Development of a Wearable Finger Exoskeleton for Rehabilitation

Round 1

Reviewer 1 Report

The work of Hernández-Santos et. al aimed to prototype a 1-DOF mechanism for finger exoskeleton. The paper presents many lacks. The scientific content of the work is poor, since authors presented a report of the prototyping activities without considering comparisons with other possible solutions. This latter aspect is of fundamental importance also to highlight the novelty of the work, which is not clearly showed. Moreover the paper results heavy to read, authors for instance, merged results with methodological aspects. It is opinion of this reviewer that the paper should be rejected.

MAJOR CONCERNS

1) Introduction, row 36 to 61; authors listed a series of papers that are topic-focused however the readability of the entire section is poor, the authors should integrate in a homogeneous way these statements with those reported in section 1.1. Indeed this latter is more readable, presenting a better organized perspective on the topic. Thus, this reviewer suggests the authors to render uniform the Introduction without any subdivision, reorganizing completely the entire section. Moreover, the authors should better describe 1-DOF systems since they should be used as a comparison with respect to their implemented solution (see also next comment). 

2) Introduction, Table 1; authors reported only multiple actuated systems, however there are many systems that actuate just one DoF. For instance, HandSOME (Brokaw et al.), WaveFlex  (Otto Bock), Hand Robot Alpha-II (MIT), IntelliArm (Ren et al.). Such systems should be better highlighted since one point of the paper is the only one actuated DoF of the implemented solution. 

3) Section 3.3 (Dynamic analysis); a better formalization with the use of generalized coordinates in vector notation can improve the  readability of the section.

4) Experimental results; the entire section should be highly reviewed. Indeed some parts should be described in the Methods and not introduced here (i.e. Solidworks, Simulink SimMechanics toolbox and ANSYS). Many repetition are present ( concepts reported in rows 250-251 were also rewritten in rows 261-262).

5) Simulink block schemes (see figures 6,7,8) are certainly well presented but they do not add anything to the work from a scientific perspective. They are not neither a result nor a methodological aspect (the result of the solidwork model importing in Simulink it is not proper in this context)  

6) Discussion is absent, no comparisons with other 1-DoF systems are reported. In the conclusion section authors did not report anything regarding the novelty of the work with respect to the state of the art.  

MINOR CONCERNS

1) Introduction, Table 1; authors should substitute GDL with DoF.

2) Structure design for the Finger Exoskeleton, row 104 to 121; support the section with citation.

3) Section 3.2 (Inverse Kinematics) The inverse kinematic problem for a three-link open chain has been widely studied (see B. Siciliano et. al., Robotics or the paper authors cited in [14]). This reviewer suggests to short the section.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Authors paper appears well built, the subject is interesting, mainly the use of 1 dof.

I suggest to improve references with the following papers: 

Borboni A., Mor M., Faglia R. “Gloreha – Hand Robotic Rehabilitation: Design, Mechanical Model, and Experiments” J. Dyn. Sys., Meas., Control 138(11), 111003 (Jul 11, 2016)   and  Effects of Soft Tissue Articraft in the Measurement of Hand Kinematics, April 2020, International Review of Mechanical Engineering 14(4): 230. DOI: DOI: 10.15866/ireme.v14i4.18253 Authors: Federica Ragni, Cinzia Amici, Alberto Borboni, Rodolfo Faglia, Paolo Pedersini, Valter Cappellini, Jorge H Villafrane.             Regards.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

First of all, I want to acknowledge the author's detailed work on presenting the information in terms of tables. As a reviewer/ reader it's very easy to understand the content. However, the paper is still very very long and will fail to keep the readers engaged.

The authors have discussed the mathematical analysis in great detail, however, some of the information is repetitive( section 3: position analysis, dynamics analysis......), authors can simply cite the previous paper. I don't feel the need n this topic.

