Internet of Health Things for Quality of Life:

Open Challenges based on a Systematic Literature Mapping

Pedro Almir M. Oliveira

1 a

, Rossana M. C. Andrade

1 b

Pedro A. Santos Neto

2 c

and Breno S. Oliveira

1 d

Group of Computer Networks, Software Engineering and Systems (GREat), Federal University of Cear

a, Cear

a, Brazil

Laboratory of Software Optimization and Testing (LOST), Federal University of Piau

ı, Piau

ı, Brazil

Keywords:

eHealth, Quality of Life, Internet of Health Things, Systematic Mapping.

Abstract:

Internet of Health Things (IoHT) related papers has produced valuable knowledge concerning applications

such as monitoring vital signs and predicting diseases. However, this knowledge is dispersed in the literature

and, to the best of our knowledge, we could not ﬁnd a recent study summarizing it. Thus, this work presents

a systematic mapping conducted to organize the challenges regarding IoHT applied to QoL. As a result, we

highlight a growing interest in developing health monitoring tools, but without many real-world validations.

The most mentioned challenges were well-known IoT challenges, security and privacy, data science, and

networks. Moreover, despite many studies discussing proposals for improving QoL, few papers sought to

measure this gain, and none addressed the semantic organization of QoL data obtained from smart objects.

Finally, it is expected for future a strengthening regarding elderly healthcare solutions, data science usage for

personalized systems; smart models to predict health problems; and QoL continuous monitoring.

1 INTRODUCTION

Quality of Life can be deﬁned as the perception of

life in a sociocultural context, concerning goals and

personal standards (WHOQoL Group, 1994). World

Health Organization (WHO) states that it is crucial

to measure the QoL because it has a close relation-

ship with the health status, and it provides valuable

data to medical practice (WHO, 1998). However, de-

spite the expressive number of initiatives to improve

the citizens’ QoL, there is still room for opportunities,

especially regarding the continuous measurement of

these data and strategies for adapting the environment

to improve the QoL (Oliveira et al., 2021).

Thus, due to the need for solutions that provide

broad access to healthcare and even more accurate

monitoring methods, the Internet of Things (IoT)

has been applied in healthcare (Islam et al., 2015).

The IoT enables interaction among physical things

through the Internet to achieve common goals. Thus,

the Internet of Health Things (IoHT) emerges from

https://orcid.org/0000-0002-3067-3076

https://orcid.org/0000-0002-0186-2994

https://orcid.org/0000-0002-1554-8445

https://orcid.org/0000-0003-0079-8799

the application of IoT in healthcare (Rodrigues et al.,

2018). As IoHT examples, there are non-invasive glu-

cose sensing, electrocardiogram monitoring, oxygen

saturation monitoring, medication management, and

elderly fall detection (Araujo et al., 2020).

The previously mentioned studies represent a tiny

snapshot of IoHT applications. Thus, considering the

studies published in this area, valuable knowledge is

spread in the IoHT literature. Due to this, many re-

views have been published to summarize it.

However, despite the high interest in this area, it

was not found a recent study that presents a compre-

hensive picture of the IoHT applied for the Quality

of Life. Some reviews mention the QoL term, but

they are not focused on this topic. Also, the works

published in this area bring a common idea that tech-

nologies such as IoT or Machine Learning improve

the users’ Quality of Life. Nevertheless, few studies

present measures that corroborate this idea or strate-

gies to monitor the QoL variation over time.

Therefore, in this work, we conducted a Sys-

tematic Literature Mapping (SLM) to summarize the

challenges in this area. Our main contribution is to

present a systematic map of the IoHT literature ap-

plied on Quality of Life, grouping the results and

identifying gaps that can be further investigated.

Oliveira, P., Andrade, R., Neto, P. and Oliveira, B.

Internet of Health Things for Quality of Life: Open Challenges based on a Systematic Literature Mapping.

DOI: 10.5220/0010812400003123

In Proceedings of the 15th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2022) - Volume 5: HEALTHINF, pages 397-404

ISBN: 978-989-758-552-4; ISSN: 2184-4305

397

2 MATERIAL AND METHODS

The process of this study follows the guidelines pro-

posed by (Petersen et al., 2008). In addition, to sup-

port the replicability and expansion of this study, all

artifacts and raw data can be publicly accessed (see

our Data Availability section).

