Next Article in Journal
Tomato Fungal Disease Diagnosis Using Few-Shot Learning Based on Deep Feature Extraction and Cosine Similarity
Previous Article in Journal
Development of an Automatic Harvester for Wine Grapes by Using Three-Axis Linear Motion Mechanism Robot
 
 
Article
Peer-Review Record

A New Way to Identify Mastitis in Cows Using Artificial Intelligence

AgriEngineering 2024, 6(4), 4220-4232; https://doi.org/10.3390/agriengineering6040237
by Rodes Angelo Batista da Silva 1, Héliton Pandorfi 1,*, Filipe Rolim Cordeiro 2, Rodrigo Gabriel Ferreira Soares 3, Victor Wanderley Costa de Medeiros 3, Gledson Luiz Pontes de Almeida 1, José Antonio Delfino Barbosa Filho 4, Gabriel Thales Barboza Marinho 1 and Marcos Vinícius da Silva 1
Reviewer 1:
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4: Anonymous
AgriEngineering 2024, 6(4), 4220-4232; https://doi.org/10.3390/agriengineering6040237
Submission received: 29 July 2024 / Revised: 22 October 2024 / Accepted: 30 October 2024 / Published: 8 November 2024
(This article belongs to the Section Livestock Farming Technology)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

line 168 it should stand milk production not blood production

Mayor question is why did you use CMT instead some other test? SSC for example. CMT is considered to be insensitive and is can detect about 400.000 somatic cells per ml while SSC can detect about a 1000.

I think that you need to have a veterinarian in your team.

Author Response

Response to Reviewer 1 Comments

 

Dear Editor,

 

We appreciate the time and effort that you and the reviewers dedicated to providing feedback on our manuscript and are grateful for the insightful comments on and valuable improvements to our paper. Thank you also for giving me the opportunity to submit a revised draft of my manuscript (agriengineering-3156166) titled A new way to identify mastitis in cows using artificial intelligence. All changes in the manuscript were highlighted in red for reviewer 1, to facilitate the understanding of the editor and reviewer. We have incorporated the suggestions made by the reviewer. Below is a point-by-point response to the reviewer's comments and concerns.

 

  1. line 168 it should stand milk production not blood production

Author´s response: the term “blood production” was replaced by “milk production” (Line 190).

 

  1. Mayor question is why did you use CMT instead some other test? SSC for example. CMT is considered to be insensitive and is can detect about 400.000 somatic cells per ml while SSC can detect about a 1000.

Author´s response: the California Mastitis Test (CMT) is a widespread technique for identifying cases of subclinical mastitis. The main advantages of the CMT are its low cost, the fact that it can be carried out by the milker during milking, the speed with which the results can be interpreted and the indication of the level of infection in each mammary quarter. The option for more precise and modern techniques has not been adopted due to the technological level of the production unit, the distance from the analysis centers and the financial resources associated with the need to collect samples for laboratory analysis. In this sense, precision livestock farming is an emerging field that includes computer technologies used to develop tools aimed at real-time monitoring and management of the smallest production units (animals), improving production processes such as early detection of diseases and automation of tasks such as milking, grazing, feeding and efficient resource management.

 

  1. I think that you need to have a veterinarian in your team.

Author´s response: the study was submitted to the Ethics Committee for the Use of Experimental Animals in which we had the participation of a veterinarian, however, for the specific approach of this manuscript the team decided to count on the collaboration of the professional in other developments of the research.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

The topic of this document is related of the most important healthy problem in the dairy herds around the world. AI is the new technology that promise to be applied as diagnosis tools, but I think it is necessary  to establish an accuracy protocol that included representative images or pictures of subclinical mastitis mammary glands.

Also, you should add references to describe the thermo-morphological anatomy of the mammary gland body, not only teats.

I think that  the objective of this paper is so important as diagnostic tool to increase the possibilities of cure rate and earlier diagnostic.

Please, improvement the information and references. 

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 2 Comments

 

Dear Editor,

 

We appreciate the time and effort that you and the reviewers dedicated to providing feedback on our manuscript and are grateful for the insightful comments on and valuable improvements to our paper. Thank you also for giving me the opportunity to submit a revised draft of my manuscript (agriengineering-3156166) titled A new way to identify mastitis in cows using artificial intelligence. All changes in the manuscript were highlighted in red for reviewer 2, to facilitate the understanding of the editor and reviewer. We have incorporated the suggestions made by the reviewer. Below is a point-by-point response to the reviewer's comments and concerns.

