Applications of Technological Solutions in Primary Ways of Preventing Transmission of Respiratory Infectious Diseases—A Systematic Literature Review

Abstract

With the growing concern about the spread of new respiratory infectious diseases, several studies involving the application of technology in the prevention of these diseases have been carried out. Among these studies, it is worth highlighting the importance of those focused on the primary forms of prevention, such as social distancing, mask usage, quarantine, among others. This importance arises because, from the emergence of a new disease to the production of immunizers, preventive actions must be taken to reduce contamination and fatalities rates. Despite the considerable number of studies, no records of works aimed at the identification, registration, selection, and rigorous analysis and synthesis of the literature were found. For this purpose, this paper presents a systematic review of the literature on the application of technological solutions in the primary ways of respiratory infectious diseases transmission prevention. From the 1139 initially retrieved, 219 papers were selected for data extraction, analysis, and synthesis according to predefined inclusion and exclusion criteria. Results enabled the identification of a general categorization of application domains, as well as mapping of the adopted support mechanisms. Findings showed a greater trend in studies related to pandemic planning and, among the support mechanisms adopted, data and mathematical application-related solutions received greater attention. Topics for further research and improvement were also identified such as the need for a better description of data analysis and evidence.

1. Introduction

According to Baldominus et al. [1], infectious diseases are the result of the invasive action of microscopic organisms (e.g., bacteria or viruses) in the body, and may be presented in many different types with different effect ranges. For example, while some infected bodies can remain asymptomatic, others can reach high risks of death.

Given the importance of the issue and its impact on human life, medicine has developed a variety of mechanisms for the prevention, prediction, diagnosing and treatment of infections [1]. However, over the years, new microorganisms have emerged, generating constant challenges for science in combating the action of these organisms towards humanity.

An important example of this scenario is the current pandemic crisis brought about by the new coronavirus (SARS-CoV-2), responsible for COVID-19 disease. Discovered in Wuhan and rapidly spread since December 2019 within China to other countries of the world [2,3], the newly identified coronavirus has generated considerable challenges both in terms of safety in public health, as well as economic and social impacts to society [4].

Due to the high occurrence rate, as well as severe health symptoms and high fatalities worldwide, on 31 January 2020, the World Health Organization (WHO) announced a global pandemic and on 11th March, the COVID-19 disease was recognized as a pandemic [5].

Since then, the number of cases and fatalities have been constantly making headlines around the world, where on 9 August 2021, the number of confirmed COVID-19 cases reached over 203 million with more than 213 countries and regions affected by the pandemic [6,7].

Figure 1 presents the evolution of the total cumulative count of identified COVID-19 cases around the world in the period of 22 January 2020 to 1 August 2021, and Figure 2 presents the evolution of the total number of deaths around the world in the period of 23 January 2020 to 1August 2021.

Figure 1. Total cumulative count of coronavirus cases (199,560,514) worldwide. Period: 22 January 2020 to 1 August 2021. Publication date: 2 August 2021. Source: https://www.worldometers.info/coronavirus/worldwide-graphs/#case-distribution (accessed on 4 August 2021).

Figure 2. Total cumulative count of deaths caused by coronavirus (4,240,374) worldwide. Period: 23 January 2020 to 1 August 2021. Publication date: 2 August 2021. Source: https://www.worldometers.info/coronavirus/worldwide-graphs/#case-distribution (accessed on 4 August 2021).

From the emergence of new infectious diseases, new research studies are also being carried out in order to contribute to their treatment motivated not only because of health crisis, but also social and economic impacts. However, until new medications or vaccines are produced, preventive measures are recommended by health organizations in order to reduce transmission among the population, such as social distancing, mask usage, isolation and quarantine [8,9,10].

Being a topic of considerable importance, especially due to the social, health and economic impacts to society, studies focused on the application of technology in the primary forms of prevention of new infectious diseases have attracted much attention and concern from institutions and researchers.

Despite the existence of several publications presenting approaches and different uses of technology in this context, to the best of our knowledge, there are no records of research aimed at the identification, registration, selection, and rigorous analysis and synthesis of this literature.

Additionally, due to the large volume of studies, and the fact that they are published in several conferences and journals, it is difficult to locate these works.

To assist current and future researchers in discovering these studies, as well as to identify, select, rigorously analyze, and synthesize this literature, a systematic literature review (SLR) is presented in this paper.

The scope of this SLR was to identify relevant studies that adopt information technology solutions in the primary ways of preventing respiratory infectious diseases transmission/spread.

This study also aims to assist in understanding what is being carried out and studied, discovering new directions, as well as having a better understanding of how technologies are being used in the context proposed in this work, its main objectives, support mechanisms adopted, level of evidence reported, gaps that need to be deepened in research, as well as to organize the knowledge to support the technological transition.

In this paper, the design, execution and findings of a systematic review of the literature are presented, aiming at a systematic identification, selection and summarization of a comprehensive set of approaches that adopt technological solutions in the primary ways of respiratory infectious diseases transmission prevention.

In this review, 219 relevant papers (from 1139 initially retrieved) were selected and rigorously analyzed performing data extraction, analysis, and synthesis according to predefined inclusion and exclusion criteria, in order to answer a set of research questions that motivated this review.

The SLR took place in three stages. In the first one, a proposed search string (see Section 3.2.2) was executed in four digital libraries obtaining 1139 papers. In the second stage, an initial filter was performed in the title, keywords and abstract of the studies, applying the inclusion and exclusion criteria (see Section 3.3) resulting in 239 articles. Finally, performing a full text reading of the remaining articles, the inclusion and exclusion criteria were applied again, resulting in the 219 relevant studies.

From the findings, it was possible to identify six application domain categories in which there was a greater trend in studies related to pandemic planning and, among the support mechanisms adopted, data and mathematical application-related solutions received greater attention.

The two significant contributions of this study are:

• This work presents the design, execution, and results of a comprehensive systematic literature review of relevant studies on information technology applications in the primary ways of new respiratory infectious diseases transmission prevention. The study was based on predefined selection criteria, where rigorous analysis and synthesis of the approaches and associated support mechanisms were carried out, reporting the evidence in an easily accessible format.
• In this study, the approaches, support mechanisms and available evidence were structured and classified using different formats that are expected to be useful to practitioners and researchers. It is also expected that the findings will also identify issues relevant to interested researchers, and that can be used as an evidence-based guide to select appropriate technologies, approaches, solutions, or support mechanisms based on the different needs or scenarios.

2. Background and Related Work

Having an important role in several actions in support of treatment, combat and prevention of new infectious diseases, the adoption of information technology in this scenario has attracted great attention and concern from researchers and practitioners.

However, concerning new infectious diseases that may emerge, until the production of immunizers or medications that will make their treatment possible, preventive actions must be taken by the population to reduce the rates of contamination and fatalities, such as social distancing, mask usage, isolation, or quarantine [8,9,10].