2.1 Research Questions

This review’s need relies on the increasing interest in

IoHT and the absence of studies focused on the sys-

tematization of IoHT applied in Quality of Life. Thus,

three RQ were deﬁned: RQ1 What is the context of

the papers published in the QoL-related IoHT litera-

ture? RQ2 What are the challenges related to IoHT

for QoL? RQ3 What is the evidence that IoHT can

monitor and improve people’s Quality of Life?

2.2 Search Strategy

Four scientiﬁc databases were chosen based on their

representativeness: Scopus, Web of Science (WoS),

Compendex, and PubMed.

This work applied the PICO methodology (Pai

et al., 2004) to elaborate the search string due to its

wide acceptance in the literature. According to PICO,

the search string must be composed of four parts:

Population, Intervention, Comparison, and Outcome.

Then, the Population was described as “IoT papers”,

the Intervention as “Quality of Life”, and the Out-

come was deﬁned as “Challenges”.

2.3 Eligibility Criteria

This work included papers that discuss IoHT solu-

tions, challenges, or open questions focused on Qual-

ity of Life. In contrast, we excluded papers that: i)

do not present an IoHT system focused on Quality of

Life; ii) be available only in the form of abstract; iii)

do not be written in English; iv) be a short paper; v)

do not be available on the Web; vi) not published in a

workshop, conference, or journal.

2.4 Study Selection

Two researchers performed the study selection in

three steps. The major concern here is to mitigate the

researcher’s bias during this process (Wohlin et al.,

2012). Thus, it was executed the following steps:

1. Read the Title and Abstract: the studies were

selected by the reading of titles and abstracts. In

this step, an agreement analysis was conducted

considering 10% of the papers. This agreement is

essential to check the protocol’s consistency. The

agreement level achieved was considered good

with a Kappa value of 0.8.

2. Full Reading: a full reading step was conducted

to verify the relevance to answer our questions.

3. Final Review: responsible for resolving doubts

and reviewing the selection process.

2.5 Data Extraction

Data extraction included many ﬁelds, such as goals,

research questions, and challenges. Also, topic mod-

eling was performed to cluster the studies.

Regarding topic modeling, it is widely used to get

insights into textual data. In this work, we adopted

one of the most used algorithms called LDA (Blei

et al., 2003), which was implemented using the de-

fault parameters and English stopwords provided by

the Scikit-learn and using as input the paper’s title and

abstract.

2.6 Synthesis Strategy

After Data Extraction, synthesis was performed to

build the systematic map. This phase includes i) a

classiﬁcation analysis considering the primary data

and the ﬁelds with closed options, and ii) a summa-

rization of the textual attributes using manual content

analysis.

3 RESULTS AND DISCUSSION

Initially, it was recovered 378 papers in June 2020.

After duplicates removal, 187 studies remained.

Then, 55 were removed while reading titles and ab-

stracts, and another 20 were excluded because they

were only available as abstract (1), were short papers

(13), and were not available for download (6). Fi-

nally, 18 were excluded for not meeting the inclusion

criteria. Thus, 94 relevant works were selected.

In this section, we present our results, discuss their

implications, and answer our research questions. All

data extracted for this research and data visualizations

can be publicly available (see Data Availability). It is

also important to highlight that the papers selected in

this mapping were referenced using a numerical cita-

tion due to space restrictions. In our repository, there

is a table with all the 94 selected papers.

3.1 The Context of the Papers

The papers’ context is one of the most relevant aspects

to understand a research area. However, this context

HEALTHINF 2022 - 15th International Conference on Health Informatics

398

can be complex because it is possible to involve many

aspects. In order to provide greater detail without in-

creasing complexity, this work considered eight (8)

aspects for the context:

• Aspects obtained directly from scientiﬁc bases:

– year of publication

– venue

• Aspects built using clustering algorithms:

– papers’ hot topics

• Aspects that consider published taxonomies:

– research type (Wieringa et al., 2006)

– contribution (Breivold, 2017)

– empirical validation (Wohlin et al., 2012)

– QoL data (WHOQoL Group, 1994)

• Aspects extracted as open ﬁelds:

– health issue addressed

Together, these aspects assist in building a robust

answer for RQ1. We argue that without this compre-

hensive analysis, it would not be possible to observe

research gaps. In addition, these aspects enable us

to analyze the temporal distribution of the papers, in

which venue they are being published, which topics

are most discussed, the research type and proposed

contributions, the validation strategies, the focus re-

garding QoL domains and facets, and what health is-

sues are being addressed.

