**3.2 Strategies used in search engines**

To continue the research procedure and obtain the answers to the questions, the keywords listed below were defined:


Based on these keywords, the following search string was defined at first: "serious games" OR "serious games" AND "RCP" OR "cardiopulmonary resuscitation" As the search return appeared to be too broad (187 items in the Google Scholar and 221 in Capes Journals), in addition to presenting several articles not related to the research interest, the string was redone.

*The Use of Serious Games for Learning Cardiopulmonary Resuscitation Procedures: A Systematic… DOI: http://dx.doi.org/10.5772/intechopen.102399*

The new string took the following form: ("serious games" OR "serious games") AND ("RCP" OR "cardiopulmonary resuscitation" OR "Cardiopulmonary Resuscitation" OR "CPR") AND ("SBV" OR "basic life support" OR "BLS" OR "Basic life support"), returning 110 items in Google Scholar.

On the Capes Journal Portal, the string had to be adapted, as it presented incompatibility with the previous format, obtaining the following form: (serious games) OR (serious games) AND CPR OR BLS OR ((Cardiopulmonary Resuscitation) AND (basic life support)), returning 176 items. After some observations in the listed articles, it was noticed that certain terms should be excluded from the string, as they have the same acronym of BSL and CPR, they were integrated in the search result.

Then, after a new adjustment to the string, this format was agreed to: (serious games) OR (serious games) AND CPR OR BLS OR ((Cardiopulmonary Resuscitation) AND (NOT (Common-Pool Resource)) AND (NOT (biomedical laboratory) science))), with the option to return only peer-reviewed journals, which resulted in 76 items.

The databases that returned in the search performed within the Capes Portal were PubMed, Scopus, Advanced Technologies & Aerospace Database, Directory of Open Access Journals—DOAJ, Web of Science, SpringerLink, Sage Journals, ERIC, and ACM Digital Library.

A private search was also performed in the IEEE database, which returned 21 items with the adapted string: ("All Metadata": "serious games") AND ("All Metadata": "Cardiopulmonary Resuscitation") OR ("All Metadata": "CPR training") OR ("All Metadata": "basic life support").

Finally, in the SBGames repository, which are the Annals of the Symposium, the search was performed manually, having found four items that included the chosen keywords. And this result was manually entered into START. BIBTEX files were also generated with the results of each search performed in the databases (Portal da Capes, Google Academic, and IEEE) to be inserted in an automated way in the support tool for the management of systematic literature reviews.

Searches were carried out between February and July 2021, through the academic registration valid on the CAPES portal and the mapping database, extracted during this period, and the final selection of chosen studies was systematized through MS Excel spreadsheet, which later also contributes to the elaboration of the graphs for this work. Also, during this period, learning the technique of Systematic Mapping was contemplated, through the subject of Scientific Methodology, which was administered by Professor Viviane Santos, collaborator of this work.

#### **3.3 Inclusion and exclusion criteria**

This item was also developed based on Kitchenham and Charters [15] with the formulation of the following definitions below, presented in **Table 1**:

In order to exclude erroneous terms that would invalidate the search results within the consulted databases, this list of criteria was created. In particular, the terms Common-Pool Resource/Cascaded Pose Regression (CPR) and Biomedical Laboratory Science (BLS) were included because they have the same abbreviation as the terms Cardiopulmonary Resuscitation (CPR) and Basic Life Support (BLS), respectively.

#### **3.4 Criteria for selection of studies and identification of qualities**

To obtain the works that contained the object of study of interest to the researchers in this review, the criteria were as follows: (a) whether the study uses volunteers to test


#### **Table 1.**

*List of chosen criteria.*

the software's teaching efficiency; (b) whether the study is specifically about basic life support with a focus on CPR; and (c) if there are details on how the software was built, with information such as platform, the tool used, as well as additional hardware.

Two researchers from this study were delegated to do the peer review, in the final choice. In addition to the complete reading of the individual selected, the researchers discussed each article, with the aim of both being in agreement on the fulfillment of the established criteria.

**Figure 1**, which is the PRISMA flow above, shows all stages of the study selection process until reaching the definition of the chosen works, which were 35 and are listed in **Table 2**:

#### **Figure 1.**

*Chapter selection process (prepared by the authors).*


*The Use of Serious Games for Learning Cardiopulmonary Resuscitation Procedures: A Systematic… DOI: http://dx.doi.org/10.5772/intechopen.102399*

#### **Table 2.**

*Number of articles per publication year.*

### **3.5 Data extraction and synthesis**

The process of filling out the protocol generated an extensive file, of more than 100 pages, the result of data collection carried out with the START tool [16], cataloging information such as title, abstract, authors, keywords, journal, year, publication link, and others comments. After the complete reading, the comprehension of the selected works is also included.

