able physiological sensors. In International Work-

Conference on the Interplay Between Natural and Ar-

tiﬁcial Computation, pages 526–532. Springer.

Sano, A., Phillips, A. J., Amy, Z. Y., McHill, A. W., Tay-

lor, S., Jaques, N., Czeisler, C. A., Klerman, E. B.,

and Picard, R. W. (2015). Recognizing academic

performance, sleep quality, stress level, and mental

health using personality traits, wearable sensors and

mobile phones. In IEEE 12th International Confer-

ence on Wearable and Implantable Body Sensor Net-

works (BSN), pages 1–6. IEEE.

Sano, A. and Picard, R. W. (2013). Stress recognition us-

ing wearable sensors and mobile phones. In Humaine

Association Conference on Affective Computing and

Intelligent Interaction, pages 671–676. IEEE.

Stafford, M., Chandola, T., and Marmot, M. (2007). As-

sociation between fear of crime and mental health

and physical functioning. American journal of pub-

lic health, 97(11):2076–2081.

Stone, L. and Waldron, R. (2019). Great expectations and e-

mental health: The role of literacy in mediating access

to mental healthcare. Aust J Gen Pract, 48(7):474–79.

Str

ohle, A., Hoeﬂer, M., Pﬁster, H., Mueller, A.-G., Hoyer,

J., Wittchen, H.-U., and Lieb, R. (2007). Physical ac-

tivity and prevalence and incidence of mental disor-

ders in adolescents and young adults. Psychological

medicine, 37(11):1657–1666.

Sun, F.-T., Kuo, C., Cheng, H.-T., Buthpitiya, S., Collins, P.,

and Griss, M. (2010). Activity-aware mental stress de-

tection using physiological sensors. In International

conference on Mobile computing, applications, and

services, pages 282–301. Springer.

Wijsman, J., Grundlehner, B., Liu, H., Hermens, H., and

Penders, J. (2011). Towards mental stress detec-

tion using wearable physiological sensors. In An-

nual International Conference of the IEEE Engineer-

ing in Medicine and Biology Society, pages 1798–

1801. IEEE.

Yamaguchi, A., Ogawa, M., Tamura, T., and Togawa, T.

(1998). Monitoring behavior in the home using posi-

tioning sensors. In 20th Annual International Confer-

ence of the IEEE Engineering in Medicine and Biol-

ogy Society. Vol. 20 Biomedical Engineering Towards

the Year 2000 and Beyond (Cat. No. 98CH36286),

volume 4, pages 1977–1979. IEEE.

HEALTHINF 2020 - 13th International Conference on Health Informatics

600