Figure 1: The distribution of papers over the years.

Regarding the temporal distribution, Figure 1

presents a graph with the number of works published

over the years. It is possible to observe an increase

in the number of publications except for 2020. How-

ever, this low number in 2020 is due to the period in

which the search was performed. It is highly proba-

ble (given the analysis of the works already accepted

for publication but which are not yet available in the

databases) that the number of papers in 2020 will ex-

ceed the other years. Moreover, concerning the venue,

55% (52/94) of the studies were published in journals.

Concerning the hot topics, nine clusters were pro-

posed based on the LDA results. Figure 2.A shows the

distribution of papers by each cluster. In this graphic,

it is possible to observe a high interest in IoT Health-

care Services (38%), followed by Elderly Healthcare

(17%), Big Data (11%), Sensors and Wearable (9%).

The Security and Privacy, Network and Communica-

tion, and Health Activity Monitoring clusters have six

studies each, and the last two categories are Health

Prediction (3%) and Well-being and Comfort (1%).

These results indicate an interest in IoT services

for healthcare, activity monitoring, and disease pre-

diction with a special focus on the elderly. Other re-

search areas have also been strengthened to support

the development of these services, such as Big Data,

Sensors and Wearable, Security and Privacy, and Net-

work and Communication. Another interesting point

to highlight is that although Machine Learning and

Cloud Computing did not appear as topics, they are

fundamental in IoHT. Many works mention the use of

Machine Learning techniques (Study IDs: 78, 89) and

cloud capabilities (Study IDs: 14, 48).

As regards the research type, we used the taxon-

omy proposed by (Wieringa et al., 2006) that has

six categories: solution proposal; validation research;

evaluation research; experience papers; and, opin-

ion papers. In Figure 2.B, it is possible to observe

the number of papers for each research type. Most

works were classiﬁed as solution proposal (29), fol-

lowed by evaluation research (21), validation research

(17), conceptual proposal (9), experience paper (5),

and two studies were classiﬁed as opinion papers. For

11 papers, this aspect was not clearly identiﬁed.

The contribution also has six categories accord-

ing to (Breivold, 2017): method; model; tool; formal

study; experience; and others for those that do not

ﬁt in any of these categories. Usually, this last cat-

egory encompasses secondary studies and papers fo-

cused on discussing challenges. This mapping found

25 tools, 21 models, 13 methods, 7 formal studies,

and 3 experiences (Figure 2.C).

To deepen the analysis of the contributions, we

decided to classify them into only monitoring, or

monitoring and acting. In this way, 47 papers pro-

posed only monitoring solutions, and 6 studies pre-

sented monitoring and acting solutions. For 41 stud-

ies, this aspect was not clearly identiﬁed. Thus, this

result reinforces the discussion made by (Al-Fuqaha

et al., 2015) that there are many information ag-

gregation services, and the IoT needs to evolve for

more collaborative-aware services and ubiquitous ser-

vices. Also, the low number of collaborative-aware

and ubiquitous services can be seen as a gap for the

Internet of Health Things area.

Regarding the empirical validation, we selected

the taxonomy proposed by (Wohlin et al., 2012),

including usability evaluation, proof-of-concept

(PoCs), and simulation. These three last categories

were included to expand our classiﬁcation scheme.

Internet of Health Things for Quality of Life: Open Challenges based on a Systematic Literature Mapping

399

Figure 2: Number of papers considering (A) the proposed clustering, (B) research type, and (C) contribution type.

Thus, only 68 papers (72%) present a well-described

empirical validation. There are 23 case studies, 19

PoCs, 13 experiments, 8 surveys, and 5 simulations.

This result directly correlates with the research type

since case studies usually evaluate a solution in

practice. Moreover, two papers used case studies to

present lessons learned. Hence, considering that 28%

of works did not present a strong validation and that

54,4% were evaluated in controlled environments

with PoCs, experiments, and simulations, there is

a large room of challenges for partnerships with

industry to conduct practical validations.

Concerning QoL data (domains and facets), we

considered the WHO’s QoL deﬁnition (WHOQoL

Group, 1994). For WHO, there are four domains

(physical, psychological, social, and environment)

and 24 facets in QoL. The results showed that despite

the large number of studies that use the term “Quality

of Life”, few studies (11,7%) correlate this term to the

deﬁnition proposed by WHO. In addition, although

there are proposals for individualized monitoring of

aspects related to QoL, no initiatives have been found

for comprehensive Quality of Life monitoring using

intelligent objects in IoT environments.