A synthesized document was built in Microsoft Excel to solidify the information of interest to this study, found in the reading of each article. These were the items: (1) if there is the identification of the application of artificial intelligence techniques (yes or no); (2) year of publication of the material; (3) platform used (PC, mobile devices, virtual or augmented reality glasses and other peripherals); (4) existence of volunteers in the study (yes or no); and (5) country that carried out the survey. There

#### **Figure 2.** *Cloud of words by the title of selected works.*

was space for extra information such as type of peripheral, manufacturer, and tool/ language used to create the software.

Due to the use of a tool to support the management of systematic reviews (START), it was possible to obtain the word cloud graph according to the chosen studies, with a sampling of the words that were most present in the titles (**Figure 2**).

In the following section of results and analysis, it is possible to check the graphs that were generated with this information. Out of curiosity, it is worth mentioning that among the excluded works, even with the generated search string, there were themes a little out of contexts, such as a study on fictions such as "Boys Love" [51] and collective consciousness [52].

### **4. Results and analysis**

When analyzing the relationship between the origin of publications and quantity, **Figure 3** shows that most of the selected works are of European origin.

Another interesting observation is the representation by region, with 11 works from Europe (Denmark, Spain, Netherlands, England, Italy, Portugal, and Sweden), six in Asia (India, Japan, Pakistan, Thailand, and Taiwan), four in North America (Canada, United States), three in the Middle East (Iran and Turkey), two in Africa (South Africa and Algeria) and, finally, one work in Oceania (Australia), respectively. Brazil is the only representative of South America, with nine works highlighted.

In **Figure 4**, which shows the number of publications per year, it is possible to note that the years 2019 and 2020 were the most productive for the topic, with a total of 20 works, eight of which for mobile devices. Perhaps because studies on games are increasingly in evidence, accompanied by the popularization of smartphones, which favors and encourages the emergence of new applications. According to an annual study by the company NewZoo, released in May this year, the number of active users on this device specifically reached the amount of 3.8 billion around the world, with a growth of 6% compared to the previous year. And a 56% share of that audience is found in the Asia-Pacific region [53].

It was possible to observe that in **Figure 5**, referring to the platforms used in the selected studies, there is almost a tie between the option of implementation on PC and on mobile devices (smartphones and tablets). One of the chances that the PC is *The Use of Serious Games for Learning Cardiopulmonary Resuscitation Procedures: A Systematic… DOI: http://dx.doi.org/10.5772/intechopen.102399*

**Figure 4.** *Number of publications per year.*

**Figure 5.** *Platforms used in selected studies.*

among the most chosen is due to the convenience of generating builds in the default configuration of most game development tools. Mobile devices, on the other hand, follow slightly in the lead as they are the platform with the greatest ease of access among users, even game development tools do not offer build generation by default. And according to another study by NewZoo, the global gaming market is dominated by mobile devices, with 52% of revenue [54].

In third place, there are the Virtual Reality Glasses (VR), with nine studies, representing almost 21%. They appear to be gaining in popularity, particularly in the European, Asian, and North American regions. The models that appeared in this selection were HTC Vive [55], Rift Glasses, and Go Glasses [56]. There were also three studies with the MS HoloLens [57], which makes use of mixed reality. These are modern and very interesting technologies, but they have a high market value, increasing costs, and access for the vast majority of users.

Anyway, it's worth opening a parenthesis here to detail VR devices, head-mounted displays (HMDs). The HTC Vive was used in the works A17—Canada [32] and A22— Italy [25]. Both also made use of complementary peripherals, the HTC Trackers [58], to improve the experiment's effectiveness, as shown in **Figure 6** below. Just like the Unity3D software [60] was chosen for the development of the applications.

HTC Vive has a total of 70 sensors spread out in its set (glasses and controls), allowing the user to perform fluid and high-precision movements. However, for good functioning, the device needs to be calibrated for each new user, especially if there is a difference in height. It is classified as PC VR (tethered VR) as it is dependent on a rugged computer for its operation [61, 62]. It is still possible to expand the number of sensors, by inserting HTC Trackers in the set [58].

The model used in both studies (A17—**Figure 7**/A22—**Figure 8**) was the first version from 2016, which arose from a partnership between North American Valve and Taiwanese HTC. Currently, the manufacturer already has other models on the market, such as Vive Flow, Vive Pro, Vive Focus, and Vive Cosmos [55].

