Towards an IoHT Platform to Monitor QoL Indicators

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:

Internet of Health Things, Smart Quality of Life, Automated Monitoring.

Abstract:

The Quality of Life has been studied for a long time, and the World Health Organization deﬁnes it as the indi-

vidual perception about life regarding four major domains: physical, psychological, social, and environmental.

The relevance to study QoL lies in the search for strategies able to measure a patient’s well-being. Without

these strategies, treatments, and technological solutions that aim to improve people’s QoL would be restricted

to physicians’ implicit and subjective perceptions. Thus, there are many instruments for formal QoL assess-

ment (usually questionnaires). However, the use of these instruments is time-consuming, non-transparent, and

error-prone. Considering this problem, in this work, we discuss the proposal to use the Internet of Health

Things (IoHT) to collect data from smart environments and apply machine learning techniques to infer QoL

measures. To achieve this goal, we designed an IoHT platform inspired by the MAPE-K loop. Our literature

review has shown that this idea is promising and that there are many open challenges to be addressed.

1 INTRODUCTION

Over the past few years, the use of technologies in

healthcare and medical care has grown (Oliveira et al.,

2021b). For example, years ago, it was practically un-

feasible to continuously monitor a patient without the

intrusive need of a robust hospital infrastructure (Is-

lam et al., 2015). However, advances in the Internet of

Things (IoT) such as sensors miniaturization, efﬁcient

communication protocols, expansion in data process-

ing capacity, application of intelligent algorithms, and

even more secure approaches to guarantee the pri-

vacy of patients have made possible a revolution in

the healthcare area (Mesk

o, 2014).

In this way, the Internet of Health Things (IoHT)

emerges from the application of IoT in healthcare

(Rodrigues et al., 2018) and it is possible to cite many

examples of IoHT solutions. For example, there are

non-invasive glucose sensing (Istepanian et al., 2011),

electrocardiogram monitoring (Agu et al., 2013), el-

derly fall detection (Almeida et al., 2016), and many

others.

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

In general, these solutions have been proposed to

achieve i) individual beneﬁts, by increasing safety

and well-being; ii) economic beneﬁts, by improving

the cost-effectiveness of limited healthcare resources;

and, ﬁnally, iii) social beneﬁts by promoting better

living conditions (Dohr et al., 2010).

Regarding the cost-effectiveness of healthcare

systems, world society has experienced a phe-

nomenon that has put much pressure on them: pop-

ulation aging (Nations, 2019). The United Nations

(UN) projected that, in 2050, there will be 1.5 bil-

lion of older persons. The population aging process is

highly positive since it indicates that we are progress-

ing as a society. However, this new scenario brings

new challenges. Among them, we can highlight the

need for a healthcare system focused on preventive

care (Gmeinder et al., 2017). Today, it is common to

act only when the patient becomes ill (Marvasti and

Stafford, 2012).

The Organization for Economic Co-operation and

Development (OECD) pointed out that only 2.8% of

health spending goes on prevention, and only 7% of

this part was focused on early disease detection (Fig-

ure 1) (Gmeinder et al., 2017). Furthermore, the last

World Health Organization (WHO) report indicates

that this percentage grew only to 5% in 2020 (Vri-

jburg and Hern

andez-Pe

na, 2020).

438

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

Towards an IoHT Platform to Monitor QoL Indicators.

DOI: 10.5220/0010823500003123

In Proceedings of the 15th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2022) - Volume 5: HEALTHINF, pages 438-445

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

Figure 1: Fraction of health expenditure applied in preven-

tive health strategies (Gmeinder et al., 2017).

Thus, one of the strategies that can improve the

effectiveness of healthcare systems is the continuous

monitoring of people’s Quality of Life (QoL) in order

to anticipate the development of health issues related

to physical, psychological, social, and environmental

aspects. According to the WHO, it is crucial to mea-

sure the QoL because it has a close relationship with

the health status, and it provides valuable data to med-

ical practice (Estrada-Galinanes and Wac, 2018).

Nevertheless, given the available techniques and

tools, it is possible to state that seamless Quality of

Life (QoL) monitoring is complex, expensive, non-

transparent, and error-prone. It is complex because

there are at least 150 strategies to measure QoL (Gill

and Feinstein, 1994), and it remains difﬁcult to ﬁrmly

deﬁne QoL (Karimi and Brazier, 2016). It is ex-

pensive and non-transparent because these strategies

(usually questionnaires) should be self-rated by the

patient or should be applied by a healthcare profes-

sional (Bowling, 2005). Furthermore, it is error-prone

since an active answer to a questionnaire can add a

bias in the outcomes (Crane et al., 2016).

