A New Era in Stress Monitoring: A Review of Embedded Devices and Tools for Detecting Stress in the Workplace

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper reviews embedded devices and tools for stress detection, a topic of significant importance given the growing need to monitor stress in today's society. While the paper provides a valuable overview of existing literature, there are several key limitations that need to be addressed to strengthen the review:

1) The paper primarily summarizes previous studies without critical analysis, resulting in a superficial review. While the inclusion of sensors and signals in Table 1 is good, the paper should go further by analyzing and comparing different implementations, data processing techniques, and analysis methods. Additionally, it is important to discuss the challenges associated with these implementations and compare the accuracy and effectiveness of various methods to provide  insights.

2) The paper suffers from organizational issues, particularly in the related studies section, which may confuse readers. A well-structured review paper should have clear, logically organized sections.

3) This review includes only 14 papers listed in Table 1, and the total number of studies reviewed is unclear. This raises concerns about whether the paper has covered a sufficient number of studies to provide a comprehensive overview of the field.

Comments on the Quality of English Language

The writing part of this paper is fine, but it could be improved by removing some repetitive sentences.  

Author Response

This paper reviews embedded devices and tools for stress detection, a topic of significant importance given the growing need to monitor stress in today's society. While the paper provides a valuable overview of existing literature, there are several key limitations that need to be addressed to strengthen the review:

1) The paper primarily summarizes previous studies without critical analysis, resulting in a superficial review. While the inclusion of sensors and signals in Table 1 is good, the paper should go further by analyzing and comparing different implementations, data processing techniques, and analysis methods. Additionally, it is important to discuss the challenges associated with these implementations and compare the accuracy and effectiveness of various methods to provide  insights.

We want to thank the reviewer for thoughtful feedback. In response to your suggestions, we have made significant revisions to the manuscript. We expanded the scope by providing a comprehensive overview of all studies mentioned in Table 1 and conducted a deeper analysis of the hardware platforms involved into Section 5.1.1 (highlighted red). Additionally, we included two studies: Leone et al., which proposed a framework for monitoring mental load using wearable devices, and Rescio et al., which developed an unsupervised learning framework for stress detection with wearable devices. We have recreated Table 1 to include key details such as the components used, sensing capabilities, data transmission methods, sampling rates, and the main features of the hardware. Furthermore, we performed a comparative analysis of these studies to highlight their distinctions and contributions.

Additionally, we developed a new table in the section 5.2.1 Advanced Techniques in Neural Network-Based Stress Detection (highlighted red) to streamline this section. In this table, we addressed various neural network techniques, the data used, and the accuracy of the models. A comparative analysis was also performed to offer greater insight into these methods. Furthermore, we included studies involving EOG signals to provide a more comprehensive view of stress detection techniques.

We believe that these revisions provide a more detailed and in-depth analysis, adequately addressing the concerns you raised. Thank you again for your valuable input.

2) The paper suffers from organizational issues, particularly in the related studies section, which may confuse readers. A well-structured review paper should have clear, logically organized sections.

We want to thank the reviewer for valuable feedback. Based on your advice, we made significant changes to improve the organization of the manuscript. We created a new section titled Comprehensive Overview of Related Studies and Their Key Contributions, where we discuss all the relevant studies (Section 5 highlighted red). This section is divided into two clear subsections: Hardware Platforms for Stress Detection and Advanced Techniques in Neural Network-Based Stress Detection.

In the Hardware Platforms subsection, we included a specific sub-subsection called Comparative Overview of Hardware Platforms for Stress Detection to provide a more structured analysis. Similarly, in the Neural Network subsection, we added a sub-subsection titled Neural Network Techniques for Stress Detection to better organize and clarify the discussion of these methods.

 

We believe these revisions address the organizational concerns you highlighted, making the paper clearer and more logically structured. Thank you again for your helpful suggestions.

3) This review includes only 14 papers listed in Table 1, and the total number of studies reviewed is unclear. This raises concerns about whether the paper has covered a sufficient number of studies to provide a comprehensive overview of the field.

We want to thank the reviewer for raising this important concern. To clarify, our review analysed a total of 75 papers published between 1992 and 2024. The studies listed in Table 1 represent a selection of key studies that are most relevant to the specific focus of our review. These papers were chosen to highlight significant advancements and representative approaches within the field.

We have incorporated a paper selection analysis into Section 1.1 (highlighted red) of the manuscript. This section provides a overview of the literature search process, including the sources and keywords used, and the number of papers analysed. We hope this additional information addresses your concern about the coverage and scope of the studies reviewed.

Reviewer 2 Report

Comments and Suggestions for Authors

The paper is very interesting and I would like to congratulate the Authors for their contribution to the research field.

The paper is well structured and the Sections are well defined.

However, the Authors have to complete their state-of-the art description by adding a description to this manuscript regarding the contributions of stress coming from Electrooculography (EOG).

