Submited on: 17 Jul 2012 12:36:37 PM GMT
Published on: 17 Jul 2012 04:55:01 PM GMT
 
Invited review of this paper
Posted by Anonymous Reviewer on 23 Jul 2012 07:18:32 PM GMT

  • What are the main claims of the paper and how important are they?

    This paper describes a blood vessel segmentation method, which uses a two-state approach: perception and interpretation. The perception stage uses two-scale Laplacian of Gaussian, which is inspired by a low-level vision mechanism, to detect (both vascular and non-vascular) objects of different sizes. The interpretation stage includes partitioning the binary structures into vessel segments and extracting a number of their features, and then classifying these features with artificial neural networks (ANN). Finally a vascular tree is assembled with rule-based tracking. The proposed method is validated on STARE and DRIVE databases.


  • Are these claims novel? If not, please specify papers that weaken the claims to the originality of this one.

    Many methods for retinal vessel segmentatino have been proposed in the literature; nonetheless, the combindation of a two-scale filtering with vescular tree analysis might be considered somewhat novel. 


  • Are the claims properly placed in the context of the previous literature?

    Yes. The relevant work is presented properly. 


  • Do the results support the claims? If not, what other evidence is required?

    Yes. The proposed method is validated using the STARE and DRIVE databases, and it is compared with other state-of-the-art methods. The experimental results appear to be comparable to (and occasionally better than) several state-of-the-art methods. 


  • If a protocol is provided, for example for a randomized controlled trial, are there any important deviations from it? If so, have the authors explained adequately why the deviations occurred?

    N/A. 


  • Is the methodology valid? Does the paper offer enough details of its methodology that its experiments or its analyses could be reproduced?

    Yes, it appears so. The paper provides enough details and the results appear to be reproducible. 


  • Would any other experiments or additional information improve the paper? How much better would the paper be if this extra work was done, and how difficult would such work be to do, or to provide?

    A successful, real-world application might be an ultimate validation of the usefulness of the method. 


  • Is this paper outstanding in its discipline? (For example, would you like to see this work presented in a seminar at your hospital or university? Do you feel these results need to be incorporated in your next general lecture on the subject?) If yes, what makes it outstanding? If not, why not?

    Not really. Many methods have already been proposed in this field. The improvement is not as significant.  


  • Other Comments:

    No. 

  • Competing interests:
    No
  • Invited by the author to review this article? :
    No
  • Have you previously published on this or a similar topic?:
    No
  • References:
  • Experience and credentials in the specific area of science:

    Good.

  • How to cite:  Anonymous.Invited review of this paper[Review of the article 'Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels. ' by La Torre A].WebmedCentral 2012;3(7):WMCRW002116
1 2 3 4 5 6 7 8 9
Report abuse
 

  • What are the main claims of the paper and how important are they?

    NA


  • Are these claims novel? If not, please specify papers that weaken the claims to the originality of this one.

    NA


  • Are the claims properly placed in the context of the previous literature?

    NA


  • Do the results support the claims? If not, what other evidence is required?

    NA


  • If a protocol is provided, for example for a randomized controlled trial, are there any important deviations from it? If so, have the authors explained adequately why the deviations occurred?

    NA


  • Is the methodology valid? Does the paper offer enough details of its methodology that its experiments or its analyses could be reproduced?

    NA


  • Would any other experiments or additional information improve the paper? How much better would the paper be if this extra work was done, and how difficult would such work be to do, or to provide?

    NA


  • Is this paper outstanding in its discipline? (For example, would you like to see this work presented in a seminar at your hospital or university? Do you feel these results need to be incorporated in your next general lecture on the subject?) If yes, what makes it outstanding? If not, why not?

    NA


  • Other Comments:

    An accurate segmentation of retinal blood vessels has to be done in order to reach a useful diagnostic conclusion about their morphological  changes  in important diseases like diabetes and hypertension. All this process, if manually done, is largely time consuming and presents difficulties in patient’s results comparisons. The main paper’s claim is an automatic procedure including both  vessel tree segmentation (detection of objects against the retina image background) and its consequent diagnostic analysis performed by an artificial neural network operating according to a “human vision inspired” approach. Owing to the still difficult automatic achieving of  reliable diagnostic assessment, it may be promising.

    Many algorithms and methods of segmentation, included the use of artificial neural networks, have been already introduced; however the automatic integration of conventional multiscale segmentation and “human vision based” vessel tree analysis, presented in this paper, may be considered a novel approach.

    The paper could refer also to some recent review on the topic, as for example:
    MM Fraz et al. Blood vessel segmentation methodologies in retinal images – A survey.
    Computer Methods and Programs in Biomedicine, Available online 21 April 2012. http://dx.doi.org/10.1016/j.cmpb.2012.03.009 or some old basic paper published for the ophthalmological community: Chanjira Sinthanayothin et al. Automated localisation of the optic disc, fovea, and retinal blood vessels from digital colour fundus images. Br J Ophthalmol 1999;83:902–910

    The results support the claims but it needs a better way of detecting OD in pathological retinas.

    The authors use STARE and DRIVE databases offering sensitivity, specificity and accuracy evaluation of the test; this presents a solid reproducibility.

    Despite being the exploitation of STARE and DRIVE very useful to test algorithms, a clinical trial on  patients should be introduced as soon as possible to assess drawbacks, limits and advantages of this method.

    The paper is a good contribution to the field of automatic diagnosis of retinal pathologies but is not outstanding. However it should be included in review lectures about this topic.