Thus, this work presents ongoing research that

aims to create an IoHT platform able to support the

development of self-adaptive applications (Vansyckel

et al., 2013) focused on monitoring and improving

users’ Quality of Life. This study is part of a larger

STIC-AmSud

project - called Angel - which aims

to provide a robust Internet of Things (IoT) platform

to Ambient Assisted Living, offering support to the

improvement of Quality of Life (QoL), especially for

persons with chronic diseases, older adults and per-

sons with acute diseases under medical monitoring.

The idea is to use the Internet of Things (IoT) to ob-

tain and enrich environmental data in order to infer

QoL levels, monitor health vital signs, and identify

atypical situations such as falls and nocturia.

In this paper, we discussed the advances in our

data collection module. The following sections dis-

cuss our background, proposal, ongoing proof of con-

cept, and ﬁnal remarks.

STIC-AmSud: sticmathamsud.org/stic/proyectos.

2 BACKGROUND

Before any formal deﬁnition, having a better Quality

of Life is probably the greatest desire of humankind.

Naturally, this desire has driven the development of

studies focused on improving people’s QoL (Baker

et al., 2017), mainly because there is a close relation-

ship between health and QoL (Guyatt et al., 1993).

However, despite being discussed for a long time

(Elkinton, 1966), the term Quality of Life is con-

fusing and can be observed from many perspectives

(Karimi and Brazier, 2016). The QoL can be related

to the absence of chronic diseases, perception of lone-

liness, physical well-being, and understanding of the

aging/death process. In this work, the WHO deﬁ-

nition for Quality of Life was considered the main

reference. For WHO, QoL can be described as the

individual perception of life in a sociocultural con-

text and concerning goals, expectations, and personal

standards (WHOQoL Group, 1994).

From this deﬁnition, many mechanisms to evalu-

ate QoL have been proposed. One of the most cited

is the WHOQOL-BREF questionnaire (Skevington

et al., 2004) due to its reliability and cross-cultural

validity. The WHOQOL-BREF was evaluated in 23

countries (including Brazil), and it is available in 19

different languages. It has twenty-six (26) questions

distributed into four domains: physical, psychologi-

cal, social, and environment.

The “Physical” domain assesses motor facets such

as daily living activities, medicines’ dependence, mo-

bility, sleep quality, and work capacity. The “Psycho-

logical” domain is related to the bodily image, neg-

ative and positive feelings, self-esteem, spirituality,

and other mental health aspects. The “Social” do-

main observes personal relationships, social support,

and sexual activity. Finally, the “Environment” do-

main aims to evaluate the environmental facets such

as freedom, safety, security, participation in leisure

activities, pollution, noise, trafﬁc, and climate.

Unfortunately, the continuous application of

this kind of instrument is tedious and bothersome

(Sanchez et al., 2015), which makes it challenging

to engage the participants. Therefore, the QoL con-

tinuous monitoring is still an open problem due to

the complexity of the measurement instruments and

the invasive approaches that do not preserve privacy

(Oliveira et al., 2021b). The relevance of this problem

emerges from the health beneﬁts that can be achieved

from up-to-date and accurate QoL information (e.g.,

early interventions). Due to this, some studies have

been conducted to ﬁnd strategies to use the Internet

of Things for seamless QoL monitoring.

Towards an IoHT Platform to Monitor QoL Indicators

439

3 RELATED WORK

In order to compose our related work, a literature re-

view was performed on papers indexed on Elsevier’s

Scopus database. Our search string was composed of

the following terms and their synonyms: “smart qual-

ity of life, passive sensing, internet of health things,

platform and machine learning”. The ﬁrst two terms

were included to ensure the retrieval of three con-

trol papers (previously identiﬁed by the authors). The

other terms are directly associated with our ﬁnal goal:

to build a platform to support the development of

IoHT systems that use Machine Learning to monitor

and improve users’ QoL.

The Scopus database was selected based on its

coverage

of software engineering venues and rele-

vant digital libraries such as ACM, IEEExplorer, Sci-

ence Direct, and Springer. Thus, the selected papers

represent a suitable sample to describe this study area.

Furthermore, it is also important to note that we did

not include any date restriction.