 

 

 

Author Response

The paper is very interesting and I would like to congratulate the Authors for their contribution to the research field.

The paper is well structured and the Sections are well defined.

However, the Authors have to complete their state-of-the art description by adding a description to this manuscript regarding the contributions of stress coming from Electrooculography (EOG).

We want to thank the reviewer for a valuable feedback. In response to your suggestions, we have updated the manuscript by adding two studies on EOG in Section 5.1 (highlighted red), which focuses on hardware systems. Additionally, we included two more studies in Section 5.2 (highlighted red), which addresses neural network techniques for stress detection. These updates improve the coverage of EOG applications and recent developments in neural network techniques. Thank you again for your insightful comments.

Reviewer 3 Report

Comments and Suggestions for Authors

The article has focused on stress detection techniques using embedded devices and tools. It has only pointed out desk jobs. It has included enough references to related works. Along with hardware, it described artificial intelligence algorithms. The work is interesting for stress monitoring. 

Figures 2 and 3 are left-aligned; these would be put in the center of the page. In Section 6, it discusses neural network-based systems. In this chapter, it would be good to present the results of other papers through a comparison table. 

At the end, it adds that computer peripherals are promising tools to identify the stress of their users, and they have a significant possibility of increasing productivity. This would be significant for researchers in this field.

Author Response

The article has focused on stress detection techniques using embedded devices and tools. It has only pointed out desk jobs. It has included enough references to related works. Along with hardware, it described artificial intelligence algorithms. The work is interesting for stress monitoring. 

Figures 2 and 3 are left-aligned; these would be put in the center of the page.

We want to thank the reviewer for the observation regarding the alignment of Figures 2 and 3. We followed the template guidelines provided by MDPI, which recommend left-aligning figures.

In Section 6, it discusses neural network-based systems. In this chapter, it would be good to present the results of other papers through a comparison table. 

We want to thank the reviewer for great suggestion regarding the presentation of results from other papers in Section 6. We have addressed this by integrating a comparison table in Section 5.2.1 (highlighted red), which provides a comprehensive overview of neural network-based systems discussed in our manuscript. Additionally, we included studies involving EOG signals to provide a more comprehensive view of stress detection techniques. This table effectively summarizes and compares the key findings from relevant studies, as you recommended.

At the end, it adds that computer peripherals are promising tools to identify the stress of their users, and they have a significant possibility of increasing productivity. This would be significant for researchers in this field.

Reviewer 4 Report

Comments and Suggestions for Authors

The paper presents a review on the use of embedded sensor systems to detect human stress in workplace. I suggest the following revisions to improve the paper quality.

1) In Section 2 (Physiological stress) is stated that stress is regulated by two neuroendocrine systems: HPA and SAM. The working principle of HPA is presented in Fig. 2. However, the working principle of SAM is not adequately discussed.

2) In Fig. 2 there is a reference to CRF that I think is CRH, instead.

3) The authors presents the working principle of human stress (in section 2) and the most suitable biosignals to detect stress (in sections 3 and 4). Then, many sensor systems from literature are presented in sections 5 and 6. I feel the authors must include a new section (between section 4 and section 5) where it is discussed in general the hardware implementation of sensor systems for the measurement of the biosignals presented in section 4. This section, in particular, must focus on the used sensors for biosignal measurements, how the sensors are applied to the human body, the sampling rate, the power consumption and, in general, the authors must present advantages and disadvantages of the different approaches.

4) Line 340 “…. development board was used to drive the sensors”. I think “…. development board was used to acquire the sensors data” is more appropriate.

5) I suggest to include a table in section 7 (Discussion), where the performance of the systems from literature presented in sections 5 and 6 are compared in terms of performance, accuracy of stress detection, intrusiveness, cost, etc.

Author Response

The paper presents a review on the use of embedded sensor systems to detect human stress in workplace. I suggest the following revisions to improve the paper quality.

1) In Section 2 (Physiological stress) is stated that stress is regulated by two neuroendocrine systems: HPA and SAM. The working principle of HPA is presented in Fig. 2. However, the working principle of SAM is not adequately discussed.

We want to thank the reviewer for valuable feedback. We appreciate your observation regarding the need for a more detailed explanation of the SAM system. In response, we have provided a deeper explanation of the working principle of the SAM system (in Section 2 highlighted red). We believe this additional detail will offer a clearer understanding of how the SAM system functions alongside the HPA axis.

2) In Fig. 2 there is a reference to CRF that I think is CRH, instead.

We want to thank the reviewer for pointing out this mistake. We appreciate your attention to detail and have corrected the reference from CRF to CRH in Fig. 2. Thank you again for your careful examination.