Finally, we also decided to extract the

health issues have been addressed in the QoL-

related IoHT literature. It was found 33 health issues

and were clustered into six groups: diseases, health

monitoring, elderly healthcare, illness detection,

ethics, and health at work.

Regarding the diseases, studies about sleep apnea

syndrome, stress, cancer, cardiovascular and cardio-

respiratory diseases, chronic diseases, hypertension,

and diabetes were found. In the health monitoring cat-

egory, the most recurrent issue is how to provide con-

tinuous and real-time monitoring of vital signs. For

elderly healthcare, dementia and falls were the most

mentioned. Finally, it was also found studies focused

on early illness detection, ethical responsibility over

health data, and how to provide a better environment

to employees. Thus, the answer for RQ1 can be sum-

marized as follows.

• RQ1: What is the context of the papers published

in the QoL-related IoHT literature?

Summarized Answer:

– Year and Venue: in the last years, the num-

ber of studies has grown. Also, 55% and 27%

of the papers were published in journals and

conferences, respectively. However, no venue

stood out.

– Hot Topics: IoT Healthcare Services (38.3%),

Elderly Healthcare (17%), Big Data (11.7%),

Sensors and Wearable (9.6%), Security and

Privacy (6.4%), Network and Communication

(6.4%), Health Activity Monitoring (6.4%),

Heath Prediction (3.2%), and Well-being and

Comfort (1%).

– Research Type and Contribution: many pa-

pers were classiﬁed as solution proposals, and

that have monitoring tools as a contribution.

– Empirical Validation: the distribution was

33,8% of case studies, 27,9% of proofs-of-

concept, 19,1% of experiments, 11,8% of sur-

veys, 7,4% of simulations, and was not found

usability evaluations. This result reinforces the

need to conduct practical validations.

– QoL Domains and Facets: many studies

have investigated strategies to improve people’s

QoL, but these strategies are generally focused

on a speciﬁc QoL aspect or speciﬁc health is-

sues. Moreover, it was not found studies fo-

cused on seamless QoL monitoring.

– Health Issues: it was found 33 health issues

into six categories: diseases, health monitoring,

elderly healthcare, illness detection, ethics, and

health at work.

3.2 Challenges

The challenges were identiﬁed by paper’ excerpts us-

ing an open ﬁeld on the extraction form. Then, they

were iteratively reﬁned in order to make them more

concise and to categorize them. This process resulted

HEALTHINF 2022 - 15th International Conference on Health Informatics

400

in 182 challenges grouped into eight categories, and

due to the large number of challenges found, the dis-

cussion will be made by category mentioning the most

recurring challenges.

3.2.1 IoT Challenges

This category, which had the most mentions (71), en-

compasses well-known and intrinsic IoT challenges

that have been studied. Unfortunately, despite the ad-

vances, there is still no “killer solution” for them.

The Lack of Interoperability is a well-known

difﬁculty for those who work with the IoT. This chal-

lenge has many facets, and one of its major causes is

the huge heterogeneity found in IoT environments. It

is possible to ﬁnd devices from many different compa-

nies with speciﬁc protocols and data structures. Het-

erogeneity combined with the lack of widely accepted

standards and the vendor lock-in improve this prob-

lem. In this way, middleware platforms (e.g., FI-

WARE) and data standardization (such as FHIR) are

the most common solutions. The study (Study ID: 32)

used a subset of FHIR to overcome the data interop-

erability in an IoT system for dementia care, and this

kind of initiative has been strengthened by private sec-

tor initiatives like the Argonaut Project. Other authors

have proposed smart gateways capable of supporting

interoperability among heterogeneous sensors (Study

ID: 48), and solutions based on the idea of plug-and-

play, extensible components (Study ID: 50).

Trustworthiness is a critical issue in IoT systems,

and even more when these systems are responsible

for people’s healthcare. This aspect deals with the

user’s expectation of the service competency (Study

ID: 81). Moreover, the trustworthiness is a challenge

in this context due to its close relation with the data

quality (Study ID: 55), privacy (Study ID: 81), and

quality of network (Study ID: 69). The solution to

this problem involves strengthening veriﬁcation and

validation techniques, fault tolerance strategies, in ad-

dition to regulations that seek to protect users in case

of failures.