Initially, 122 papers were recovered, but only 13

were chosen after full reading. The eligibility criteria

were: be a primary study, written in English, fully

available on the Web, and with more than 5 pages;

be published in conferences or journals; and discuss

IoHT solutions to monitor QoL automatically.

Table 1 summarizes the 13 selected papers and

our work. However, before starting the discussion of

these works, we would like to highlight the difﬁculty

in building a search string capable of differentiating

studies focused on strategies to automate QoL mea-

surements or that seek to correlate data collected by

IoHT systems with measurements made by QoL ques-

tionnaires; from the studies that bring a speciﬁc pro-

posal for a health problem. In general, both kinds of

these studies use the terms Quality of Life, monitor-

ing, and improvement. Thus, we argue that the term

“Smart Quality of Life” is a suitable candidate to rep-

resent this research area. Within this context, the ﬁrst

authors

to use this expression were (Qiu et al., 2020).

Unfortunately, the authors did not provide a formal

deﬁnition for this term. Therefore, we present here a

formal deﬁnition inspired by the WHO statement.

Smart QoL can be described as the perception of

a person’s Quality of Life from individual and contex-

tual data acquired in Smart Environments using ubiq-

uitous technologies. Usually, this perception is built

through intelligent algorithms, and, over time, it can

be used to detect health issues. Now, in light of this

deﬁnition, it is possible to discuss our related work.

Scopus Coverage: elsevier.com/?a=69451.

Other authors have used this term, but in the context of

smart cities.

We decided to group on the top of Table 1 case

studies. Thus, the criteria “Deal with heterogene-

ity?”, “Use any semantic model?”, “Provide AI for

data analysis?”, and “Allow strategies to adapt user’s

environment” were ﬁlled with a dash because they are

not applicable for them.

The studies presented by (Bade et al., 2018),

(Kim et al., 2019), (Lee et al., 2019), (Angthong and

Veljkovic, 2019), (Oliveira et al., 2021a), and (Brudy

et al., 2021) were classiﬁed as longitudinal studies

because they involve analyzing the participants data

through an extended period in order to prove the cor-

relation between health data and the patients’ QoL.

Although these studies do not present software arti-

facts as the main contribution, their discussion is rele-

vant to indicate strategies for evaluating solutions that

use health data to infer the quality of life of their users.

All these works (excluding only the study conducted

by (Lee et al., 2019)) used commercial smart bands

and their native applications. This decision is prob-

ably related to the costs of these devices (values be-

tween $15 and $60 dollars). In general, devices with

higher processing power that allow the development

of native apps for their platforms are expensive (val-

ues above $300 dollars). Another difﬁculty observed

in these studies is the absence of APIs for data extrac-

tion, which makes this extraction an arduous process.

Regarding the data analysis, all of these longitu-

dinal studies present statistical analyzes to validate

their hypotheses. (Bade et al., 2018) and (Oliveira

et al., 2021a) proved that there is a correlation be-

tween physical activity data and the QoL of people

with cancer. Similarly, (Kim et al., 2019) shown this

correlation for hospitalized patients with spinal is-

sues; (Lee et al., 2019) for patients with ﬁbromyalgia;

(Angthong and Veljkovic, 2019) for adults with foot-

ankle condition, and (Brudy et al., 2021) for children

with congenital heart disease. The results of these

studies can be generalized to state that it is possi-

ble to use data collected by smart objects to mea-

sure the Quality of Life of patients even with dif-

ferent QoL questionnaires and for different health

conditions. This opportunity has also been rein-

forced in renowned medical journals (Huckvale et al.,

2019). Unfortunately, none of these studies made

their datasets available, which hinders the advance-

ment in this study area. Currently, there are many data

silos without a semantic representation that allows its

use in further investigations.

To conclude this ﬁrst group of works, (Concheiro-

Moscoso et al., 2021) brings a protocol to assess the

impact of stress in workers’ QoL. Their main contri-

bution is to present a guide on conducting studies that

seek to correlate health data with QoL facets.

HEALTHINF 2022 - 15th International Conference on Health Informatics

440

Table 1: Papers selected in the literature review compared with our proposal.

Work Contribution

Deal with

hetero-

geneity?

Use any

semantic

model?

Provide

AI for

data

analysis?

Allow

strategies

to adapt

user’s

environ-

ment?