To this end, several research studies have been proposed and carried out regarding the application of technology in this context, as mentioned by Chen et al. [11], who conducted a review on the developments and challenges of current contact tracing technologies. According to Chen et al. [11], contact tracking is one of the key technologies that may be adopted in the prevention and control of infectious diseases, which may be helpful in the location and isolation of infected people and high-risk individuals, preventing further spread of the diseases.

Also, in the line of contact tracing study, Ahmed et al. [12] presented an overview of proposed tracing app examples adopted in the fight against COVID-19, discussing the concerns reported by users regarding their usage, and outlining potential research directions for next-generation app design. Regarding challenges and future directions of contact tracing in the assistance of the fight against coronavirus, Chowdhury et al. [13] reviewed data-driven solutions and apps to identify their strength and weakness, and Hasaini et al. [14] presented a literature review of contact tracing approaches and applications adopted in governments around the world to monitor and control the spread of the COVID-19 disease.

According to Ricci et al. [15], several blockchain studies are also emerging in order to assist in the fight against COVID-19. In a survey conducted by the author, ways in which blockchain technology can be useful in supporting health actions were presented, including in contact tracing and vaccine support.

Regarding machine learning applications, Chamola et al. [16] provided a review of several machine learning algorithms that can be adopted in disaster and pandemic management, also presenting a tutorial on machine learning algorithms. Mathematical models, including compartment, statistical and machine learning models for COVID-19 transmission and diagnosis were also reviewed in a work presented by John et al. [17].

Other researches related to data science are also emerging in terms of applications in the prevention of infectious diseases. Regarding Big data, Sudana et al. [18] performed an analysis of the use of Big Data in the health domain in order to identify the benefits of its application in preventing the spread of infectious diseases such as Tuberculosis. A survey of the state of the art of researches based on data science process application for Dengue infection combat was also carried out by Siriyasatien et al. [19]. The work presented some issues to be explored and analyzed such as data sources, data preparation and representation techniques, and forecasting models.

Deep learning-based techniques were analyzed in a survey performed by Ikram et al. [20], which are classical techniques that can be used to detect COVID-19 Standard Operating Procedures (SOP), such as wearing masks or social distancing, were explored.

Studies that adopt Internet of Things Technologies have also been gaining space in actions applied in the health domain and prevention of infectious diseases. In a review carried out by Manavi et al. [21], the authors explored various Internet of Things technologies and applications adopted in contact tracing, screening, and surveillance, aiming to provide an overall understanding of the identified solutions in the fight against COVID-19.

Regarding Computational Intelligence applications, Baldominos et al. [1] performed a systematic literature review aiming to find studies that adopted computational intelligence to predict infections in patients using physiological data as features. The study analyzed 101 relevant documents between the period of 2003 and 2019, showed that automatic diagnosis of these diseases is well documented in medical literature and concluded that sepsis, Clostridium difficile infection and surgical site infections were the most addressed. Most of the identified studies adopted machine learning techniques.

Focusing on wearable devices, unobtrusive sensing systems and telehealth technologies, Ding et al. [22] presented a review on technologies and systems with various application scenarios for handling the COVID-19, and, regarding applications (apps and systems) developed by Government Institutions, Private Firms, and Individual Citizens across the world, Gupta et al. [23] performed a survey encompassing more than 100 apps to identify the different categories where technology is being used for decision making. The major areas of application covered in the study were Contact Tracing, Social Distancing, Mask Detection, Information Searching, Big Data, among others.

Regarding digital interventions for fighting COVID-19, Nazrul and Najmul Islam [24] performed a review to compare the Bangladeshi perspective with other countries. As the topic was still emerging and there was not much academic literature available, the authors reviewed online content using Google and Yahoo search engines. A total of 57 online e-resources were found, including news and blogs articles, web contents of organizations and online press releases. After identifying some digital interventions in the fight against COVID-19, both in different parts of the world and in Bangladesh, a comparative analysis was carried out and areas were proposed where Bangladesh could focus to strengthen the fight against COVID-19.

Also considering applications of technology in the COVID-19 pandemic scenario, Whitelaw et al. [25] made available a viewpoint with a framework for the application of digital technologies in pandemic management and response. The authors highlighted successful technologies applications in different countries regarding pandemic planning, surveillance, quarantine, health care, contact tracing and testing.

Thus, over time, several surveys and reviews focused on technical solutions to provide assistance in the fight against infectious diseases have appeared in conferences or journals. However, to the best of our knowledge, there were no records of works aimed at systematic reviewing (identifying, selecting, rigorously analyzing and synthesizing) the literature focusing on the application of technological solutions in the primary ways of respiratory infectious diseases transmission prevention.

Through this systematic literature review, we are interested in finding out what technological solutions and support mechanisms are available in scientific works, and how they can contribute to the primary forms of prevention of the spread of new respiratory infectious diseases.

This work is structured as follows: in Section 3, the systematic literature review method is described and the review protocol is defined. Demographic information, quality assessment and research questions analysis are presented in Section 4. Threats to the validity, implications and limitations are discussed in Section 5 and, finally, conclusions are presented in Section 6.

3. Research Method

Being one of the most widely used research methods in Evidence-Based Software Engineering (EBSE), the Systematic Literature Review (SLR) provides a well-defined process for identifying, evaluating, and interpreting all available evidence relevant to a specific research question or topic, as well as evaluates existing studies on a specific phenomenon in a fair and credible manner [26].

This study was performed following the guidelines of Kitchenhamet et al. [26] involving three main phases: definition of a review protocol, conduction of the review and the review report. The adopted review protocol consists of the following elements: (i) research questions, (ii) search strategy, (iii) inclusion and exclusion criteria, (iv) study selection, (v) evaluation of study quality and (vi) data extraction and synthesis, which will be discussed in the following subsections.

3.1. Research Questions

Through this systematic literature review, the aim is to summarize and provide an overview of current research on “which approaches that adopted information technology in the primary ways of prevention of respiratory infectious diseases were reported in the peer-reviewed literature?”. For this, a set of research questions (RQs) were formulated (see Table 1) in order to be answered through this SLR.

Table 1. Proposed research questions (RQ).

Research Question	Motivation
RQ1: Which approaches have been adopted or suggested in the primary ways of preventing respiratory infectious diseases using information technology solutions?	The purpose is to identify infectious diseases transmission prevention approaches that are proposed or applied with the use of information technology
RQ2: What are the application domains of the identified approaches?	Identify the application domains of the approaches. With this information, researchers and practitioners can identify the application domains that have gained the most interest in the primary forms of prevention of infectious diseases
RQ3: Which support mechanisms are proposed or used?	What technologies, systems, techniques, among other mechanisms have been proposed or adopted to support or achieve the objectives of the approaches? This knowledge can help practitioners and/or researchers in the identification of trends in the use of technological solutions as support mechanisms.
RQ4: How much evidence to support the adoption of the approaches is available?	Obtain knowledge about the maturity of the identified approaches, in order to assist researchers and practitioners in further adoption or evaluation of existing approaches in this systematic review, maturity was measured based on levels of evidence (see Section 3.4).
RQ5: Which contexts are addressed?	Identify the contexts (academy or industry) in which the studies were applied, validated or evaluated. For industrial context, real case data or evaluation in real case scenarios is considered. If validations were described in both contexts, the industrial context is considered for the purpose of evaluating the works.