Some medical services are restricted to rich peo-

ple in underdeveloped countries due to their high-cost

(Study ID: 87). Then, the Cost Efﬁciency is the de-

sired goal for IoT systems applied on remote health

(Study ID: 57). Thus, we observed the development

of low-cost solutions.

Healthcare should be highly personalized (Study

ID: 30). In this way, the Personalized IoT-Health

is a challenge that has been more explored recently.

The authors of the paper (Study ID: 19) stated, for

example, that in general-propose elderly monitoring

systems, several presumptions are made, and these

presumptions can result in inefﬁciencies in the long-

term. Therefore, investigating these challenges rep-

resents an opportunity to develop self-adaptive IoHT

systems or reinforce data analytics strategies.

Standardization is another well-known challenge in

the IoT-related area. The deﬁnition of standards for

data representation, data exchange, communication

protocols, quality of service, development method-

ologies, and many others has the potential to re-

move barriers in the development of many IoHT so-

lutions. In this way, there are initiatives such as the

ISO/IEEE 11073 standards for point-of-care medical

device communication; HL7, FHIR, and OpenEHR

standards for electronic health records; and the DI-

COM for medical images. However, this area still has

opportunities to improve the existing standards or to

propose new ones.

3.2.2 Security and Privacy

Security and privacy challenges were the second most

mentioned (64). The challenges found in this category

are closely related to the three primary security goals:

conﬁdentiality, integrity, and availability. In this way,

the study (Study ID: 87) mentioned the Data Secu-

rity as a critical requirement for IoHT systems and

that it is necessary both develop new solutions to keep

the data consistent and train healthcare professionals

to be aware of this criticality. In addition, problems

with the data can hinder decision-making regarding

the treatment of a patient.

For the Access control, it was found mentions

for the Identify Establishment and Capability-based

Access Control (IECAC) protocol and the Elliptical

Curve Cryptography (ECC) to protect the IoT from

the man-in-the-middle, replay, and denial-of-service

(DDoS) attacks (Study ID: 29). The paper (Study

ID: 52) proposed a framework to preserve privacy

in patient-doctor communication based on public and

private keys.

For the Conﬁdentiality, the investigation (Study

ID: 29) mentioned solutions using the Datagram

Transport Layer Security (DTLS) protocol, and cryp-

tography based on symmetric encryption and elliptic

curve. Other researchers have also investigated the

usage of Blockchain for this purpose (Study ID: 90).

In addition to the previously mentioned chal-

lenges, it was found papers mentioning more spe-

ciﬁc issues such as the problems with methods to

de-identify data without introducing noise and re-

identiﬁcation attacks in anonymization techniques

(Study ID: 17), and security issues in scenarios with

heterogeneous resource-constrained devices (Study

ID: 76). Considering this context, it is crucial to adopt

an expanded view for security and privacy in IoHT

(Study ID: 33).

Internet of Health Things for Quality of Life: Open Challenges based on a Systematic Literature Mapping

401

3.2.3 Data Science

The Data Science category was mentioned 55 times,

and it includes challenges related to Big Data, Data

Analytics, and the usage of Artiﬁcial Intelligence to

support decision making in healthcare. These three

areas are fundamental to move from reactive health,

in which the diagnosis and treatment are deﬁned in

response to symptoms, to proactive health, which is

focused on early warnings (data inference) using the

data collected by smart objects (Study IDs: 82, 17).

Regarding Big Data, it is estimated that the

healthcare industry produces 30% of the entire

world’s data volume. In addition, this data also has a

great variety as it may involve medical images, moni-

toring vital signs, sleep data, location, medical notes,

laboratory test results, among others. Furthermore,

the veracity is related to the quality of data, and it can

impact clinical decision-making (Study ID: 23). So

then, the volume, variety, velocity, and veracity are

concerns that should be addressed during the IoHT

development (Study ID: 62).

Concerning the Data Analytic, the challenge is to

analyze the massive amount of data. Usually, this in-

volves data acquisition, ﬁltering, cleaning and trans-

formation, application of statistical methods and data

mining algorithms, interpretation, and formatting of

results (Study ID: 10). In the results, it was found pa-

pers discussing new algorithms for data cleaning and

improvements in the mining approaches to deal with

heterogeneous data. Also, the work (Study ID: 17)

highlighted issues about data silos.