Following the details of the models found in the selection of studies, we have two variants of Facebook property: the Oculus Rift, which was used in study A23—India [42]; *The Use of Serious Games for Learning Cardiopulmonary Resuscitation Procedures: A Systematic… DOI: http://dx.doi.org/10.5772/intechopen.102399*

**Figure 7.** *HTC vive kit and trackers used on study A17 [32].*

#### **Figure 8.**

*Study A22 [25] uses the HTC vive kit with their software. Figure (A) is related to the measurement of the individual's carotid pulse. The figure (B) is related to cardiac compressions. Both on VR and on real life.*

and the Oculus Go, in study A34—Australia [27]. The Rift model was the manufacturer's debut product, having been launched in 2016. It is a tethered VR type, comes with a sensor called Constellation, which needs to be placed in front of the user to enable tracking of the position, both sitting and sitting foot. Optionally, it can be used together with two peripherals for inserting hands in the environment, called Oculus Touch. It also has native Xbox console controller support for Windows only [63]. In study A23, only the HMD was used, with no complementary accessories.

In contrast, the other model, the Oculus Go used in study A34, is classified as an all-in-one (standalone VR) type. It does not need a powerful computer to work, as it is self-sufficient: it contains all the hardware components required to work, without the need for additional accessories, such as the aforementioned trackers [62]. But it has hand controls (Oculus Go hand controllers) for handling menus and other interactions.

The Unreal Game Engine [64] was chosen for the development of the application in the A34 study, as it allows the use of textures, has editing capacity, great customization power, in addition to providing a final product with high visual and realistic quality [27]. According to Hillman (2019), the development environment in this software has support for the most important VR platforms, with a unified set of tools that uses consistent procedures for the interaction between components [65].

Finally, among the HMDs devices, we have the Microsoft HoloLens, having been used in three of the selected studies: A19 [23] and A21 [33], both of American origin (USA); and in the European A26—Italy [34]. This device was introduced to the market in March 2016. But it acts a little differently from its competitors, making use of augmented reality (AR), by projecting for the user through the glasses on the device, information, and 3D digital graphics, overlaid on the image viewed in the real world [57].

And here comes the concept of mixed reality (MxR). He appears to be somewhat confused at first, but Bekele (2021) presents a study that details the nuances between these definitions [66]. The HoloLens is also considered a wearable device, as it allows the user to use it as a complementary accessory: as smart glasses, being called Augmented Reality Smart Glasses (ARSGs) [67].

Studies A19 and A21 only mention the use of the device, without providing further details about its application. By contrast, study A26 details his experiment, serious game modeling, and application in detail. It is also the only one who mentions the tool used for development, which was Unity3D. A common detail for the three studies was the use of a mannequin to assist in immersing the simulation for the user.

Following the analysis of peripherals used in the selected studies, Microsoft's motion sensor, Kinect [68], remains tied with Arduino, with two incidences each. Probably because of the difficulty of access and the cost involved. To use Kinect, you must have an Xbox 360 console [69] or an XOne. It is also possible to perform implementations on the PC, but it has the extra cost of acquiring the device [43, 49].

Arduino [59] is hardware with the characteristic of open source and low cost to implement. Provides great freedom for the developer. In the studies mentioned, it was used together with VR, mobile, and PC devices [24, 25].

The Nintendo Wii Balance Board [70] was used in one studies of Japanese origin. The implementation was in conjunction with MS Kinect and the PC [43]. And finally, the option without platform registration/citation, with six incidences.

**Figure 9** clearly shows the number of studies that have the participation of volunteers in order to validate the experiment: 74% yes and 25% no. This was one of the criteria for selection of studies and identification of qualities, item (a) of Section 3.4 this work. Therefore, the importance of this criterion for validating the software developed for any study is noted.

*The Use of Serious Games for Learning Cardiopulmonary Resuscitation Procedures: A Systematic… DOI: http://dx.doi.org/10.5772/intechopen.102399*

**Figure 9.** *Presence of volunteers in the study.*

#### **4.1 Quality assessment result**

As listed in Section 3.4 (criteria for selection of studies and identification of qualities) of this chapter, three quality criteria were defined. The first refers to letter (a) if the study uses volunteers to test the efficiency of teaching the software. **Figure 6** shown in the previous item brings this answer and 26 articles make use of this guideline. The second refers to letter (b) if the study is specifically about basic life support with a focus on CPR. Yes, the 35 selected studies deal specifically with this theme, and only five of them are indirect, presenting different approaches, but still, the CPR theme is present. The third refers to letter (c) if there are details on how the software was built, with information such as platform, the tool used, as well as additional hardware. Yes, these data were included, and it is possible to check the information generated in the figures and graphs in this article.