AUTHORS SHOULD ADD MORE DETAIL IN SECTION 4.4 STRUCTURAL ANALYSIS, this section should be the key highlight. Authors can also discuss are current setbacks for 3D printing technology for exoskeleton applications

Author Response

Please see the attachment 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Authors provided reasonable response to the comments. They rendered the manuscript available for publication. However the PDF version of the manuscript does not appear clear (see for instance eq 33 and so on). Moreover I suggest a refine of the English employed.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

This paper presents the mechanical design of a finger exoskeleton. It presents the forward and inverse kinematics, and the dynamics of a single finger exoskeleton. A structural analysis is also performed. All of these methods follow standard engineering design procedures.

Although it is clear that a significant amount of work has been performed, this paper is not ready for publication. It reads as a student paper and not as a journal publication. Some of the reasons for this assessment are included below:

1. There are many grammatical, punctuation and formatting mistakes that need to be fixed. The write up can be significantly improved by providing better flow between sentences and sections.
2. When presenting a literature review, there is no point in just listing everything that has been developed without elaborating on their contribution. Instead of just listing what is out there, refer to the key pieces of literature that are relevant, identify their limitations and the gap that exists within the state of the art, thereby setting the motivation for the work you are presenting.
3. What is the objective, goal or purpose of the design?
4. How does this work contribute to the state of the art?
5. How does this work solve issues currently existing with previous exoskeletons?
6. How is this exoskeleton to be used in rehabilitation setting?
7. An important limitation is that although a prototype is developed, there is no demonstration that the prototype works, or that the equations presented are proven to be correct in the real system. Even though they are validated in simulation, without validation in the real prototype, the equations are not useful to the community.
8. Finally, there are significant limitations to the design, which are not discussed, or even presented. For example, how to you adapt to different hand sizes, how do you handle the slip motion of the exoskeleton components on the finger and what effects can those have on motion support.

Reviewer 2 Report

The paper introduced a design of a wearable finger exoskeleton for rehabilitation. By modeling the design to a 1-DOF bar linkage, the paper calculated the range of motion of the finger. The paper also tried to find the inverse kinematics by simplifying it to an open 3 bar linkage and to conduct dynamics analysis by simplifying it to a 2 bar linkage. The paper also showed the simulation using Matlab/Simulink with SimMechanics and structural analysis of the design using ANSYS. Finally, the paper showed a 3D printed design without the linear actuator. 

The paper tried to show a complete package of analysis of a mechanical device. However, the components of the analysis are not quite consistent and more physical and mathematical clarifications and considerations are desired. 

1. Fig. 3 should show both the initial position and the final position as the calculations refer to them. They will also help readers understand early on how the mechanism works. The Fig. 9 showed the two positions, but is a little bit late. Also, Fig. 3b would be much clearer if it can show the drawings of two positions. It is not crystal clear with the current format. 
2. It is not very clear the reason why the authors use the expression in Eq. 1. θ₂=atan2(d1, D2) will get the same result but with a much simpler format. 
3. Many variables used in the paper are not clearly marked in the figures or explained in the content. For example, θ1i,  θ19,  θ20, θ24, all need to be guessed by the reader. 
4. It’s not clear how the inverse kinematic model is obtained as shown in section 3.2. Where are the joints in the design (Fig. 3) and what are the link length L1, L2 and L3? Since they are different from the lengths L1, L2, L3 used in section 3.1, it is better to choose a different letter to distinguish them. Meanwhile, the inverse analysis shown here is for a general four bar linkage, which is well known. It doesn’t provide too much information unique to this paper and the authors need to elaborate more how this is applied. 
5. In section 3.3, the third link is removed without specifying reason. Meanwhile, the force and torque analysis can’t be directly applied to the mechanisms described in this paper. Again, the authors need to explain how this section is related to the mechanism described earlier (Fig. 3). 
6. Corresponding to the models in sections 3.2 and 3.3, the simulation results in 4.2 & 4.3 need to be explained as where they are shown in the design and the physical meaning in the design. BTW, in line 302, the “joints vs the weather” is a typo. 
7. The FEA analysis in 4.4 should show where and how the external forces are applied. As shown in Fig. 3, friction forces can be significant in this kind of mechanism, a reader may wonder whether they are included, and if yes, how they are modeled in the analysis. Also, it is desirable to obtain the force on the tip of the finger when the 40N is exerted from the linear actuator. 
8. Figure 14 should show the initial position and the final position instead of two initial positions.