Domains Proﬁle # of Part. Env. Devices Analysis

(Bade et al.,

2018)

Longitudinal

study

- - - - Physical

Patients with

Lung Cancer

Not

speciﬁed

Fitbit Zip and Smartphone

Spearman rank

correlation

(Kim et al.,

2019)

Longitudinal

study

- - - - Physical

People with

spinal issues

22 Hospital Fitbit Charge

Pearson

correlation and

regression

analysis

(Lee et al.,

2019)

Longitudinal

study

- - - - Physical

Fibromyalgia

patients

Not

speciﬁed

Speciﬁc wearable built for

this study

Statistical

analysis

(Angthong

and

Veljkovic,

2019)

Longitudinal

study

- - - - All

Adults with

foot- and

ankle-related

conditions

Not

speciﬁed

Foot pod (Garmin)

Pearson’s

correlation

(Oliveira

et al., 2021a)

Longitudinal

study

- - - - All

Lymphoma

patients

Not

speciﬁed

Microsoft Band 2

Statistical

analysis

(Concheiro-

Moscoso

et al., 2021)

Study

Protocol

- - - - Physical Adults 11 Work Xiaomi Mi Band 3

Statistical

analysis

(Brudy et al.,

2021)

Longitudinal

study

- - - - All

Children with

congential

heart disease

343

Not

speciﬁed

Garmin vivoﬁt Jr

Logistic

regression

(Merilahti

et al., 2012)

Model No No Yes No Physical Older adults 19

Not

speciﬁed

Actigraphy, Bed sensor,

Omron Walking Style II

pedometer and Omron

705IT

Spearman

correlation and

k-means

clustering

(Vargiu et al.,

2014)

Methodology Yes No Yes No All

People with

disabilities

Not

informed

Home

Brain/Neural Computer

Interface (BNCI), inertial

sensors, environmental

sensors, smart home devices

C4.5 and k-NN

(Bono-Nuez

et al., 2014)

System No No Yes No Physical Older adults

Not

informed

Smart

Kitchen

Kitchen appliances, Zigbee

sensors, RFID and portable

devices

Self-organizing

maps (SOM)

(De Masi

et al., 2016)

Platform No No Yes No All Generic

Not

informed

Indoor

and

outdoor

Smartphone and Wearables

User data

timeline

(Dobre et al.,

2019)

Architecture Yes No No No All Older adults

Not

informed

Indoor

and

outdoor

Smartwatch, smart shoes,

camera

Statistical

Analysis

adulescu

et al., 2019)

Framework No No No No All Older adults 17

Not

speciﬁed

Not speciﬁed

Spearman

correlation

Our work Platform Yes Yes Yes Yes

Physical

and Psy-

chological

Adults

progress

Indoor

and

Outdoor

Smartphones, Amazﬁt Bip,

and Smart Home Devices

Machine

Learning

The second group brings methods, frameworks,

systems, and platforms as their main contribution.

In (Merilahti et al., 2012), the authors present

a study about the performance of health monitor-

ing technologies to estimate the physical function of

older adults. They present a hypothesis that health

data would predict pre-clinical measures. Thus, 19

older adults were analyzed through 84 days using

wrist-worn activity monitors, bed sensors, pedome-

ters, weight scales, and blood pressure monitors. The

acquired raw data were transformed into 16 features,

and they were analyzed using statistical correlation

and clustering methods. Unfortunately, the results

were not promising, indicating only a correction with

the daily steps. However, this work brings interesting

insights about which features can be used in this type

of investigation and points out issues in data collec-

tion. Compared to our proposal, this paper presents

only a speciﬁc model for the physical domain, which

does not consider self-assessment questionnaires and

does not concern itself with other aspects of IoHT

(such as heterogeneity, semantic model, and environ-

mental adaptation).

(Vargiu et al., 2014) propose a context-aware

methodology to telemonitor QoL concerning the

physical and social autonomy of people with disabili-

ties. Thus, they adapted the EQ-5D-5L questionnaire

to assess mood, health status, mobility, self-care,

usual activities, and pain/discomfort. On the other

hand, health data were collected by Brain/Neural

Computer Interface (BNCI), inertial and environmen-

tal sensors, and smart home devices. Finally, the au-

thors achieved good results using the C4.5 and k-NN

algorithms. Nonetheless, due to the lack of real data,

synthetic data was used.

(Bono-Nuez et al., 2014) focused their contri-

bution in creating a QoL evaluation system to sup-

port the work of caregivers. The idea is to provide

QoL assessments of older adults periodically to help

decision-making of caring actions. Unlike other stud-

ies, the authors did not choose a QoL questionnaire

as a reference. Instead, they decided to cluster the

data using self-organizing maps (SOM). Neverthe-

less, their proposal was focused on smart kitchens,

and it requires the analysis of a domain expert to in-

terpret the results obtained by the SOM.