Another challenge found was the use of Artiﬁ-

cial Intelligence (AI) techniques in healthcare. These

techniques can bring advantages both for the quality

of the services and cost reduction. The challenges re-

lated to AI can be summarized as the application of

Machine Learning and Deep Learning to monitor pa-

tients, recognize user activities, and predict diseases.

For example, the study (Study ID: 5) has used logistic

regression and artiﬁcial neural networks (ANN) for

early detection of hypertension. However, as stated

by (Study ID: 87), there are still challenges to guar-

antee accuracy since the false alerts or the absence of

warnings are critical factors for the IoHT.

3.2.4 Network and Communication

We found 50 mentions of network and communica-

tion issues. The ﬁrst and most mentioned challenge

was Real-time probably because the latency is criti-

cal for healthcare systems. It is currently common to

observe applications in which the data are collected

by sensors, then transmitted to gateways or cloud in-

frastructures to be processed. In some cases, this ap-

proach can introduce an impracticable latency. To

tackle this issue, it has been proposed many strate-

gies, such as the usage of efﬁcient processing units

close to the sensors (Study ID: 49), new system ar-

chitectures to use concepts of edge and fog comput-

ing (Study IDs: 42, 48, 56), or adaptive data trans-

mission policies using mist, fog, and cloud Comput-

ing (Study ID: 84). Although 5G technology was not

much mentioned (only two papers), it is expected that

this new technology will make profound changes in

digital healthcare through its high-throughput, low-

latency wireless connectivity (Study ID: 27).

In addition, we also found mentions for scalabil-

ity (Study ID: 62), availability, and network design is-

sues. These issues are relevant because it is expected

that IoHT systems be able to keep their availability

even with more smart objects and users, and this re-

quirement impacts the network design (such as the

choice for wireless communication and architecture)

(Study ID: 77).

3.2.5 Sensors and Wearable

This category has many challenges due to the grow-

ing interest in wearables. For example, the authors

of the paper (Study ID: 16) estimated that, in the fu-

ture, wearables could represent 30% of health track-

ing. The mapping result found 38 papers discussing

challenges, mainly focused on resource constraints.

Regarding the Device Resources Constraint, the

energy consumption is the major issue (Study ID:

62). The limited power can restrict the transmis-

sion and processing capabilities. As stated by the

study (Study ID: 77), wearables should operate con-

tinuously with minimal human intervention, and the

adoption of large-capacity batteries makes them un-

comfortable. As a practical example, the paper writ-

ten by (Study ID: 76) reinforces these difﬁculties dur-

ing the development of healthcare systems in Canada.

This context paves the way for developing protocols

that support low energy consumption (Study IDs: 74,

87) and for strategies to harvest energy (Study ID:

77). In this way, the work (Study ID: 74) proposed

a routing protocol for body area sensor networks that

uses the bio-inspired multi-objective algorithm to im-

prove data reliability, reducing power consumption.

Another relevant characteristic for wearable is its

User Acceptance. In some case, this requirement can

be a barrier to its usage (Study ID: 38). Thus, it is

also important to investigate less invasive methods to

collect data (Study ID: 16).

Finally, the recent emergence of various ﬁtness

tracking, there is a demand for the development of

approaches for Testing and Validating these devices

(Study ID: 82). In addition, the (Study ID: 72) also

HEALTHINF 2022 - 15th International Conference on Health Informatics

402

reinforces the need for methods to verify the effec-

tiveness of mobile health applications.

3.2.6 Software Engineering

In the studies selected by this mapping, seventeen

(17) pointed out software engineering challenges.

The most mentioned were reliability (Study ID: 42),

reconﬁgurability and remote conﬁguration (Study

IDs: 36, 7), easy installation (Study ID: 80), legacy

systems and technical debts (Study ID: 82), system

compatibility (Study ID: 41), and fault tolerance

(Study ID: 57). In this way, many solutions have been

proposed. For example, reliability and fault tolerance

can be addressed with testing techniques and adaptive

models for this kind of system. The reconﬁgurabil-

ity and easy installation require design solutions; the

legacy systems and technical debts can be mitigated

by decoupling the data from the legacy system and

adopting debt management processes. Finally, there

are also many solutions using middleware platforms

to overcome heterogeneity.