Another interesting information raised was that eight studies [22, 24, 25, 32, 34, 44, 45, 50] explained in their texts the use of the Unity3D game engine [60], a wellknown tool as a game engine and quite popular among game developers.

There were also mentions of other technologies such as MS Visual Studio [43, 71] for programming in the C# language, the use of the Japanese RPG Maker VX game engine [37, 72], and the robust Unreal Game Engine [64], software owned by Epic Games.

#### **4.2 Critical comparisons**

Some articles presented simulators as a serious gaming resource. Define the difference between simulators and serious games, what is a simulator and what is a game. If it's a simulator, is it considered a game too?

Other articles addressed the competition of learning using online videos and serious games, where the effectiveness was greater when using videos, but volunteers reinforced that learning with games was more interesting. But games are software that uses multimedia resources (audio, video, and images), there is even a genre that applies this resource in the game: the visual novels. The popularization of games focused on RCP, which is of this genre, would not be more interesting for the target audience? [73].

Twenty-nine (29) works mentioned the platform used. A tie-in incidences between the PC and mobile devices was registered. What will be the ideal platform to implement serious health games? Each platform has its own advantages and

disadvantages, including in-game features and mechanics. Or would the focus be on reaching all of them? But the more cross-platform the game is, the higher production costs will be involved, which can make the creation of the software unfeasible.

It is worth mentioning that Pavkov et al. (2017) present a comparison between five game engines, aimed at the development of serious games. With the advantages and disadvantages of each, the study proposes a criterion in order to facilitate the choice of the tool by the future developer [74]. In contrast, Cowan and Kapralos (2017) argue that it is common for educators to need to hire well-skilled developers to assist in programming serious game projects [75]. The authors even recommend other game engines not identified here in this study, for those who aspire to develop games but lack programming knowledge.

Another interesting observation is that nine studies were registered that present games aimed at VR devices. It shows an interesting trend, but for developing countries with scarce financial resources for research (aka Brazil), unfortunately, this is a difficult trend to follow here. In addition to limiting the target audience too much, VR devices have a high market value compared to other platforms such as PC or smartphones, which makes production even more impossible.

On the other hand, studies that mentioned the use of Microsoft Kinect and Nintendo Balance Board devices (A24 and A25) may become meaningless, as both peripherals left the manufacturing line, no longer being marketed [76].

#### **4.3 Existing gaps**

Conducting this systematic mapping raised some interesting points to be mentioned. One was to note that there is no academically validated research regarding existing commercial implementations of serious CPR-focused games. A quick search of game app stores such as Steam for PC and Google Play for Android mobile devices reveals a plethora of health-oriented games (Heath Games). Some even had great commercial success, with millions of units sold.

It is remarkable the fact that a commercial game has a greater appeal for playfulness/entertainment, dispensing with realistic educational rigor, with concepts to be worked on and solidified. But it is an undeniable fact that they are much more popular among the general consumer public, compared to serious games, mostly with academic backgrounds.

Another interesting perception for future works would be a literary, integrative, or systematic review focused on some of the platforms identified in the research, such as only mobile devices, only consoles, VR devices, or PC.

It was also possible to notice the need for curation of games available in digital stores. In order to get a guide or perhaps a list of health game recommendations. Not just with a commercial view, but with academic validation.

Finally, it was noted the need for more systematic reviews and mappings within the area, not only focused on CPR, but on other aspects that Health Games may have examples.

### **5. Conclusion**

The selected works were of great value for the understanding of this study, of how a systematic mapping is developed, with all its detailed and careful process that must be carried out.

*The Use of Serious Games for Learning Cardiopulmonary Resuscitation Procedures: A Systematic… DOI: http://dx.doi.org/10.5772/intechopen.102399*

Next, the answers for each research question mentioned in item 3.1 follow:


To improve this work, the evaluation criteria must be revised, listing weights for each one, in order to have numerical data to define the quality of the selected articles. Review some criteria and perhaps define further details between them. Also expand the number of databases searched, for those who know how to find more jobs in the South American region.

It was also possible to notice that the area of the Serious Games focused on health is relatively new when observing the age of the relevant events in the area: SBGames will complete 20 editions by the end of this year, but only 5 years ago it has a track dedicated to the theme; SEGAH is reaching its 9th edition. This fact opens up great possibilities for studies and research that have not yet been carried out, as the field is relatively new.

It was also possible to note that there is still a lot of room for research using new technologies, such as VR glasses, Microsoft Kinect, which have only one study each. *Computer Game Development*