The questions were defined in order to cover the objectives of this SLR, which are: identifying what approaches have been adopted or suggested in the primary ways of preventing respiratory infectious diseases using information technology solutions (RQ1), identification of the application domains of the approaches (RQ2), identification of which support mechanisms are proposed or used (RQ3), how much evidence to support the adoption of the approaches is available (RQ4), and the addressed contexts (RQ5).

The answers to these questions can provide systematic insight and overview to researchers and practitioners regarding approaches and support mechanisms proposed in scientific studies, also helping to identify missing gaps and opportunities for improvement.

3.2. Search Strategy

According to Kitchenhamet et al. [26], in order to help researchers to get as many relevant studies as possible, the search strategy is essential. In this SLR, the research was conducted with various combinations of derivative terms related to the subject of the study, where the adopted search strategy was composed of the following elements: search method, search items and data sources.

3.2.1. Search Method

For the search strategy, automatic searches on electronic database engines or digital libraries (listed in Table 2) were performed using the search terms presented in Section 3.2.2.

Table 2. Electronic databases adopted in the automatic searches.

Electronic Database	Search Terms Matched With	Web Address	Publications Found
IEEE Xplore Digital Library	Paper title, keywords, abstract	http://ieeexplore.ieee.org	117
ACM Digital Library	Paper title, keywords, abstract	http://dl.acm.org	94
El Compendex	Paper title, keywords, abstract	http://engineeringvillage.com	487
Elsevier Scopus	Paper title, keywords, abstract	http://www.scopus.com	441

3.2.2. Search Terms

The search terms adopted in this study, which were used to match paper titles, keywords and abstracts in the performed automatic search, followed the guidelines proposed Kitchenhamet et al. [26]. To define the most relevant search terms for the search, the following strategies were performed:

• Definition of terms from research questions and study topics
• Identification of synonyms, plurals and related terms
• Adoption of the logical operator “OR” to incorporate synonyms
• Concatenate parameters using the logical operator “AND”
• Check the terms in the titles of papers, keywords and abstracts

The resulted search string is composed of synonyms and terms related to “infectious diseases” AND “transmission” AND “prevention” AND “technology”, as presented as follows:

• (“infectious disease” OR “infectious diseases” OR “COVID” OR “COVID-19” OR “SARS-CoV-2”) AND (“spread” OR “transmission” OR “propagation”) AND (“prevention” OR “prevent”) AND (“technology” OR “information technology”)

The terms related to “COVID-19” were also added since the inclusion of these terms was also of interest to the research.

3.2.3. Data Sources

As presented in Table 2, four electronic data sources were selected. Being cited by Kitchenhamet et al. [26] and Chen et al. [27] as relevant sources, the digital libraries were also selected because of the ease of access, the possibility of obtaining full text publications, and the fact that they are used for indexing journals and conference proceedings.

In order to allow a broader scope of the SLR, no limitations for the period of the publications were defined and only papers in English were selected for been considered the standard language of most international journal and conference proceedings.

It should be added that Google Scholar was not included as a data source because of the high possibility of returning inaccurate results and the considerable overlap with ACM and IEEE electronic databases [27]. Table A1 (Appendix A) presents the selected studies retrieved after the execution of the research string, and Table A2 (Appendix A) the mapping regarding the selected digital libraries.

3.3. Inclusion and Exclusion Criteria

In order to allow only relevant studies that met the objectives of the SLR to be returned after the execution of the search string, the inclusion and exclusion criteria presented in Table 3 were adopted.

Table 3. Adopted inclusion and exclusion criteria.

Inclusion Criteria

I1: Infectious diseases transmission/spread prevention-related works/approaches addressing the use of information technology solutions.

Exclusion Criteria

E1: Duplicate publications (including different references)

E2: Standards, models, industry standards

E3: Editorials, reports, position papers, keynotes, reviews, perspectives, surveys, summaries tutorials, books, courses or workshops, panel discussions

E4: Non-scientific publications

E5: Publications not related to infectious diseases transmission/spread prevention in humans

E6: Publications that do not cover “respiratory infectious diseases”, or whose approaches could not be applied in the primary ways of preventing transmission of these diseases

E7: Publications that do not have sufficient information to solidly answer the research questions

E8: Publications that do not meet the inclusion criteria

3.4. Study Selection and Data Extraction

Regarding the publication’s selection process, three stages were performed. In the first stage, the electronic bases were selected, the research string was executed in each digital library (on 30 June 2021), and the returned publications were compiled resulting in 1139 papers. Figure 3 presents the steps of the performed study selection and data extraction.

Figure 3. Publication’s selection process stages.

In the second stage of the publication’s selection process, duplicate papers were discarded, and the first filter was performed, where title, keywords and abstract were read and inclusion and exclusion criteria were applied, resulting in 293 candidate publications.

In the third stage, a second filter was performed where, after all papers were downloaded, a full-text reading of each article was performed applying the inclusion and exclusion criteria, which resulted in 219 papers. Finally, the proposed research questions were applied in all studies where data extraction and answers recording were performed based on the terms presented in Table 4 (results and discussions are presented in Section 4).

Table 4. Summary of items extracted from each study including research questions and quality criteria.

Objective	Item	Objective	Item
General Data	Title	RQ5	Context
	Author(s)	Q1	Objective of the Study
	Publication Year	Q2	Description of the Context
	Venue	Q3	Description of the Research Project
	Paper Summary	Q4	Analysis of the Data
RQ1	Approach	Q5	Conclusions Presentation
RQ2	Application Domain	Q6	Critical Analysis Description
RQ3	Adopted Support Mechanisms	Q7	Description of Limitations and Bias
RQ4	Level of Evidence

Table 4 presents the items adopted in the study in order to document the work, meet the research questions and evaluate the quality of the studies. The adopted quality criteria (Q1 to Q7) are described in Section 4.2 and the evidence levels, adopted in order to evaluate the maturity of the techniques described in the selected publications, are listed in Table 5.

Table 5. Levels of Evidence.

Level	Classification	Description
0	No evidence	No evidence was presented regarding evaluation or validation
1	Example or demonstration	Application description is provided with an example to aid its description
2	Specialists Notes	Qualitative or textual assessments are provided. Example: advantages and disadvantages contrasts/comparation
3	Experiment in laboratory	Results are reached from simulations with artificial data in real experiments. Evidence collection is performed formally or informally.
4	Empirical Investigation	Real context investigation of the behavior of the proposed approach
5	Strict analysis	Evaluation/validation of the study is performed using a formal methodology. Example: questions and variables definition for analysis after the application of the approach

Based on the work presented by Chen et al. [28], Table 5 presents the classification items in which the selected publications were validated or evaluated in order to identify the level of evidence of the solutions described.

