Holo-BLSD was developed jointly by the SIMNOVA simulation center (Novara, Italy) and the Department of Computer Engineering of Politecnico di Torino, in collaboration with Logosnet’s e-REAL Immersive Simulation Labs in Lugano, Switzerland. The app uses Microsoft’s HoloLens device, a wearable headset for AR experiences, and meets the recent American Heart Association guidelines [2]. Holo-BLSD can be used in different real environments, and its contents can be adapted to where the system is being used: virtual elements can be placed and anchored in the desired position.

Since the majority of potential users are likely to have little or no experience with AR apps or the HoloLens, Holo-BLSD provides a specific training session to help users get acquainted with the system. During interaction training, all interaction types are introduced individually with detailed instructions that include vocal and visual clues.

After the interaction training, users can begin the BLS training. Holo-BLSD guides users step-by-step through the resuscitation procedure of an adult experiencing cardiac arrest. The simulated activities include the following: scene safety, in which any potential hazard should be removed; responsiveness check, performed by shaking and calling the victim; activation of local emergency medical services by using a public phone booth and interacting with a simulated operator; automated external defibrillator (AED) retrieval by asking a witness to find an AED; CPR, using a real manikin torso and superimposing a person’s full body using AR (Figure 1); use of AED, once available, by directing pad placement and delivery of shock, if recommended. All generated data are logged, and a feedback sheet can be generated, allowing trainees and instructors to intuitively identify errors and strengths of the performance, thus supporting debriefing sessions and enabling the creation of a library of training events.

Participants were recruited on a voluntary basis during a national simulation-based training event for medical residents [15]. The selection of subjects represents a convenience sample, and medical background was not a required inclusion criterion. Experiments took place at the SIMNOVA Simulation Center in Novara, Italy.

The experience was divided into three phases, in which volunteers (1) learned how to use the device through the interaction training session, (2) performed a BLS simulation using the Holo-BLSD app and a standard torso manikin, and (3) filled out the evaluation questionnaire.

Participants completed a survey of 61 questions divided across six categories (user input, system output, system usability, fidelity of simulation, immersivity, likeability), and graded on a scale from 1 (strongly disagree) to 5 (strongly agree). The complete questionnaire can be found in Multimedia Appendix 1.

The results are presented with relative percentages for each of the five possible grades and summarized as mean and standard deviation. Given the explorative nature of the study, we did not plan any statistical inference tests.

Feasibility was assessed using the user input scale (interaction with gaze and gestures); the system output scale (quality of the display and sound); and the system usability scale [16], which is a standardized tool to evaluate interfaces based on the ISO 9241-400 guidance on ergonomic factors.

The remaining scales of the survey (fidelity, immersion, and likeability) were designed to assess acceptability.

Written informed consent was obtained from all participants, and the study results were presented in aggregate with no identifiers. The study was conducted in accordance with the principles of the Declaration of Helsinki. Descriptive statistics were calculated using Microsoft Excel (Version 2003, Microsoft Corporation).

Responses to the feasibility scale are shown in Multimedia Appendices 2 and 3 and aggregated data for the system usability scale are presented in Table 2. The cognitive load required to operate the HoloLens was minimal (mean 1.77, SD 0.86) and no high physical effort was required (mean 1.19, SD 0.40). Users felt confident using the software (mean 3.62, SD 1.06). Voice (mean 4.48, SD 0.87), gaze (mean 4.12, SD 0.97), and gesture interaction (mean 3.84, SD 1.14) were evaluated positively. Words and symbols (mean 4.70, SD 0.56) and audio instructions (mean 4.43, SD 0.79) were easy to understand. Users were most critical of the quality of the display (mean 2.45, SD 1.47), although they rated it as appropriate for the function (mean 3.62, SD 1.17). Finally, users rated the system as easy to use (mean 4.00, SD 0.94).

Responses to the acceptability scale are summarized in Multimedia Appendices 4-6. The sensorial information provided by the AR gave participants the impression of physically being in the scenario (mean 3.52, SD 0.95). Users reported that the experience was pleasant (mean 4.13, SD 0.81) and enjoyable (mean 4.65, SD 0.57) and that the virtual contents were realistic (mean 3.74, SD 1.05). Users judged the system as capable of providing real benefit as a training tool (mean 4.22, SD 0.67).

Experts are paying increased attention to the realism and scenario design of CPR training. The physical features of manikins and simulators by themselves are insufficient in suspending learners’ disbelief and positively influencing learning outcomes [4]. Despite ongoing advances in resuscitation science, cardiac arrest survival rates remain suboptimal and the educational efficiency of caregivers is still critical, as highlighted in 2003 [17] and 2018 [3]. Education facilitated through technology has been identified as a strategy to improve the effectiveness of BLS training. We developed the Holo-BLSD app using Microsoft’s HoloLens technology as an AR self-instruction learning environment for training and assessment of CPR and AED use, by using high definition holograms to immerge trainees in realistic scenarios, and a standard low-cost torso manikin to deliver tactile feedback. In this pilot study, we measured the feasibility and acceptability of our first prototype of the training app.

An excessive cognitive load may impair participants’ perceptions and performance, decreasing attention and problem-solving skills [6], and physical effort and fatigue may reduce user enjoyment of the experience. Despite these conditions, users rated the system as easy to use and they judged the learning experience as pleasant and enjoyable. Volunteers reported that the mental effort required to operate the device was minimal. Similarly, users indicated that the head-mounted display was comfortable. Unlike other head-mounted displays, such as the Google Glass used in Chaballout’s study [6], no delay between the real and virtual environment was experienced.

Some users noted that ambient light affected the quality of the holograms and made visual instructions difficult to interpret. Users judged the lack of peripheral view, which is known to increase cognitive load [18], as a major limitation. The assembled environment was considered complex enough to be of use, allowing users to focus on the tasks without leaving them disoriented. It is worth noting that trainees found the virtual contents realistic, stating that the system could provide a real benefit as a training tool and help practitioners be more effective.

In this study, we presented preliminary evidence that validated the Holo-BLSD app as a BLS training tool. Additional studies comparing traditional instructor-led training with low and high-fidelity simulation or other models will be useful.

The limitations of the study are the small number of participants, which makes it difficult to draw clear conclusions about the benefit of the proposed technology on learning outcomes. Nevertheless, this study was intended to provide a proof of concept and to measure the feasibility and acceptability of AR technology in a life support simulation. To this end, positive user responses indicate that future studies are warranted.

AR is in the early stages of application within health care education, but it has enormous potential. In this study, we presented Holo-BLSD, an AR system for BLS training that offers realistic haptic feedback through a manikin and a virtual scenario that can be easily reconfigured to generate different situations, including extreme and dangerous ones. We found the proposed application to be feasible and acceptable as a tool for self-instruction training. The positive outcomes of this preliminary study make this prototype worthy of future testing.