Towards an IoHT Platform to Monitor QoL Indicators

441

The work proposed by (De Masi et al., 2016) is

probably the most related to our proposal. The signiﬁ-

cant difference is that the authors proposed a platform

to support interdisciplinary studies related to Quality

of Life. In contrast, we are focused on helping the

development of IoHT systems capable of using data

from Smart Environments to provide a closed adapta-

tion loop. The authors present the ﬁrst version of the

mQoL Living Lab and the requirements for the sec-

ond version in their paper. Unfortunately, this plat-

form is not publicly available, and it was not possible

to conduct a more in-depth test.

(Dobre et al., 2019) propose an IoT architecture

to deliver non-intrusive monitoring and support older

adults’ healthcare. One of the most interesting points

of this work lies in the authors’ concern about inher-

ent Internet of Things challenges, such as interoper-

ability. The architecture was designed with a mod-

ular structure and, similar to the work proposed by

(De Masi et al., 2016), the data analysis module is

aimed at conducting scientiﬁc studies. However, this

work did not address semantic models for QoL data,

intelligent techniques to infer QoL, and strategies to

act in the environment.

The work proposed by (R

adulescu et al., 2019)

brings a framework to ﬁnd a correlation of health pa-

rameters with QoL questionnaires. The authors deal

with this problem using mathematical models. They

selected the Technique for Order Preference by Sim-

ilarity to an Ideal Solution (TOPSIS), which uses the

concept of “ideal” and “anti-ideal” solutions and com-

putes the Euclidean distance to ﬁnd an overall health

index for the elderly. Nonetheless, the method was

evaluated only with synthetic data, and its complexity

makes its adoption in several contexts difﬁcult.

Finally, the analysis of these papers highlights that

there are still opportunities in developing platforms

that enable the collection of health data, the pro-

cessing of these data using intelligent algorithms, the

planning of interventions in case of risky situations,

and the environment adaptation. In other words, a

closed loop of health care that seeks to identify health

issues before they become a reality.

4 OUR PROPOSAL

This investigation has the ﬁnal goal to develop an

IoHT solution to collect data from smart environ-

ments and apply machine learning to infer QoL mea-

sures. Moreover, this solution should allow high-level

health interventions to adapt the user’s environment.

To achieve this goal, we are proposing a platform

to support the development of this kind of system.

The rationale for building a platform lies in the bene-

ﬁts obtained by software reuse since other researchers

or practitioners can use it to implement solutions for

speciﬁc contexts (Ara

ujo et al., 2018). In addition,

the platform is being designed to address challenges

such as lack of interoperability, high volatility, high

development cost when it involves AI, absence of

a semantic model for QoL data, and difﬁculties to

bring healthcare professionals closer to the develop-

ment process (Oliveira et al., 2021b).

Figure 2 presents an overview of this proposal. We

decided to organize the platform modules following

the MAPE-K loop framework (IBM, 2005). Hence,

the platform will support collecting and analyzing

data, plan what actions should be taken based on the

analysis result, and adjust the environment. Further-

Figure 2: IoHT platform design inspired on the MAPE-K loop.

HEALTHINF 2022 - 15th International Conference on Health Informatics

442

more, it will be possible to represent the knowledge

using a QoL ontology.

In the monitor are the modules responsible for ob-

taining user data. These data can be collected from

mobile devices (e.g., smartphones), context brokers,

or EHR (Electronic Health Record) systems. As dis-

cussed in our related work, the literature already has

evidence that data collected from wearables can infer

QoL aspects. In this way, this proposal argues that an

expanded view of devices present in smart environ-

ments allows a broad and accurate QoL inference.

Additionally, it will be integrated a previous work

called LoCCAM-IoT (Andrade et al., 2020), which is

a multifaceted infrastructure to support the develop-

ment of self-adaptive IoT systems. This infrastructure

has three major modules: i) CoAP-CTX, an extension

of the CoAP protocol for context-awareness device

discovery; ii) LoCCAM, a middleware for acquisition

and context management that uses smartphones as a

decision center; and iii) SUCCEEd, which was cre-

ated to support the self-adaptation using workﬂows.