During the execution of the publication’s selection process, the selected studies were presented to another two research studies where, after selection agreement, the publications were classified and categorized (divergences cases in paper selection and/or classification were solved after discussions in order to ensure the inclusion of relevant papers in this study).

4. Results and Discussions

In this section, results and discussions regarding the synthesis and analysis of the data extracted from the selected studies to answer the research questions are presented (including demographic data information).

4.1. Demographic Data

In order to provide an overview of the studies regarding publication venues, citation count, distribution by year of publication and countries, this section presents the demographic information on the selected studies. All included publications are listed in Table A1 (Appendix A).

Publication Venues and Citation Count

Information regarding publication venues and citations may be potentially useful for researchers interested in conducting research on a relevant topic, as well as partially show the impact of a study, the quality or the maturity of the proposed techniques. This is why it is also important to provide information on the distribution of the selected works on publication venues (as presented in Table A3—Appendix A), as well as an overview of the citation count.

Table A4 (Appendix A) presents an overview of the citation count of the selected publications sorted in descending order (information obtained from Google Scholar on 1 August 2021).

From the descending ordered list in Table A4, it is possible to identify the publications which were most cited where, comparing the 10 most cited studies with the application domain categories as presented in Section 4.3.1, it is possible to identify that six publications (S5, S101, S102, S116, S125, and S132) adopted approaches focusing on pandemic planning application domain (CD4). Tracking, surveillance and Contact tracing (CD6) application domain contained three studies (S114, S208, and S212), and Healthcare and Clinical management (CD1) one study (S104).

Regarding adopted support mechanisms, analyzing the table mapping presented in Section 4.3.2, from the 10 most cited studies, seven publications (S5, S101, S102, S104, S116, S125, and S132) adopted data and mathematical application related solutions, two studies (S114 and S212) adopted internet of things and hardware, and three studies adopted Software/Systems/Apps/Programing languages as support mechanisms (some studies used more than one category of support mechanism).

In Figure 4, the number of selected papers published per year is presented where, from 2020 onwards, a considerable increase in studies can be observed which, as mentioned before, is the period that the COVID-19 pandemic began. In other words, based on these data, it is possible to note that only after a global epidemic crisis emerged that the number of studies focused on the application of technology in primary ways of preventing the transmission of respiratory infectious diseases increased.

Figure 4. Number of selected papers published per year.

Table A5 (Appendix A) presents the distribution of the papers regarding the domain application categories (see Section 4.3.1) and the authors’ institution country. China, United States and India were the countries that presented the largest number of papers (48, 38 and 31, respectively), representing 53.42% of the publications selected in this study.

4.2. Study Quality Assessment

To perform the study quality assessment, the 219 studies were evaluated by the authors adopting the set of questions listed in Table 6, which were adopted and adjusted from the work presented by Chen et al. [28] and Dyba et al. [29]. Unlike the quality study proposed by the authors, the questions were not used to select the studies, but to validate the results.

Table 6. Quality assessment questions.

ID	Quality Assessment Question	Yes	Partially	No
Q1	Are the study objectives and goals clearly specified?	218 (99.5%)	1 (0.5%)	0 (0.0%)
Q2	Is the study context clearly defined?	113 (51.6%)	89 (40.6%)	17 (7.8%)
Q3	Does the research design support the objectives/goals of the study?	135 (61.6%)	71 (32.4%)	13 (5.9%)
Q4	Does the study have an adequate description of the analysis of the data?	96 (43.8%)	67 (30.6%)	56 (25.6%)
Q5	Does the study present a clear statement of the findings and provide enough data to support them?	79 (36.1%)	81 (37.0%)	59 (26.9%)
Q6	Do researchers critically examine potential bias and/or influence in the study?	3 (1.4%)	33 (15.1%)	183 (83.6%)
Q7	Study limitations are discussed explicitly?	51 (23.3%)	66 (30.1%)	102 (46.6%)

During the analysis of the studies (see Section 3.4), each of the questions was answered according to a ratio scale (“Yes,” “No” and “Partially”), in order to obtain information about the credibility of the results. As mentioned by Kitchenhamet et al. [26] and Chen et al. [28], the result of quality assessment may also reveal potential limitations of current research and guide future field studies.

As presented in Table 6, the answers regarding the objectives and goals of the studies (Q1) were all positive and, regarding the context definition (Q2), more than 92% of the studies defined them clearly.

Information regarding the nature and type of the organization, adopted software, team experience and research design to achieve the objectives (Q3) was provided in almost 94% of the studies, and about 75% presented an adequate description of the data analysis (Q4) as well as presented a clear statement of the findings (Q5).

However, the greatest concern was about the examination of bias or influence in the study (Q6), where about 83% did not provide enough (or any) information, as well as in discussions about the study limitations (Q7), where 46.6% of the papers did not present discussions.

4.3. Question Analysis

In the following sub-sections, the analysis and discussions of the research questions will be presented.

4.3.1. Available Approaches (RQ1) and Application Domains (RQ2)

During the studies analysis and information extraction phase, the proposed approaches were identified (descriptions presented in Table A1), and the application domains were recognized and grouped into six main categories. Table 7 presents the proposed application domain categories (the number of studies for each category is presented in parentheses after the category name).

Table 7. Application Domain Categories.

Category	Studies
CD1: Healthcare and Clinical management (59)	S4, S11, S14, S21, S24, S25, S27, S31, S37, S38, S39, S40, S56, S57, S58, S59, S60, S61, S64, S70, S77, S80, S88, S89, S92, S93, S95, S103, S104, S112, S123, S126, S127, S133, S137, S139, S140, S141, S150, S151, S157, S158, S160, S168, S171, S181, S182, S183, S187, S191, S192, S197, S202, S203, S205, S206, S207, S218, S219
CD2: Infection Testing/Screening (14)	S26, S35, S65, S82, S118, S122, S128, S148, S159, S163, S177, S185, S193, S198
CD3: Mask Detection (16)	S6, S8, S47, S63, S74, S75, S76, S106, S107, S108, S121, S142, S143, S144, S152, S178
CD4: Pandemic Planning (75)	S1, S2, S3, S5, S7, S15, S17, S23, S28, S29, S32, S36, S41, S43, S44, S45, S46, S48, S51, S53, S54, S55, S62, S66, S68, S71, S78, S81, S83, S85, S91, S96, S97, S98, S99, S100, S101, S102, S105, S111, S113, S116, S117, S120, S124, S125, S129, S130, S132, S138, S145, S146, S147, S153, S155, S156, S161, S164, S165, S169, S170, S173, S180, S186, S188, S189, S190, S199, S200, S201, S204, S209, S211, S215, S216
CD5: Quarantine/isolation/containment/social distancing (24)	S10, S16, S19, S22, S30, S50, S52, S69, S79, S84, S90, S109, S110, S119, S131, S135, S167, S172, S174, S179, S184, S194, S195, S213
CD6: Tracking, surveillance, and Contact tracing (31)	S9, S12, S13, S18, S20, S33, S34, S42, S49, S67, S72, S73, S86, S87, S94, S114, S115, S134, S136, S149, S154, S162, S166, S175, S176, S196, S208, S210, S212, S214, S217