3.2.7 Human-Computer Interaction

Fourteen papers mentioned the following challenges:

usability (Study IDs: 87, 32, 93), non-invasive care

technologies (Study IDs: 75, 34), empowered users

(Study ID: 90), engagement in health interventions

(Study ID: 86), and the acceptance of the elderly

(Study ID: 25). As the adoption of IoHT systems

increases, interest in technologies that provide a bet-

ter user experience should also grow, leading to stud-

ies focused on the impact of functional and non-

functional requirements in that experience.

3.2.8 Cloud Computing

The last category is cloud computing with 6 men-

tions. For example, the complexity of integration

and management of different layers of cloud and

IoT for healthcare systems (Study ID: 62), delay in

cloud computing (Study ID: 91), ofﬂoading (Study

ID: 70), the usage of fog computing in IoT-Health

(Study ID: 43), and the synchronization between

different cloud vendors (Study ID: 46).

• RQ2: What are the challenges related to IoHT for

Quality of Life?

Summarized Answer: Among the large room of

challenges to be addressed, there is a high inter-

est in personalized IoT-Health applications, data

security and privacy, the usage of wearables to

monitor patients, and machine learning to predict

health issues. Also, the strengthening of mobile

health is expected. Naturally, this strengthening

will demand new software engineering methods,

mainly focused on testing and systems’ usability.

3.3 IoHT to Monitor and Improve the

People’s Quality of Life

It is a consensus that health and QoL are closely re-

lated (Study ID: 72). However, there is no signiﬁcant

interest in strategies to measure this QoL gain. Cur-

rently, the most known strategies to measure QoL are

based on questionnaires, which are tiring, and hard to

engage the user (Sanchez et al., 2015). In this way,

a seamless and unnoticeable IoHT-based monitoring

can be more helpful to promote early interventions.

The RQ3 was proposed from the hypothesis that

many works are proposing IoHT solutions to improve

people’s QoL, but only a few studies are concerned

with measuring this indicator. In our results, only

11 papers explicitly mentioned some QoL domain or

facet. However, none have proposed a method, or tool

to ubiquitously infer users’ QoL using IoHT data.

The study published by (Study ID: 75) presents

a broad discussion about QoL for the elderly and its

relation to health. In addition, it was identiﬁed a set

of instruments to measure QoL, such as EQ-SD-3L,

SF-36, and WHOQOL-BREF. Finally, they conclude

by proposing an architecture for non-intrusive mon-

itoring of older adults. The main drawback of this

proposal is that the monitoring module still uses ques-

tionnaires and was not presented any strategy focused

on the semantic structure of the QoL domain and how

the IoHT can produce data to infer the QoL.

The other works address speciﬁc points, such as

QoL of hospitalized children (Study ID: 3), plant

management for indoor comfort (Study ID: 51), sleep

monitoring for apnea treatment (Study ID: 48), and

recognition of emotions (Study ID: 70).

• RQ3: What is the evidence that IoHT can monitor

and improve the people’s Quality of Life?

Summarized Answer: most studies did not seek to

measure QoL gain using IoHT. Few studies have

proposed QoL automated monitoring approaches.

Finally, there is a lack of works focused on QoL,

providing models for the semantic relation of the

QoL-related data, and using AI to infer its value.

4 CONCLUSION

This work was conducted to summarize the literature

about the IoHT applied to the Quality of Life. As

our main highlights, we found a growing interest in

Internet of Health Things for Quality of Life: Open Challenges based on a Systematic Literature Mapping

403

IoHT studies, mainly focused on the elderly. In gen-

eral, there is still room for more partnerships with the

industry to perform validations in practice. Finally,

several solutions for monitoring and diagnosing dis-

eases have been proposed, but an increase in machine

learning solutions for early diagnosis is expected.

Regarding the challenges, there is a desire for per-

sonalized and intelligent services that provide contin-

uous, fast, secure, and effective health data monitor-

ing. Regarding our last research question, the results

show that few studies seek a general approach to deal-

ing with QoL, and the works closer to this proposal

still use questionnaires to collect data. Probably, this

happens due to the difﬁculty in developing this kind of

approach since a large amount of user data is needed,

in addition to the validation complexity, which is done

through longitudinal studies. Thus, this question can

validate the beginning of a more in-depth investiga-

tion towards a semantic structure of the QoL domains

and facets and an intelligent model for capturing and

inferring this metric.