    Some difficulty arises to interpret correctly the analytic notations  reported at page 4 and 5 of pdf file, due to unusual characters/symbols used (typo or font errors? Both in the online writeup and in pdf text a question mark is typed, e.g. w = 2?2·3? or I?1, I?2 )

  • Competing interests:
    No
  • Invited by the author to review this article? :
    No
  • Have you previously published on this or a similar topic?:
    No
  • References:

    NA

  • Experience and credentials in the specific area of science:

    I was involved in Ophthalmological research years ago

  • How to cite:  Fonda S .Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels[Review of the article 'Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels. ' by La Torre A].WebmedCentral 2012;3(7):WMCRW002113
1 2 3 4 5 6 7 8 9
Report abuse
 

  • What are the main claims of the paper and how important are they?

    An accurate segmentation of retinal blood vessels has to be done in order to reach a useful diagnostic conclusion about their morphological  changes  in important diseases like diabetes and hypertension. All this process, if manually done, is largely time consuming and presents difficulties in patient’s results comparisons. The main paper’s claim is an automatic procedure including both  vessel tree segmentation (detection of objects against the retina image background) and its consequent diagnostic analysis performed by an artificial neural network operating according to a “human vision inspired” approach. Owing to the still difficult automatic achieving of  reliable diagnostic assessment, it may be promising.


  • Are these claims novel? If not, please specify papers that weaken the claims to the originality of this one.

    Many algorithms and methods of segmentation, included the use of artificial neural networks, have been already introduced; however the automatic integration of conventional multiscale segmentation and “human vision based” vessel tree analysis, presented in this paper, may be considered a novel approach.


  • Are the claims properly placed in the context of the previous literature?

    Yes. However the paper could refer also to some recent review on the topic, as for example:
    MM Fraz et al. Blood vessel segmentation methodologies in retinal images – A survey.
    Computer Methods and Programs in Biomedicine, Available online 21 April 2012. http://dx.doi.org/10.1016/j.cmpb.2012.03.009

    or some old basic paper published for the ophthalmological community:
    Chanjira Sinthanayothin et al. Automated localisation of the optic disc, fovea, and retinal blood vessels from digital colour fundus images. Br J Ophthalmol 1999;83:902–910

     


  • Do the results support the claims? If not, what other evidence is required?

    Yes, but it needs a better way of detecting optic disc in pathological retinas


  • If a protocol is provided, for example for a randomized controlled trial, are there any important deviations from it? If so, have the authors explained adequately why the deviations occurred?

    NA


  • Is the methodology valid? Does the paper offer enough details of its methodology that its experiments or its analyses could be reproduced?

    The authors use STARE and DRIVE databases, offering sensitivity, specificity and accuracy evaluation of the test; this presents a solid reproducibility


  • Would any other experiments or additional information improve the paper? How much better would the paper be if this extra work was done, and how difficult would such work be to do, or to provide?

    Despite being the exploitation of STARE and DRIVE very useful to test algorithms, a clinical trial on  patients should be introduced as soon as possible to assess drawbacks, limits and advantages of this method


  • Is this paper outstanding in its discipline? (For example, would you like to see this work presented in a seminar at your hospital or university? Do you feel these results need to be incorporated in your next general lecture on the subject?) If yes, what makes it outstanding? If not, why not?

    The paper is a good contribution to the field of automatic diagnosis of retinal pathologies but is not outstanding. However it should be included in review lectures about this topic

     


  • Other Comments:

    Difficulty to interpret correctly the analytic notations  reported at page 4 and 5, due to unusual characters/symbols used (typo or font errors? Both online and in pdf text a question mark is typed, e.g. w = 2?2·3? or I?1, I?2 )

     

  • Competing interests:
    No
  • Invited by the author to review this article? :
    Yes
  • Have you previously published on this or a similar topic?:
    Yes
  • References:

    I was involved in ophthalmological research years ago. Sergio Fonda, MS; Bruno Bagolini, MD, Relative Photometric Measurements of Retinal Circulation (Dromofluorograms)_A Television Technique. Arch Ophthalmol. 1977;95(2):302-307. doi:10.1001/archopht.1977.04450020103017

  • Experience and credentials in the specific area of science:
    None
  • How to cite:  Fonda S .Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels[Review of the article 'Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels. ' by La Torre A].WebmedCentral 2012;3(7):WMCRW002111
1 2 3 4 5 6 7 8 9
Report abuse
 

  • What are the main claims of the paper and how important are they?

    A computer vision approach which  mimics the image reading by expert people.


  • Are these claims novel? If not, please specify papers that weaken the claims to the originality of this one.

    so so


  • Are the claims properly placed in the context of the previous literature?

    Yes


  • Do the results support the claims? If not, what other evidence is required?

    Yes


  • If a protocol is provided, for example for a randomized controlled trial, are there any important deviations from it? If so, have the authors explained adequately why the deviations occurred?

    It is not the case


  • Is the methodology valid? Does the paper offer enough details of its methodology that its experiments or its analyses could be reproduced?

    Yes


  • Would any other experiments or additional information improve the paper? How much better would the paper be if this extra work was done, and how difficult would such work be to do, or to provide?

    No


  • Is this paper outstanding in its discipline? (For example, would you like to see this work presented in a seminar at your hospital or university? Do you feel these results need to be incorporated in your next general lecture on the subject?) If yes, what makes it outstanding? If not, why not?

    It is a good paper. The paper presents a new method and experiments to supprort that.


  • Other Comments:

    It would better to utilize diagrams for methodology explaination to increase readibility of the paper

  • Competing interests:
    No
  • Invited by the author to review this article? :
    No
  • Have you previously published on this or a similar topic?:
    No
  • References:
  • Experience and credentials in the specific area of science:

    Good

  • How to cite:  Balafar M .Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels.[Review of the article 'Multiscale Filtering and Neural Network Classification for Segmentation and Analysis of Retinal Vessels. ' by La Torre A].WebmedCentral 2012;3(7):WMCRW002098
1 2 3 4 5 6 7 8 9
Report abuse