In the analysis area, the platform will provide ma-

chine learning techniques adapting the Athena tool,

which is a visual, customizable, cloud-based tool to

support the development of systems that require Com-

putational Intelligence (CI) techniques. It uses the ab-

straction of visual modules to encapsulate the CI al-

gorithms allowing their interconnection to solve com-

plex problems (Oliveira et al., 2018). All these tech-

niques should support the creation of an intelligent

model to infer QoL.

Monitoring and data analysis will provide a QoL

proﬁle for the user. With this proﬁle, it will be pos-

sible to identify risk situations and plan health inter-

ventions. These interventions - deﬁned by health pro-

fessionals - can be recommendations for changes in

habits; or even the execution of adaptations in the en-

vironment. In critical cases, it will be possible to re-

quest medical interventions.

Finally, the execution of this loop generates a vast

amount of data, which represents the knowledge ac-

quired in that context. Thus, the platform also pro-

vides an ontology for representing and storing data.

5 PROOF-OF-CONCEPT

A Proof-of-Concept (PoC) is in progress in order to

validate our proposal. This PoC aims to anticipate

the best strategies for collecting data from users and

which machine learning algorithms are suitable for

the inference process.

For this, we developed an Android application –

called QoL Monitor – to collect contextual and health

data from users. In this version, we collect sociode-

mographic and anthropometric data, sleep duration,

daily steps, calories spent, physical activities, heart

rate, location, apps usage time, and the number of

calls made or received. To do that, it was necessary to

integrate our app with the Google Fit API. Thus, users

who participate in this study can use different wear-

ables as long as they are integrated with the Google

Fit account.

Since we are looking to validate the creation of

intelligent models to infer users’ QoL, it will be nec-

essary for them to answer questionnaires informing

their QoL perception periodically. We are working

to reduce the use of these questionnaires. However,

today, they are our best alternative to calibrate our

machine learning models, as they have been validated

with many participants in different countries.

Figure 3 presents the QoL Monitor data collec-

tion workﬂow. It is possible to observe some impor-

Figure 3: QoL Monitor data collection workﬂow.

Towards an IoHT Platform to Monitor QoL Indicators

443

tant points: i) the questionnaire can be externally cus-

tomized, making easier changes throughout the study

or even the validation of another questionnaire; ii)

historical health data is obtained via Google Fit API;

and iii) before sending the data to the server, they are

anonymized and encrypted using AES-256 algorithm,

in addition to the exchange of RSA keys for secure

communication with the server.

Currently, the QoL Monitor has been ﬁnished and

internally validated by the researchers. Thus, our next

step is to request permission for the ethics commit-

tee to start a pilot study with a larger number of par-

ticipants. Our purpose is also to create a public and

anonymous database to improve QoL inference mod-

els. Unfortunately, to the best of our knowledge, no

databases were found able to correlate contextual and

health data with self-report QoL questionnaires.

6 FINAL REMARKS

This work is just beginning, and there are many points

to be deﬁned. However, we argue that this paper dis-

cusses an interesting research topic to guide the scien-

tiﬁc community towards an IoHT platform to monitor

and improve people’s Quality of Life.

In this paper, we present the platform’s design in-

spired by the MAPE-K loop. Using this approach, it is

possible to support IoHT applications able to monitor

and analyze user data and plan and execute interven-

tions in the environment.

Furthermore, we state that there is a growing need

for investments in solutions capable of anticipating

health issues. This kind of solution has the poten-

tial to move our healthcare system from passive care

to active care, increasing its cost-beneﬁt. Due to this,

we also discussed 13 related works and proposed a

deﬁnition for the term Smart Quality of Life.

As our future work, we highlight the conclusion

of our PoC; the investigation of which health issues

can be early detected by analyzing QoL over time; the

development of the ﬁrst version of our IoHT platform;

and the conduction studies to validate the platform.

ACKNOWLEDGMENTS

We would like to thank CNPQ for the Productiv-

ity Scholarship of Rossana M. C. Andrade DT-2 (N

315543 / 2018-3), for the Productivity Scholarship of

Pedro A. dos Santos Neto DT-2 (N

315198 / 2018-4).

REFERENCES

Agu, E., Pedersen, P., Strong, D., Tulu, B., He, Q., Wang,

L., and Li, Y. (2013). The smartphone as a medical de-

vice: Assessing enablers, beneﬁts and challenges. In

IEEE Int. Workshop of Internet-of-Things Networking

and Control, pages 48–52. IEEE.