The categories descriptions are presented below:

• Healthcare and clinical management (CD1): Category that covers approaches that seek to adopt technological solutions focusing on healthcare, case investigation, medical supplies, among others like the ones that seek to diagnose infected individuals, monitor clinical status, predict clinical outcomes, and provide capacity for telemedicine services, virtual care, and hygiene surveillance.
• Infection testing/screening (CD2): A category that covers approaches focusing on screening/testing individuals for diseases, either assessing for signs of disease in an apparently asymptomatic population, for example, or adopting technology with medical procedures to confirm the diagnosis in individuals.
• Mask detection (CD3): Covers approaches that adopt information technology solutions aiming to detect people who are (or are not) using the protective mask.
• Pandemic Planning (CD4): Covers approaches that aim at the identification/obtainment of new information that can be used or contribute to the prevention and/or control of transmission of infectious diseases, including anticipation of behaviors, transmissions, new outbreaks of epidemics, among others.
• Quarantine/isolation/containment/social distancing (CD5): Category of approaches involving the application of technology in order to restrict the spread of infection through the contribution to social distancing, containment or isolation of indeciduous; for example, monitoring quarantine patients, restricting social contact using global positioning systems or mobile phone applications, among others.
• Tracking, surveillance, and Contact tracing (CD6): Include approaches that aim at the identification, tracking or tracing of individuals who might have come into contact with an infected person in order to tracks viral spread; for example, monitors the spread of infection across locations, or to prevent onward transmission by alerting those who came in contact with the positive case.

Of the selected studies, we can highlight that most of the works (about 61.19%) focused on pandemic planning (CD4) and healthcare and clinical management (CD1) related application domains, obtaining a percentage of about 34.25% for the CD4 category, and 26.94% for the CD1 category.

Regarding tracking, surveillance, and contact tracing (CD6)-related approaches, 31 studies (14.15%) were found and, with respect to Quarantine/isolation/containment/social distancing (CD5) application domain, 24 studies (10.96%) were identified.

Mask detection (CD3) and infection testing/screening (CD2)-related application domains were the ones with the fewest studies with only 7.31% and 6.39%, respectively.

Figure 5 presents the quantitative distribution of studies by application domain category regarding the period of publication, where it is possible to observe that studies related to the categories CD3 and CD5 only appeared after 2020 and it was only in 2019 that studies related to the CD2 category emerged in publications.

Figure 5. Distribution of studies by application domain and year of publication.

It should be added that the studies were categorized regarding application domains that received greater prominence; although, in some cases, some fell into more than one category. Studies with more than one application domain were: S10 and S693 (CD3 and CD5); S53 and S79 (CD5 and CD6); S82 and S185 (CD2 and CD3); S109 (CD2 and CD5); and S121 (CD1 and CD3).

4.3.2. Adopted Support Mechanisms (RQ3)

After the identification of application domain categories of the selected studies, another important step would be the identification and categorization of the adopted/proposed support mechanisms. The mapping of the identified mechanisms according to the proposed categories (and sub-categories) is presented in Table 8 (the number of studies for each support mechanism is presented in parentheses).

Table 8. Support Mechanisms.

