kumoh national institute of technology
Networked Systems Lab.

Review Comment

NSL > Education> Review Comment
Alif - [CCNC 2017]
By :
Date : 2016-10-27
Views : 116

Dear Mr. Alif Pranata:

Congratulations - your paper #1570314234 ('Detrended Fluctuation Analysis on ECG Device for Home Environment') submitted at the "Workshop - Ambient Assisted Living and eHealth" has been accepted and will be presented and published in the Proceedings of IEEE CCNC 2017 and IEEE Xplore (subject to the fulfillment of conditions outlined below). Information about the format of your presentation, the date/time of the your presentation, and other program schedules will be announced later. However, the general program-at-a-glance of IEEE CCNC 2017 is available at: http://ccnc2017.ieee-ccnc.org/content/program-glance

Please read the reviews carefully. They are attached below. The reviews can also be accessed at http://edas.info/showPaper.php?m=1570314234.

CAMERA READY PREPARATION:
Please prepare the final (camera ready) version of your manuscript according to the IEEE template for conferences:
http://www.ieee.org/conferences_events/conferences/publishing/templates.html

Your final manuscript must undergo a pdf-express check before submission, and then the pdf-express generated file must be uploaded to EDAS. This process is explained in more detail at the "authors" web page: http://ccnc2017.ieee-ccnc.org/authors

A pdf-express generated file is mandatory to ensure that your paper will be uploaded on IEEE Xplore.
Please note that the page length limit is 6. Pages 7, 8 and 9 are allowed with a charge of USD 100 for each additional page.

DEADLINE AND UPLOAD:
The deadline for the camera ready upload is November 4th (FIRM) and has to be done through EDAS.
Please do not wait until the last day to avoid last minute problems.

REGISTRATION AND PRESENTATION:
To be published in the Proceedings of IEEE CCNC 2017 and IEEE Xplore, an author of an accepted paper is required to register for the conference at the full (member or non-member) rate and the paper must be presented by an author of that paper.

One full registration covers up to three papers (either conference, workshop, demo, poster, work in progress...) For more details, please see: http://ccnc2017.ieee-ccnc.org/registration

Regards,
Pietro Manzoni, TPC Chair
Ben Lee, TPC Vice Chair

===========================================================================
======= Review 1 =======

> *** Overall Rating: What is your overall recommendation for the paper?
Marginal (30-50%) (2)

> *** Relevance: How relevant is the topic of this paper to CCNC?
Neutral (2)

> *** Novelty: How novel is the approach/topic presented in the paper?
Minor tweak of prior schemes (1)

> *** Technical Strength: Is the paper technically sound and correct
Significant flaws that may call key result of the paper into question (1)

> *** Presentation: What is the quality of presentation of the paper?
The paper is overall well written. There are small flaws that are correctable. (3)

> *** Reviewer Confidence: How comfortable are you in the review decision?
Very comfortable and know much of the related work (my area) (3)

> *** Short Summary: Please give a short summary of the paper

This paper presents two main contributions, the development of a cheap ECG and the signal analysis of the ECG to try to identify is user is a smoker or not, using a well known algorithm.

> *** Main Strengths (Reasons to Accept): What are the main arguments to accept the paper?

It presents the use of a well-known algorithm applied in a novel area (e-health) using a "home developed" cheap ECG.

> *** Main Weaknesses (Reasons to reject): What are the main arguments to reject the paper?

- This paper presents two contributions, however these contributions are really different between them, the development of a cheap ECG looks more like a Technical Development that can be presented in a report, meanwhile signal analysis is more a data analysis problem, that actually can be reported in paper form.

- Data analysis has a lot of issues. A validation of the process is needed, along the paper is never mentioned how it can be validated, for instance the classical approach of 70% - train and 30% blind test. Figure 5 (which has interesting data, but is not explained in the paper) let us know the results of DFA depending if the user is smoker or not, but never validated.

- A "data description" is needed, if data were generated "online", at least a detailes demography data is needed, "women/men", "working/not worker", "like sports/not" etc. that can lead to and overfit of the proposed DFA method.