Almeida, R. L., Macedo, A. A., de Ara

ujo,

I. L., Aguilar,

P. A., and Andrade, R. M. (2016). Watchalert: An

evolution of the falert app for detecting falls on smart-

watches. In Extended Proceedings of the XXII Brazil-

ian Symp. on Multimedia and Web Systems, pages

124–127. SBC.

Andrade, R. M., Junior, B. R. A., Oliveira, P. A. M., Maia,

M. E., Viana, W., and Nogueira, T. P. (2020). Mul-

tifaceted infrastructure for self-adaptive iot systems.

Information and Software Technology, page 106505.

Angthong, C. and Veljkovic, A. (2019). Relationships

among subjective patient-reported outcome, quality of

life, and objective gait characteristics using wearable

foot inertial-sensor assessment in foot–ankle patients.

European Journal of Orthopaedic Surgery & Trauma-

tology, 29(3):683–687.

Ara

ujo, t., Andrade, R., and Aguilar, P. (2018). Artefatos de

Reuso para Internet das Coisas: Arquiteturas, Middle-

wares, Frameworks e Padr

oes. Thesis report, Federal

University of Cear

Bade, B. C., Brooks, M. C., Nietert, S. B., Ulmer, A.,

Thomas, D. D., Nietert, P. J., Scott, J. B., and Silvestri,

G. A. (2018). Assessing the correlation between phys-

ical activity and quality of life in advanced lung can-

cer. Integrative cancer therapies, 17(1):73–79.

Baker, S. B., Xiang, W., and Atkinson, I. (2017). Internet of

things for smart healthcare: Technologies, challenges,

and opportunities. IEEE Access, 5:26521–26544.

Bono-Nuez, A., Blasco, R., Casas, R., and Mart

ın-del Br

ıo,

B. (2014). Ambient intelligence for quality of life as-

sessment. Journal of Ambient Intelligence and Smart

Environments, 6(1):57–70.

Bowling, A. (2005). Just one question: If one question

works, why ask several?

Brudy, L., Meyer, M., Oberhoffer, R., Ewert, P., and M

uller,

J. (2021). Move more–be happier? physical activ-

ity and health-related quality of life in children with

congenital heart disease. American Heart Journal,

241:68–73.

Concheiro-Moscoso, P., Groba, B., Mart

ınez-Mart

ınez,

F. J., Miranda-Duro, M. d. C., Nieto-Riveiro, L., Pou-

sada, T., Queir

os, C., and Pereira, J. (2021). Study for

the design of a protocol to assess the impact of stress

in the quality of life of workers. Int. journal of envi-

ronmental research and public health, 18(4):1413.

Crane, M., Rissel, C., Greaves, S., and Gebel, K. (2016).

Correcting bias in self-rated quality of life: an ap-

plication of anchoring vignettes and ordinal regres-

sion models to better understand qol differences

across commuting modes. Quality of life research,

25(2):257–266.

De Masi, A., Ciman, M., Gustarini, M., and Wac, K. (2016).

mqol smart lab: quality of life living lab for interdisci-

HEALTHINF 2022 - 15th International Conference on Health Informatics

444

plinary experiments. In Proceedings of the 2016 ACM

Int. Joint Conf. on Pervasive and Ubiquitous Comput-

ing: Adjunct, pages 635–640.

Dobre, C., Bajenaru, L., Marinescu, I. A., and Tomescu, M.

(2019). Improving the quality of life for older people:

From smart sensors to distributed platforms. In 2019

22nd Int. Conf.on Control Systems and Computer Sci-

ence (CSCS), pages 636–642. IEEE.

Dohr, A., Modre-Opsrian, R., Drobics, M., Hayn, D., and

Schreier, G. (2010). The internet of things for am-

bient assisted living. In 7th int. conf. on information

technology: new generations, pages 804–809. IEEE.

Elkinton, J. R. (1966). Medicine and the quality of life.

Annals of Internal Medicine, 64:711–714.

Estrada-Galinanes, V. and Wac, K. (2018). Visions and

challenges in managing and preserving data to mea-

sure quality of life. In 3rd Int. Work. on Foundations

and Applications of Self Systems, pages 92–99. IEEE.

Gill, T. M. and Feinstein, A. R. (1994). A critical appraisal

of the quality of quality-of-life measurements. Jama,

272(8):619–626.

Gmeinder, M., Morgan, D., and Mueller, M. (2017). How

much do oecd countries spend on prevention?

Guyatt, G. H., Feeny, D. H., and Patrick, D. L. (1993). Mea-

suring health-related quality of life. Annals of internal

medicine, 118(8):622–629.