Category	Sub-Category	Support Mechanism	Studies
CS1: Data and Mathematical Application Related Solutions	Algorithms, Theories, Mathematical/Statistical Models	Bootstrap Method (1)	S170
		Dijkstra Algorithm (1)	S16
		Discrete Fourier Transform (DFT) model (1)	S65
		General Algorithms, mathematical models/equations (12)	S9, S34, S49, S52, S69, S94, S102, S130, S156, S162, S210, S214
		K-nearest Neighbor Algorithm, Nearest-neighbour distance (2)	S13, S167
		Markov Model, Spatial Temporal Method, Graph Theory, NHPP, Monte Carlo (19)	S17, S18, S23, S34, S36, S44, S51, S53, S62, S85, S99, S101, S111, S116, S129, S149, S154, S215, S216
		Multi-agent (Model/simulation), Equation-based model (13)	S2, S3, S4, S5, S7, S91, S96, S99, S100, S125, S132, S160, S211
		Multiple Signal Classification (MUSIC) Algorithm (1)	S128
		Optimal Control Theory (1)	S161
		Regression models, Short-term Prediction, RMSE, MAE (5)	S15, S68, S128, S147, S169
		SEIR model, Grey Prediction Model, DSGE Algorithm, SLIR, SIS, SIR (24)	S21, S32, S55, S66, S91, S96, S97, S98, S116, S117, S129, S130, S145, S146, S153, S155, S161, S189, S190, S199, S200, S201, S209, S215
		Self-Propelled Entity Dynamics (SPED) model, LDS—Low Discrepancy Sequence (1)	S164
	Artificial intelligence, Deep learning, Machine Learning, Big Data and Data mining	Big Data (5)	S1, S42, S81, S123, S147
		Decision Tree, Regression Tree, CART (5)	S40, S41, S46, S123, S180
		DBSCAN—Density-Based Spatial Clustering of Applications with Noise (1)	S172
		Fuzzy Logic (3)	S125, S171, S192
		Heterogeneous Diffusion Network (1)	S154
		K-means (5)	S33, S44, S180, S214, S217
		LLA—Lexical Link Analysis (1)	S138
		Logistic Regression (10)	S27, S45, S46, S55, S104, S149, S154, S169, S170, S177
		Maximum Entropy Model (1)	S105
		Naive Bayes (2)	S27, S41
		NLP—Natural Language Processing (3)	S103, S137, S158
		Neural network (CNN, MTCNN, MobileNet, others), Feature Enhancement Module (FEM), Spatial Separable Convolution, SSD (41)	S6, S8, S10, S20, S40, S41, S45, S47, S48, S55, S57, S58, S59, S60, S61, S63, S64, S65, S70, S74, S75, S76, S88, S90, S106, S107, S108, S109, S112, S120, S121, S142, S143, S144, S152, S178, S182, S184, S185, S201, S206
		Random Forest, iForest (5)	S27, S33, S40, S41, S46
		Support Vector Machine (8)	S3, S27, S40, S41, S45, S46, S61, S104
		Vector Space Model (2)	S123, S141
CS2: Software/Systems/Apps/Programing languages	Market Software/Platform (Proprietary or Free/Open Source)	Android Studio (1)	S71, S135
		AnyLogic, Django Framework (1)	S2
		ArcGIS (3)	S28, S62, S208
		Autodesk Revit/Meshmixer, Rhino3D, AutoCAD, Grasshopper (4)	S83, S110, S188, S199
		AWS—Amazon Web Services (e.g., software, and load Balancer, elastic container, lambda, Greengrass, others) (3)	S12, S62, S137
		Bootstrap, Adobe Photoshop (1)	S204
		Business Model Canvas (BMC), Service Blueprint (1)	S124
		Ethereum (1)	S67
		Google Cloud Platform (2)	S62, S68
		Hadoop (1)	S42
		Hyperledger Fabric (1)	S117
		IOTA Tangle Platform (1)	S71
		Kibana (Elasticsearch) (1)	S62
		MATLAB (1)	S113
		Microsoft Azure Cloud (2)	S70, S78
		NetLogo (1)	S4
		NLTK—Natural Language Toolkit (1)	S158
		Node.js (2)	S62, S137
		Node-RED and Grafana (1)	S136
		OpenCV (2)	S109, S151
		Ultimaker Cura (1)	S110
		Unity Platform (e.g., WebGL, 3D) (3)	S83, S131, S179
		WeChat, WhatsApp, WhatsApp Bot (4)	S93, S109, S111, S117
		Wireshark Dumpcap (1)	S195
		Zoom Platform (1)	S207
	Mobile, Desktop, WEB or Cloud Application/Framework proposed as study contributions	Cloud Application (3)	S12, S31, S35
		Desktop Application (4)	S51, S83, S96, S113
		Mobile Application (24)	S12, S20, S33, S63, S70, S72, S73, S84, S92, S111, S115, S119, S123, S135, S150, S151, S165, S166, S175, S192, S194, S203, S204, S214
		Web Application/Framework (21)	S2, S14, S28, S29, S37, S41, S42, S62, S67, S68, S78, S82, S93, S123, S131, S158, S165, S166, S176, S204, S214
	Programming Languages	C#, C++ (2)	S83, S96
		Java (J2EE, J2ME, JNI, Hibernate) (5)	S14, S28, S37, S73, S92
		JavaScript Libraries/ API (e.g., jQuery, ReactJS, AJAX, Google Web Toolkits, Google Maps) (9)	S2, S67, S28, S29, S72, S151, S166, S196, S204
		PHP (2)	S166, S204
		Python (6)	S2, S83, S96, S158, S199, S216
		Visual Basic (1)	S51
	Data Base Management System	Firebase (4)	S71, S73, S92, S194
		Influx DB (1)	S136
		MongoDB (2)	S2, S42
		MYSQL (3)	S78, S123, S204
		MS Access (1)	S51
		Oracle (1)	S29
		PostgreSQL (2)	S37, S208
		Neo4j (1)	S12
		SQLite (1)	S83
CS3: Internet of Things and Hardware	*1	Wearable devices (e.g.,smartwatches, smartphones, smartbelt, and others) (17)	S19, S27, S40, S54, S56, S69, S70, S79, S111, S115, S139, S157, S163, S175, S181, S195, S213
	Sensors (mobile or fixed), Cameras, RFID (Radio Frequency Identification)	Cameras—photo and video (Fixed and mobile) (11)	S30, S74, S82, S106, S109, S118, S121, S122, S128, S193, S194
		Environment Sensors (e.g., Passive Infrared (PIR) Sensor, and others) (26)	S11, S16, S21, S30, S38, S47, S50, S54, S74, S77, S82, S89, S109, S114, S121, S122, S126, S127, S135, S136, S172, S177, S185, S191, S192, S193
		RFID (Radio Frequency Identification) devices (9)	S13, S27, S30, S35, S50, S182, S185, S192, S196
		Wearable and/or mobile body sensors (e.g., temperature, cough, oxygen, pressure, heart rate measurement) (14)	S21, S25, S26, S27, S31, S35, S44, S69, S86, S87, S118, S157, S159, S181
	Others (e.g., Printers, Spray, Chips, GPS/GSM/Bluetooth devices, WIFI routers, UV tech, WBAN, and others)	Bluetooth/WIFI/GPS/Wireless devices (e.g., module, routers, access point, receivers, SMS gateways, GPS chips, and others) (21)	S9, S18, S26, S31, S44, S50, S84, S86, S123, S124, S134, S136, S142, S148, S172, S175, S194, S196, S207, S212, S219
		Desktops, Laptops, and computer accessories (e.g., memory cards, processors, and other boards) (21)	S25, S31, S39, S47, S74, S86, S109, S114, S121, S123, S134, S136, S137, S139, S185, S191, S192, S197, S198, S207, S219
		Printer and scan devices (3)	S110, S174, S218
		Spray/Dispenser devices (6)	S11, S82, S168, S191, S205, S219
		UV technology (e.g., UVC, UV Chip, UV Led, UV Light, UV ray) (7)	S11, S24, S38, S95, S127, S133, S148
	Robot/Drones	Robot/Drones/Unmanned Aerial Vehicles (UAV) (14)	S18, S22, S43, S80, S90, S127, S140, S167, S173, S183, S187, S198, S205, S218
CS4: Blockchain	*1	Blockchain (7)	S43, S67, S71, S117, S162, S173, S186

*¹ Fields with “*” have no value.

Finishing the studies analysis and extraction of the support mechanisms information, four general categories (and nine sub-categories) were identified and adopted in order to group the selected studies. It should be added that it is common that some studies adopted more than one support mechanism, so they were included in more than one category.

Regarding studies that adopted algorithms, theories, mathematical/statistical models, a greater tendency was observed in the application of SEIR model, Gray Prediction Model, DSGE Algorithm, SLIR, SIS, SIR (24 studies), followed by Markov Model, Spatial Temporal Method, Graph Theory, NHPP, and Monte Carlo (19 studies). This grouping of studies was due to the fact that these studies used most of these mathematical models in their approaches or evaluations.

Regarding studies that applied artificial intelligence, deep learning, machine learning, big data and data mining, the greatest trend of the approaches applications was in neural network, feature enhancement module (FEM), spatial separable convolution, and SSD with a total of 41 studies.

Most of the studies that adopted mathematical models or machine learning focused on prediction for decision-making (or simulations) regarding present or future actions in pandemics, such as vaccination, social isolation, disease transmission and control, among others (see Table A1 for descriptions of the proposed approaches). In addition to assistance in increasing the capacity and accuracy of identification of infectious diseases cases and their expansion, artificial intelligence, machine learning, and big data also received a lot of attention from studies that focused on screening, contact tracing, and diagnosis of infectious diseases.

Regarding support mechanisms related to software/systems/apps/programing languages (CS2) category (with 66 papers), studies were included that used existing market paid/free software or developed software/mobile apps as research contributions (programming languages and database management systems were also included).

With regard to studies that proposed Mobile, Desktop, WEB or Cloud applications or frameworks (see Table A1 for details of the approaches), for the case of Web and Cloud applications, there was a trend towards solutions focused on support for decision making, disease surveillance, and issuing alerts of critical areas with higher incidences of disease cases.

For Mobile applications, there was a greater tendence in contact tracing, social distancing, and body-symptoms detection/analysis. Regarding Desktop application, the proposed solutions focused on the containment of infectious disease outbreaks using geographical information with mathematical methods/models’ application.