Concerning the gaps, we realized a need for

collaborative-aware and ubiquitous services able to

anticipate events and to act in the environment to im-

prove the living conditions; partnerships with the in-

dustry to conduct validations in practice; methods and

protocols capable of guaranteeing data security and

privacy even on restricted devices; network designs

to ensure low latency and high reliability; techniques

for validation and veriﬁcation of ﬁtness tracking apps;

approaches to the development of intelligent systems

integrated with the domain experts; and studies fo-

cused on user experience.

DATA AVAILABILITY

Our data are public. It is also important to highlight

that the papers selected in this mapping were linked

using the study ID due to space restrictions.

- Protocol: bit.ly/3B6jqth

- Selected Papers: /bit.ly/3FeXzCr

- Images (higher resolution): bit.ly/3D3nP0Q

- Raw Data (codes and tables): bit.ly/3Fe79FV

REFERENCES

Al-Fuqaha, A., Guizani, M., Mohammadi, M., Aledhari,

M., and Ayyash, M. (2015). Internet of things: A

survey on enabling technologies, protocols, and ap-

plications. IEEE communications surveys & tutorials,

17(4):2347–2376.

Araujo, I. L., Andrade, R. M. C., Costa Junior, E., Oliveira,

P. A. M., Oliveira, B. S. ., and Aguilar, P. A. (2020).

Lessons learned from the development of mobile ap-

plications for fall detection. In GLOBAL HEALTH

2020: The Ninth International Conference on Global

Health Challenges (GLOBAL HEALTH 2020).

Blei, D. M., Ng, A. Y., and Jordan, M. I. (2003). Latent

dirichlet allocation. Journal of machine Learning re-

search, 3(Jan):993–1022.

Breivold, H. P. (2017). Internet-of-things and cloud com-

puting for smart industry: A systematic mapping

study. In 2017 5th International Conference on En-

terprise Systems (ES), pages 299–304. IEEE.

Islam, S. R., Kwak, D., Kabir, M. H., Hossain, M., and

Kwak, K.-S. (2015). The internet of things for health

care: a comprehensive survey. IEEE Access.

Oliveira, P. A. M., Andrade, R. M., and Neto, P. A. S.

(2021). Iot-health platform to monitor and improve

quality of life in smart environments. In 2021 IEEE

45th Annual Computers, Software, and Applications

Conference (COMPSAC), pages 1909–1912. IEEE.

Pai, M., McCulloch, M., Gorman, J. D., Pai, N., Enanoria,

W., Kennedy, G., Tharyan, P., and Colford Jr, J. M.

(2004). Systematic reviews and meta-analyses: an il-

lustrated, step-by-step guide. The National medical

journal of India, 17(2):86–95.

Petersen, K., Feldt, R., Mujtaba, S., and Mattsson, M.

(2008). Systematic mapping studies in software en-

gineering. In 12th International Conference on Eval-

uation and Assessment in Software Engineering.

Rodrigues, J. J., Segundo, D. B. D. R., Junqueira, H. A.,

Sabino, M. H., Prince, R. M., Al-Muhtadi, J., and

De Albuquerque, V. H. C. (2018). Enabling technolo-

gies for the internet of health things. IEEE Access,

6:13129–13141.

Sanchez, W., Martinez, A., Campos, W., Estrada, H., and

Pelechano, V. (2015). Inferring loneliness levels in

older adults from smartphones. Journal of Ambient

Intelligence and Smart Environments, 7(1):85–98.

WHO (1998). Whoqol: Measuring quality of life.

https://www.who.int/healthinfo/survey/whoqol-

qualityoﬂife/en/index3.html. Accessed on 11/5/2020.

WHOQoL Group (1994). The development of the world

health organization quality of life assessment instru-

ment (the whoqol). In Quality of life assessment: In-

ternational perspectives, pages 41–57. Springer.

Wieringa, R., Maiden, N., Mead, N., and Rolland, C.

(2006). Requirements engineering paper classiﬁcation

and evaluation criteria: a proposal and a discussion.

Requirements engineering, 11(1):102–107.

Wohlin, C., Runeson, P., H

ost, M., Ohlsson, M. C., Reg-

nell, B., and Wessl

en, A. (2012). Experimentation in

software engineering. Springer Science.

HEALTHINF 2022 - 15th International Conference on Health Informatics

404