Huckvale, K., Venkatesh, S., and Christensen, H. (2019).

Toward clinical digital phenotyping: a timely oppor-

tunity to consider purpose, quality, and safety. NPJ

digital medicine, 2(1):1–11.

IBM (2005). An architectural blueprint for autonomic com-

puting.

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, 3:678–

708.

Istepanian, R. S., Hu, S., Philip, N. Y., and Sungoor, A.

(2011). The potential of internet of m-health things

“m-iot” for non-invasive glucose level sensing. In An-

nual Int. Conf. of the IEEE Engineering in Medicine

and Biology Society, pages 5264–5266. IEEE.

Karimi, M. and Brazier, J. (2016). Health, health-related

quality of life, and quality of life: what is the differ-

ence? Pharmacoeconomics, 34(7):645–649.

Kim, D. H., Nam, K. H., Choi, B. K., Han, I. H., Jeon, T. J.,

and Park, S. Y. (2019). The usefulness of a wear-

able device in daily physical activity monitoring for

the hospitalized patients undergoing lumbar surgery.

Journal of Korean Neurosurgical Society, 62(5):561.

Lee, J., Park, S.-H., Ju, J. H., and Cho, J. H. (2019). Appli-

cation of a real-time pain monitoring system in korean

ﬁbromyalgia patients: A pilot study. International

journal of rheumatic diseases, 22(5):934–939.

Marvasti, F. F. and Stafford, R. S. (2012). From “sick

care” to health care: reengineering prevention into the

us system. The New England journal of medicine,

367(10):889.

Merilahti, J., P

arkk

a, J., and Korhonen, I. (2012). Esti-

mating older people’s physical functioning with au-

tomated health monitoring technologies at home: fea-

ture correlations and multivariate analysis. In Interna-

tional Conference on Grid and Pervasive Computing,

pages 94–104. Springer.

Mesk

o, B. (2014). The guide to the future of medicine: tech-

nology and the human touch. Webicina kft.

Nations, U. (2019). World population prospects: the 2019

revision. United Nations, Department of Economic

and Social Affairs, Population Division.

Oliveira, A., Silva, E., Aguiar, J., Faria, B. M., Reis, L. P.,

Cardoso, H., Gonc¸alves, J., Oliveira e S

a, J., Carvalho,

V., and Marques, H. (2021a). Biometrics and quality

of life of lymphoma patients: A longitudinal mixed-

model approach. Expert Systems, 38(4):e12640.

Oliveira, P., Neto, P. S., Britto, R., Rabelo, R., Braga, R.,

and Souza, M. (2018). Ciaas-computational intelli-

gence as a service with athena. Computer Languages,

Systems & Structures, 54:95–118.

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

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

quality of life in smart environments. In 8th IEEE Int.

Workshop on Medical Computing. IEEE.

Qiu, L., Tong, Y., Yang, Q., Sun, N., Gong, Y., and Yin,

X. (2020). Reliability and validity of a smart quality

of life scale for patients with tuberculosis. Journal of

Public Health, 28(5):575–582.

adulescu, C. Z., Alexandru, A., and B

ajenaru, L. (2019).

Health parameters correlation in an iot monitoring,

evaluation and analysis framework for elderly. In 2019

23rd International Conf. on System Theory, Control

and Computing (ICSTCC), pages 531–536. IEEE.

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 tech-

nologies 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.

Skevington, S. M., Lotfy, M., and O’Connell, K. A. (2004).

The world health organization’s whoqol-bref quality

of life assessment: psychometric properties and re-

sults of the international ﬁeld trial. Quality of life Re-

search, 13(2):299–310.

Vansyckel, S., Sch

afer, D., Schiele, G., and Becker, C.

(2013). Conﬁguration management for proactive

adaptation in pervasive environments. In 2013 IEEE

7th International Conference on Self-Adaptive and

Self-Organizing Systems, pages 131–140. IEEE.

Vargiu, E., Fern

andez, J. M., and Miralles, F. (2014).

Context-aware based quality of life telemonitoring. In

Distributed Systems and Applications of Information

Filtering and Retrieval, pages 1–23. Springer.

Vrijburg, K. and Hern

andez-Pe

na, P. (2020). Global spend-

ing on health: Weathering the storm 2020. World

Health Organization Working paper, (19.4).

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.

Towards an IoHT Platform to Monitor QoL Indicators

445