Analysis and Design for a Wearable Single-Finger-Assistive Soft Robotic Device Allowing Flexion and Extension for Different Finger Sizes
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsWork focused on design process to conceptualize and validate a soft robot for patients requiring hand rehabilitation. The process is applied for developing an actuator for tendon-gliding exercises of the finger.
To do this, an FEA is performed for the response of bi-directional pnumatic actuators for different finger sizes to determien the joint forces. 2D and 3D models are considered.
If the contribution of the paper is truly about the design process, it would be good to see the process applied to more than one joint/ such as hand and arm, or hand and leg. Or, with another set of actuators that are not the Pnumatic Network actuators. It seems as though the paper's contribution are not the design process itself, but the analysis and resulting designs informed by the analysis.
The paper is quite long to reflect the large amount of work performed in producing the analysis and design. However, the paper's contribution gets lost amongst the individual pieces of work. I would recommend the authors split the current paper into more than 1, and have very focused and clear contributions.
Figure 2 is a good diagram to outline the method of the problem. It is a good clear diagram of the steps that were taken. It does not seem to be a contribution to the field. A more generic version of this process would be to 1) Identify Problem 2) Define and Measure 3)Develop Model of system, 4) Analyze Model, 5) Provide recommendations/ design based on Model Analysis
This is not a unique process, and should not be listed as the paper's contribution.
Work was performed using hand measurements of 70 males and 69 femails. It is not clear why "a subset" of measurements were taken adn used for the study, as opposed to the whole data set. Clarity around this would be appreciated.
The diagrams and tables in the paper are very clear, and a strong positive for the article.
It is good that literature is being referenced regarding virtual marker placement, surface-level measurements, and x-ray scans beign used. What's missing is why the authors choose these methods or choose to use these particular techniques as opposed to others that may be in the literature. Or, is there a more traditional way of determining points.
It is not clear what Figure 8 is. Is this a technical 3-View drawing of the pna? It is not clear what this is until one reads the sentence that references Figure 8. A photograph of an example PNA may help in this case.
Figures 9 and onwards seem to focus on the results related to the reduced order model and the full order model. If the paper's contribution is a comparison of these two, or demonstrating the benefits of this reduced order model, having the figures combined so that they could directly compare the reduced to full models would help demonstrate the differences achieved.
Figure 21 appears to show a good test-bench set-up for testing PNA's. Perhaps this should be presented in a methods section. If this test-setup is a RESULT of the methods applied, then this aspect is not clear
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors1) Please elaborate more on how to quantify the joint moment in Figure 7. Was it directly generate from the simulation, or it was calculate based on finger tip force. Either way the procedure of obtaining the joint torque can be further explained.
2) How did you quantify the tip force in Figure 11. There was lack of details explaining it.
3) How did you relate the spring force into individual tip force at MCP, PIP, and DIP joints? It is unclear how the spring force measured in Figure 20 relates to the finger joint torque and tip force.
4) Follow by (3), it is necessary to validate the accuracy of the simulation in Figure 7 with experiments, but it seems that the spring force does not have too much relationship to it.
Overall, the authors should consider re-design rigorous experiment to validate all the simulation results. It is not sufficient for publication if the conclusion of design process is mainly determined by computer simulation without knowing the accuracy of the model.
Comments on the Quality of English Language
N.A.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis paper presents a design process for a wearable finger motion assistive soft robot to achieve tendon-gliding exercises (TGE). The authors proposed a design pipeline considering different finger sizes, modeling of joint spring forces, conceptualization of the actuator, manufacturing and verification of the actuator’s performance. Even though the soft actuator being used is Pneumatic Network Actuators (PNA), which is commonly used in soft robotics, the design process and framework is interesting in the way that it considers individual differences and medical requirements, which could be useful in practical clinic application for this technology. The musculoskeletal analysis is especially thorough and convincing. Successful results demonstrated for TGE 1 and 2, whereas unsuccessful for TGE 3.
I have some questions regarding the implementation of this design flow:
# Conditions for requirement setting
–In Fig 7, two AE conditions are modeled. It seems like all the following requirements and experimental validation are based on AE = 0.01 under fully passive conditions. Please clarify this in the paper if needed.
# parameters considered in actuator model
–Length of the actuator vs length of the finger
It seems like the length of the actuator is longer than the length of the finger. Why is this specific length chosen in actuator design?
–Chamber width and number of chambers
It seems like the design of the chamber width and the number of the chambers are not explored. However, they could be a main factor affecting the bending angle of the actuator.
–Initial position of the actuator
It seems that the initial position (on the back of the hand) of the actuator needs to be different for five different TGEs. How would this be achieved in actual wearable conditions whereas the users would be expected to wear one device to achieve all five TGEs?
# Verification of different finger sizes
The authors claim that the framework considers various finger sizes, which is a major contribution of this proposed framework. I do see finger sizes being considered in anthropometry studies. However, only 1 finger size is validated in the experimental section. It would be interesting to see the framework being validated on a significantly different finger size.
# Other minor issues
Figure 7: please label locations of CCD markers Lm, LM, Dm, DM, Bm, and DM at the MCP joint
Comments on the Quality of English LanguageThe manuscript is easy to read overall. The paper is well structured. The manuscript could be further improved by shortening the sentences and paragraphs to make the description more concise to the point.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsNo further comments at all.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsI appreciate the authors' efforts addressing the comments. I think the clarity has been improved a lot. However, I have one comment on the size validation. Even though the authors chose to only present one size, a discussion of how this method can be scaled to different finger sizes needs to be added and clarified.
Comments on the Quality of English LanguageThe writing is easy to follow.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