Although many studies have proposed software or mobile apps as contributions in their approaches, there was a lack of detailed information regarding the adopted programming languages and database management systems (see Table 8). These situations usually occurred more in studies that also proposed the application of algorithms, theories, mathematical/statistical models and/or artificial intelligence, deep learning, machine learning, big data and data mining (CS2).

Regarding mobile applications, despite the lack of detail in the adopted programming languages or database management systems, most studies reported that the applications were developed with versions available for smartphones with Android and IoS (Apple), except for studies S72, S73, S92, S150, S151, and S194.

Regarding the adoption of internet of things and hardware (category CS3), most studies adopted sensors like environment (26 studies) and body (14 studies) sensors. In the case of environment sensors, there was a greater trend in the adoption for screening potential infectious diseases carriers from distance (e.g., temperature measurement/scanning), local position or movement measurement, automation of devices for hygiene (e.g., hand hygiene), and monitoring occupancy of places (e.g., monitoring elevator occupancy using a Passive Infrared (PIR) sensor).

Regarding studies that adopted wearable and/or mobile body sensors (14 studies), there was a greater tendence towards the verification of a Pearson’s heart rate, temperature, blood oxygen level and blood pressure. Video and photo cameras (11 studies) were used in devices such as smartphones, tablets, smart gates, cabins, doors, among other places, in order to collect video and/or image for purposes such as detecting use of masks, population monitoring, among others.

To enable communication between various IoT devices (e.g., mobile devices communicating among each other or with a centralized access point or server), devices with bluetooth/wifi/wireless technology (21 studies) or with RFID technology (9 studies) were adopted.

Desktops, laptops and computer accessories such as memory cards, processors, and other boards (e.g., Raspberry pi, Arduino Uno, BeagleBoard-Xm, AT89S52 microcontroller, and others) have been adopted in 21 studies as accessories for the proposed approaches (see Table A1 for descriptions of the proposed approaches) or to be adopted with others support mechanisms, as well as printers and scan devices (3 studies). Manufacturer names and device versions were not considered for the mapping of the identified hardware.

Along with sensors and cameras, drones and/or unmanned aerial vehicles were adopted in 14 studies, with a greater tendency in population monitoring for purposes such as information collection of social distancing and contact tracing.

Regarding surface disinfection in private or public spaces (including hand hygiene), automatic sprays and robots were adopted (see Table A1 for descriptions of the proposed approaches) and, with regard to studies that adopted Blockchain (7 studies), there was also a greater concern with the security in the sending and storage of the user data.

In Figure 6, the quantitative distribution of the selected papers by sub-categories of support mechanisms is presented.

Figure 6. Quantitative distribution of the studies by support mechanisms.

Of the selected studies, there was a greater tendency (139 studies) to adopt data and mathematical application related solutions (CS1), where most studies focused on the application of algorithms mathematical/statistical models (70 studies), and artificial intelligence, deep learning, machine learning, big data and data mining (69 studies).

The second category that showed greater adherence (94 studies) regarding the use/proposal of support mechanisms was related to the general use of internet of things and hardware (CS3) such as, for example, drones, robots, smartphones, smartwatch, wearable devices, camera, sensors, among others.

4.3.3. Available Evidence (RQ4) and Context Application (RQ5)

In order to obtain information regarding available evidence (RQ4) and context application (RQ5), data extraction was performed based on the items presented in Table 4, also aiming to investigate the maturity of the selected studies.

In Table 9, the distribution of the studies regarding the evidence levels (as described in Table 5) and context application (academic or industrial) is presented (the number of studies for each context/evidence level is presented in parentheses).

Table 9. Study distribution regarding evidence level and context.

Evidence Level	Context
	Academic (121)	Industrial (98)
0: No evidence (13)	S3, S11, S20, S25, S53, S56, S70, S124, S126, S141, S150, S177, S181,	*1
1: Example or demonstration (36)	S1, S5, S9, S12, S26, S30, S31, S33, S36, S50, S67, S69, S79, S81, S82, S84, S87, S101, S107, S127, S133, S134, S140, S146, S163, S165, S166, S175, S179, S185, S191, S193, S197, S208, S213, S214,	*1
2: Specialists Notes (7)	S14, S47, S71, S72, S86, S174, S194,	*1
3: Experiment in laboratory (117)	S2, S7, S8, S10, S13, S16, S18, S34, S40, S43, S44, S49, S51, S52, S54, S59, S62, S64, S66, S75, S80, S89, S90, S91, S94, S96, S98, S99, S100, S102, S103, S104, S105, S109, S112, S115, S119, S121, S122, S129, S132, S142, S145, S151, S153, S156, S158, S159, S161, S162, S164, S171, S172, S173, S180, S182, S186, S190, S200, S206, S209, S210, S211, S215, S219	S4, S6, S17, S23, S37, S38, S39, S41, S60, S61, S65, S68, S74, S76, S77, S92, S95, S106, S108, S110, S111, S113, S116, S117, S118, S120, S123, S125, S128, S136, S138, S143, S147, S148, S152, S155, S160, S167, S170, S183, S184, S187, S188, S189, S195, S196, S198, S199, S202, S204, S216, S218
4: Empirical Investigation (24)	*1	S19, S22, S27, S28, S29, S32, S46, S55, S73, S83, S85, S88, S93, S130, S131, S135, S139, S154, S169, S176, S203, S205, S207, S212,
5: Strict analysis (22)	*1	S15, S21, S24, S35, S42, S45, S48, S57, S58, S63, S78, S97, S114, S137, S144, S149, S157, S168, S178, S192, S201, S217

*¹ Fields with “*” have no value.

From the distribution presented in Table 9, only 13 studies did not present any evidence and, regarding demonstrations or examples, 36 studies presented application descriptions in the academic context.

Most studies carried out experiments both in academic contexts (65 studies) with fictitious data, and in industrial contexts (52 studies) with data obtained from real case scenarios. Expert observations such as textual, qualitative or opinion evaluations were provided in eight studies limited to the academic context.

Adding the studies carried out in laboratories (117 papers) with the studies that carried out both empirical investigation (46 papers) and strict analysis (22 papers), it can be seen that studies aimed at preventing the transmission of infectious diseases through the use of technology have, for the most part, some evidence with tests (total of 163 papers).

As the studies analyzed in this systematic review focus on the application of technologies in the primary ways of preventing the transmission of respiratory infectious diseases, it is also interesting to identify the context in relation to the diseases. Thus, Table 10 presents the distribution of studies regarding diseases in which the use of technology was proposed to contribute to their prevention (the number of studies for each disease category is presented in parentheses).

Table 10. Papers distribution according to the studied diseases.

