Application of Augmented Reality Technology for Chest ECG Electrode Placement Practice
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe manuscript is an educational paper to assess the learning in the electrocardiography (ECG) chest placement procedure when augmented reality technology is applied. The system is composed of a smart mobile device equiped with a camera to run the augmented reality application and a mannequin where a set of virtual electrodes are placed during the training. Initially, using AR marker, the ribcage is displayed over the mannequin and the user must place the electrodes following the defined procedure. The placement of the electrodes is compared with the expected position of them in the virtual model. Iteratively, the user electrodes are marked as green (valid position) or red (wrong position). The study conducts an experiment with 62 participants, most of them females around 18 years old. Half of the participants follow a traditional learning while the rest use the AR system. The participants perfom a pre-test and a post-test to evaluate the previous knowledge and to determine the learning during the process. In addition, the authors get a measure of the accuracy in the ECG chest placement of each participant in their training to compare the outcomes of each group.
The proposal is well explained and clear, the references are also adequate, the method is sound. The results undoubtely confirm the initial hypothesis that the AR technology improves the learning outcomes with respect to a traditional methodology. However, this reviewer considers that extra information could be provided in the result section. In Figure 9, the performance during the trainig is illustrated with the mean mark per group in both the pre-test and the post-test. On the other hand, Figure 10 shows the performance during the training considering the mean error per electrode and per group. In both cases, including a boxplot in another figure or a table with the corresponding information, could be of help to identify the homogeneity of the behaviour of the participants.
Minor comment: - Figure 10. Title of the X axis, correct "Electordes".
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsPlease see the file attached.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis manuscript presents an AR application that can interact with smartphones for ECG chest electrode placement training. Experiments were done with control group (31 participants) and intervention group (31 participants). This manuscript is interesting, and its results can be considered that is a contribution in learning of ECG chest electrode placement. The presentation of the steps of the model is good and contributes to the understanding of the method. But, the comparison of this study with the previous studies in literature should be given as table.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for Authors Thank you for addressing most of the comments I have provided. I still have some observations about changes to make.Introduction:
As this is not a review on AR for medical training, the part on AR in education should be further arranged. I suggest not putting the table in the manuscript, but rather mention and comment the most relevant research in the field. This is also important as the paragraph starting at line 73 is not connected to the rest of the introduction.
Lines 56-57 and 62-63 mention two studies without actually explaining them.
Materials and Methods:
I suggest removing lines 158-161 and the title of the subsection 2.1.1.1, putting all the text in 2.1.1
Paragraph 2.1.2 explains some preliminary tests. However, I still don’t understand the difference between the tests explained in lines 208-219 and the those described in lines 219-226. Also, lines 233-234 are a repetition of the first part of the section and therefore should be removed.
Regarding the distance between the real position and the virtual electrodes, how was the measure taken? By hand? Or perhaps with a digital measuring instrument?
The year of the version of Unity is not sufficient for replicability. What is the exact version used? Was it a LTS one?
Results:
The use of paired and unpaired does not account for normal distribution, but rather for having data belonging to two different populations. A normality test must be performed to use a parametric test.
Table 4 and 5 have a wrong caption, and Figure 10 has not been changed and therefore the actual p-values have not been added. As no multiple comparisons correction has been performed, then knowing the actual p values is crucial to understand if they actually reach significance considering that 6 t tests have been performed.
The authors report that in their revised version, they will “include both sets of statistics—mean and standard deviation for clarity and median with ranges to provide a comprehensive overview of the satisfaction scores”; however I could not see these changes in the manuscript
Discussion:
I acknowledge that the discussion has improved. However, I strongly suggest to avoid a list of comments, but rather try to link the comments together, also supporting them by literature references.
In my previous comment, I suggested adding a part in the discussion about other studies using AR for training and having similar (or dissimilar) results and to comment on that. The author responded to this comment reporting a study on AR for ECG chest lead training using an HMD, that was already listed in the introduction.
I think I did not make myself clear enough on this point: the discussion usually includes a comment on the results and how they are similar or different to other existing studies. As many groups use AR for medical training, it would be interesting to compare the present results with existing one (see for instance: Ricci, S. et al. “Viewpoint: Virtual and Augmented Reality in Basic and Advanced Life Support Training.” JMIR serious games 2022; Boonbrahm et al. Interactive Marker-based Augmented Reality for CPR Training. IJTech 2019 Balian S, et al. Feasibility of an augmented reality cardiopulmonary resuscitation training system for health care providers. Heliyon 2019; Leary M, et al. A Pilot Study of CPR Quality Comparing an Augmented Reality Application vs. a Standard Audio-Visual Feedback Manikin. Front. Digit. Health 2020; Strada F, et al. Holo-BLSD – A Holographic Tool for Self-training and Self-Evaluation of Emergency Response Skills. IEEE Trans. Emerg. Topics Comput 2021; Drummond D, et al. Google Glass for residents dealing with pediatric cardiopulmonary arrest: A randomized, controlled, simulation-based study. Pediatr Crit Care Med 2017)
Conclusions:
Correcting for multiple comparisons is not a future development, but rather a correct way of statistically analyze the data. Perhaps providing a scoring system and feedback would be useful to enhance the usability of the system.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