- Finally, even when is mentioned that can not be compared with the same prior investigation, a comparison, even when the methodology differs, can let us see if the proposal is a valuable approach to try.

> *** Detailed Comments: Please input additional detailed comments below.

Major issues:
- This paper presents two contributions, however these contributions are really different between them, the development of a cheap ECG looks more like a Technical Development that can be presented in a report, meanwhile signal analysis is more a data analysis problem, that actually can be reported in paper form.

- Data analysis has a lot of issues. A validation of the process is needed, along the paper is never mentioned how it can be validated, for instance the classical approach of 70% - train and 30% blind test. Figure 5 (which has interesting data, but is not explained in the paper) let us know the results of DFA depending if the user is smoker or not, but never validated.

- A "data description" is needed, if data were generated "online", at least a detailes demography data is needed, "women/men", "working/not worker", "like sports/not" etc. that can lead to and overfit of the proposed DFA method.

- Finally, even when is mentioned that can not be compared with the same prior investigation, a comparison, even when the methodology differs, can let us see if the proposal is a valuable approach to try.

Minor issues:

Page 1, column 2

"...used in [3] was adopted to..." it is not only bad grammar, but the "[3]" should not be used as a noun, it is a sub-sentence, and the sentence should be complete if removed, just imagine if the format for citation was not brackets, but footnotes. There are a lot of these along the paper.

Page 2, column 1
"...methods includes [6][7][8]. In [6] a multifractal..." Same issue.
"...In [7] DFA..."
"...Two apparent works are [9] and [10]. In [9]..."
"...In line with [9], [10] s..."
P-R and QRS must be defined.

Page 3
Figure 5 is present but never referenced.

======= Review 2 =======

> *** Overall Rating: What is your overall recommendation for the paper?
Accept (10-30%) (3)

> *** Relevance: How relevant is the topic of this paper to CCNC?
Relevant (3)

> *** Novelty: How novel is the approach/topic presented in the paper?
Minor tweak of prior schemes (1)

> *** Technical Strength: Is the paper technically sound and correct
Technical solid (3)

> *** Presentation: What is the quality of presentation of the paper?
The paper is beautifully written. (4)

> *** Reviewer Confidence: How comfortable are you in the review decision?
The paper is way out of my expertise area (0)

> *** Short Summary: Please give a short summary of the paper

This paper investigates Detrended Fluctuation Analysis (DFA) to assess human heart electric signal by using the developed ECG device. Experimental results reveal a cardiac disparity between smoker and nonsmoker.

> *** Main Strengths (Reasons to Accept): What are the main arguments to accept the paper?



> *** Main Weaknesses (Reasons to reject): What are the main arguments to reject the paper?



> *** Detailed Comments: Please input additional detailed comments below.

In page 4, "Figure 4 illustrates this comparison in a line graph". Is it Figure 5?

What's the difference between this work and the previous work [1]?

======= Review 3 =======

> *** Overall Rating: What is your overall recommendation for the paper?
Accept (10-30%) (3)

> *** Relevance: How relevant is the topic of this paper to CCNC?
Relevant (3)

> *** Novelty: How novel is the approach/topic presented in the paper?
Some novel component (2)

> *** Technical Strength: Is the paper technically sound and correct
Technical solid (3)

> *** Presentation: What is the quality of presentation of the paper?
The paper is overall well written. There are small flaws that are correctable. (3)

> *** Reviewer Confidence: How comfortable are you in the review decision?
Comfortable reviewing the paper (close to my area) (2)

> *** Short Summary: Please give a short summary of the paper

In this work authors present measure ECG signals applying the well method detrended fluctuation analysis (DFA), and they implemented the methos in a novel device using a Rasperry pi 2.

> *** Main Strengths (Reasons to Accept): What are the main arguments to accept the paper?

Device proposed presents a novel way to study complex ECG algorithms and signal analysis in hardware.

> *** Main Weaknesses (Reasons to reject): What are the main arguments to reject the paper?

Authors should compare their device with new devices developed in this year. References are not strength related with de application proposed.

> *** Detailed Comments: Please input additional detailed comments below.