Diseases	Studies
Infectious diseases in general (using or not some disease as examples) (62)	S1, S3, S4, S5, S13, S21, S23, S28, S29, S36, S37, S42, S49, S53, S76, S85, S96, S101, S103, S105, S111, S113, S117, S118, S122, S123, S124, S126, S128, S129, S130, S132, S133, S146, S147, S149, S150, S151, S154, S156, S159, S160, S161, S162, S164, S165, S168, S172, S173, S177, S181, S182, S186, S187, S196, S197, S204, S208, S211, S213, S215, S219
COVID-19 (139)	S6, S8, S9, S10, S11, S12, S14, S15, S16, S18, S19, S20, S22, S24, S25, S26, S27, S30, S31, S32, S33, S34, S35, S38, S39, S40, S41, S43, S44, S45, S46, S47, S50, S52, S54, S55, S56, S57, S58, S59, S60, S61, S62, S63, S64, S65, S66, S67, S68, S69, S70, S71, S72, S73, S74, S75, S77, S78, S79, S80, S81, S82, S83, S84, S86, 87, S88, S89, S90, S92, S93, S94, S97, S98, S104, S106, S107, S108, S109, S110, S112, S119, S120, S121, S127, S131, S134, S135, S136, S137, S138, S139, S140, S141, S142, S143, S144, S145, S148, S152, S153, S157, S158, S163, S166, S167, S169, S170, S171, S174, S175, S176, S178, S179, S180, S183, S184, S185, S188, S189, S191, S192, S193, S194, S195, S198, S199, S200, S202, S203, S205, S206, S207, S209, S210, S214, 216, S217, S218
Influenza (H1N1, H5N1, and others) (17)	S2, S7, S17, S48, S51, S91, S99, S100, S102, S114, S115, S116, S125, S155, S190, S201, S212
Klebsiella pneumoniae (1)	S195

It should be added that the use of the term “Infectious diseases in general” was adopted to group studies that, despite using or not diseases as an example, have their approach generalized to infectious diseases in general, whose transmission can be reduced through primary forms of prevention, such as social distancing, mask usage, isolation and quarantine. Therefore, they were grouped separately from those studies whose approaches were proposed specifically for the diseases listed in Table 10.

As presented in Table 10, it is possible to identify a greater concentration of studies focused on the use of technology in the primary ways of preventing COVID-19 (139 studies), followed by infectious diseases in general (62 studies).

5. Research Implications and Limitations

As mentioned earlier, in order to obtain the greatest possible number of relevant studies in the systematic review, a comprehensive search was performed with automatic searches in digital libraries adopting terms related to the subject matter of the search. To contribute to the selection of relevant studies, inclusion and exclusion criteria were applied in two stages.

Regarding possible threats to the validity of this systematic review, we can mention the possibility of bias in the selection of publications and in data extraction where, due to the possibility of subjectivity in decision making, researchers’ bias can affect the results of the work.

To contribute to the reduction in bias in the selection of publications, the review protocol was initially developed by a researcher, and validated by two other researchers with extensive knowledge and experience in software engineering and in works related to systematic literature reviews. After completion of the review protocol, it was strictly followed.

The selection process of the studies was then conducted in three stages (see Section 3.4) to reduce the chances of exclusion of relevant studies, and to contribute to bias reduction. The study selection process was conducted by one researcher, and all included and excluded studies were examined by two other researchers (inclusion and exclusion motivations were recorded and disagreements were resolved through discussions).

The option of adopting only automatic searches in digital libraries was also chosen as a way to contribute to the reduction in bias in the search and selection of studies. The terms adopted in the searches were iteratively improved based on the evaluation searches and were carefully tested before the execution of the systematic review.

In order to reduce threats regarding data extraction inaccuracies, a data extraction form was created (as presented in Table 4) to enable a more consistent extraction and analysis of data to answer the proposed research questions of this systematic review.

To contribute to bias reduction, the data extraction procedure was performed by two researchers who executed the data extraction and verification of the selected studies performing a complete reading of the works and answering the research questions and the quality criteria according to Table 4. After recording and analyzing the extracted information, similar positions and conclusions were unified, and disagreements were resolved through discussions.

After obtaining the results, the third researcher validated them, acting as the final decision-maker for discussions when no agreement was made.

The quality assessment of the papers also contributed to the increase in accuracy and precision of data extraction process, giving more credibility to the fact that extracted data comes from reliable studies.

The major limitations of the studies are the limited number of selected sources (only four digital libraries), and the fact that only automatic searches were performed, where the use of keywords may not cover all studies that use technology in the primary forms of prevention of transmission of infectious diseases.

This may happen because the searched terms may not be explicit in the title, keywords, or abstracts of the studies, as well as the string search, which may not contain the full set of terms required to obtain all the relevant works available in the digital libraries.

Another limitation that should be added is that, although the guidelines suggested by Kitchenhamet et al. [26] were adopted in this systematic review, instead of having the participation of a group of researchers in the data extraction, only three participated. Although the performance of analysis activity by a single researcher is common in many studies, it can also lead to questions regarding the description or classification of the collected data. To contribute to the reduction in bias in this scenario, three researchers participated in the study.

6. Conclusions

In the race against the spread of transmissible infectious diseases, there has been a growing interest in the use of technological solutions in the primary ways of preventing the transmission of these diseases.

Due to the importance of the subject, mainly due to its economic and social impact to society, it is equally important to systematically identify, analyze and document what is being carried out and studied, discovering new directions, as well as having a better understanding of how technologies are being used in this scenario, its main objectives, adopted support mechanisms, level of evidence reported, gaps that need to be deepened in research, as well as to organize the knowledge to support the technological transition.

For this purpose, this work presented the design, execution, and results of a comprehensive systematic literature review of relevant studies on information technology applications in the primary ways of prevention of new infectious diseases transmission.

Based on the findings, it was possible to identify issues relevant to interested researchers and practitioners, as well as contributing to the availability of an evidence-based guide to select appropriate technologies, approaches, solutions, or support mechanisms based on the different needs or scenarios.

From the results presented in this literature review, it was possible to identify six application domain categories of the selected studies in which there was a greater trend in studies related to pandemic planning and, among the support mechanisms adopted, data and mathematical application related solutions received greater attention.

From the mapping of support mechanisms carried out, it was also possible to identify a trend towards the application of artificial intelligence, deep learning, and machine learning technologies in primary ways of preventing transmission of respiratory infectious diseases. Thus, a thorough analysis and comparison of these algorithms (e.g., analysis of the success rate of the algorithms) is proposed as future works.

Regarding quality assessment analysis, most of the studies did not provide enough (or any) information about the examination of bias or influence in the study, as well as in discussions about the study limitations. Regarding available evidence, most of the studies presented some evidence (with tests).

From the findings, a greater tendency of studies focused on the use of technology in the primary ways of preventing COVID-19 was identified, followed by infectious diseases in general.

While it cannot be said that the study is exhaustive, it is believed to be a useful resource for interested researchers and practitioners regarding the use of technological solutions in the primary ways of preventing the transmission of infectious diseases.

For future works, it is recommended to expand the scope of this study with manual searches, including with searches in the references of the selected studies (through snowball techniques).