3) The authors presents the working principle of human stress (in section 2) and the most suitable biosignals to detect stress (in sections 3 and 4). Then, many sensor systems from literature are presented in sections 5 and 6. I feel the authors must include a new section (between section 4 and section 5) where it is discussed in general the hardware implementation of sensor systems for the measurement of the biosignals presented in section 4. This section, in particular, must focus on the used sensors for biosignal measurements, how the sensors are applied to the human body, the sampling rate, the power consumption and, in general, the authors must present advantages and disadvantages of the different approaches.

We want to thank the reviewer for insightful feedback. In response to your suggestion, we have introduced a new sub-subsection 5.1.1, titled Comparative Analysis of Hardware Platforms for Stress Detection. We have recreated Table 1 to include key details such as the components used, sensing capabilities, data transmission methods, sampling rates, and the main features of the hardware. Furthermore, we performed a comparative analysis of these studies to highlight their distinctions and contributions.

This new section is located under the broader Hardware Platforms for Stress Detection within the comprehensive overview. We believe this addition addresses the need for a general discussion on hardware implementation and enhances the overall clarity of the manuscript. Thank you again for your valuable suggestions.

4) Line 340 “…. development board was used to drive the sensors”. I think “…. development board was used to acquire the sensors data” is more appropriate.

We want to thank the reviewer for the feedback. I appreciate the reviewer’s suggestion to use the phrase “development board was used to acquire the sensors’ data.” However, the term “drive the sensors” was chosen intentionally to convey that the STM32F446RE Nucleo-64 development board is responsible for managing and controlling the sensors, consistent with the terminology used in the original source (Valenti, Simone, et al. "Wearable multisensor ring-shaped probe for assessing stress and blood oxygenation: Design and preliminary measurements." Biosensors 13.4 (2023): 460.). In this context, “drive” effectively represents the concept of the development board actively interfacing with and controlling the sensors.

5) I suggest to include a table in section 7 (Discussion), where the performance of the systems from literature presented in sections 5 and 6 are compared in terms of performance, accuracy of stress detection, intrusiveness, cost, etc.

We want to thank the reviewer for the insightful feedback. We appreciate your suggestion regarding the inclusion of an additional table in Section 7 (Discussion). In response, we have reedited Table 1 and added new Table 2 to the manuscript. The first table provides a Comparative Overview of Hardware Platforms for Stress Detection, detailing aspects such as hardware components, sensing capabilities, data transmission, sampling rate, and key features. The second table offers a Comparative Overview of Neural Network Techniques for Stress Detection, which includes information on the neural network techniques, the physiological data used, and accuracy measures. Considering that our new tables already provide a detailed comparison of hardware platforms and neural network techniques, and that Section 5.1.1 thoroughly addresses the electronic parameters and Section 5.1.2 addresses the performance metrics, we believe that adding another table in the Discussion section would be redundant.

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed the previous concerns. The manuscript in its current form is appropriate for publication. 

Author Response

Thank you for your positive feedback and for taking the time to review our manuscript. We are pleased to hear that the revisions have addressed your concerns.

Reviewer 4 Report

Comments and Suggestions for Authors

The authors revised the paper according to the Reviewer comments. I think it can be published after minor revisions.

1) Line 643: there is a ")" that should not be present.

2) Lines 889-892: the sentence "The Bluetooth communication module includes a passive high-pass filter to eliminate baseline drift, a microcontroller with a 10-bit A/D converter for digitizing the PPG signal, and a Bluetooth module for data transmission" should be replaced with "The proposed system includes a passive high-pass filter to eliminate baseline drift, a microcontroller with a 10-bit A/D converter for digitizing the PPG signal, and a Bluetooth module for data transmission".

3) Some layout problems are present in the two tables (overlap of different words, etc.). This could be a problem occurred during the PDF generation from the Word file. Please, check it.

Author Response

The authors revised the paper according to the Reviewer comments. I think it can be published after minor revisions.

1) Line 643: there is a ")" that should not be present.

Thank you for your valuable feedback. You are correct that the ")" at line 643 should not be present, and we have removed it in the revised version of the manuscript. We appreciate your attention to detail.

2) Lines 889-892: the sentence "The Bluetooth communication module includes a passive high-pass filter to eliminate baseline drift, a microcontroller with a 10-bit A/D converter for digitizing the PPG signal, and a Bluetooth module for data transmission" should be replaced with "The proposed system includes a passive high-pass filter to eliminate baseline drift, a microcontroller with a 10-bit A/D converter for digitizing the PPG signal, and a Bluetooth module for data transmission".

Thank you for your insightful comment. You are absolutely correct. We have replaced the sentence on lines 889-892 as suggested.

3) Some layout problems are present in the two tables (overlap of different words, etc.). This could be a problem occurred during the PDF generation from the Word file. Please, check it.

Thank you for bringing this to our attention. We have carefully reviewed the tables and corrected the layout issues. The updated version of the manuscript ensures that the tables are now properly formatted.