 

Dear authors,

 

  1. The topic of this document is related of the most important healthy problem in the dairy herds around the world. AI is the new technology that promise to be applied as diagnosis tools, but I think it is necessary to establish an accuracy protocol that included representative images or pictures of subclinical mastitis mammary glands.

 

  1. Also, you should add references to describe the thermo-morphological anatomy of the mammary gland body, not only teats.

 

  1. I think that the objective of this paper is so important as diagnostic tool to increase the possibilities of cure rate and earlier diagnostic.

 

  1. Please, improvement the information and references.

 

Author´s response: Inserted text lines 45-49: Measuring the surface temperature of the breast skin by infrared thermography is a noninvasive method capable of detecting temperature changes due to inflammation, used in the diagnosis of subclinical mastitis in dairy cows [1, 4, 7, 8]. In these studies, the temperature intervals varied considerably reaching 5.3°C of difference between a healthy animal and those with subclinic mastitis.

 

Author´s response: Inserted text lines 50-64: “When subclinical mastitis occurs, the local inflammatory response has been shown to cause animal fever and changes in tissue blood flow [32, 33], which can be the origin of an increase in the surface temperature of the udder skin [34]. The udder with mastitis has a high temperature even before clinical symptoms appear. In addition, [36] reported an increase of 2 to 3oC in the udder surface temperature of lactating cows after inoculation with Escherichia coli in different parts of the udder. In a study carried out by the National Institute of Animal Health (NIAH), [35] observed that udder temperature measured using thermography can be a useful diagnostic tool for detecting mastitis in dairy cattle. In the same way, the authors [37] used thermographic images as a tool to detect mastitis in sheep, especially subclinical mastitis.

Screening for subclinical mastitis by measuring udder surface temperature has a high predictive diagnostic capacity similar to the California Mastitis Test (CMT). However, analysis of the reliability of surface temperature by thermography among cows with different body and physiological characteristics living in different environmental conditions must be determined in each case [7].”

 

Author´s response: Inserted text lines 65-75: In a study on dairy cows with mastitis, [1] developed a software that automatically measures udder temperature. The research demonstrated a strong correlation between these temperatures and the somatic cell count, which significantly facilitates the diagnosis of clinical mastitis. The findings indicate that elevated udder temperature is a reliable early indicator of mastitis, allowing for quicker intervention and better management of udder health. In another study, [9] compared automatic image recognition software with manual methods for detecting E. coli infections in cows. The results showed that the automated system provided comparable accuracy to manual methods, with the added benefits of increased speed and objectivity in identifying infections. This suggests that the implementation of image recognition technology could streamline the process of infection detection, improving both efficiency and reliability in herd management.

 

Author´s response: Text Inserted Lines 76-83: In a study based on convolutional networks [2], an object detector was developed using YOLOv3 as a proposal to perform the detection and provide answers about the clinical picture of the animal and obtained accuracy of 83.33%. The transfer of learning techniques by convolutional networks was reported in a study [9] developed with digital images of nipples for classification and identification of bovine mastitis reaching AUC of 0.920 or higher. However, few studies have applied this technique in thermal images of udders of animals mainly aiming to obtain the best model for their classification, because using optimized networks have improvements in classification than the manual approach.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Mastitis, or inflammation of the udder, is one of the most common and at the same time most costly ailments in dairy herds. It causes a decrease in production, a decrease in milk quality, a deterioration in the condition of the animals, and in extreme cases can lead to death. Mastitis usually develops as a result of the immune system's reaction to the penetration of bacteria (as well as fungi, viruses and algae) into the teat canal, but it can also be a consequence of chemical, mechanical or thermal trauma to the udder. Mastitis can take the form of an easily recognizable clinical inflammation of the udder, but it can also occur in a subclinical form with few or no symptoms. The authors of the manuscript entitled "A new way to identify mastitis in cows using artificial intelligence" describe the use of the latest thermal imaging techniques in mastitis diagnostics.

The content of the manuscript is very interesting, but a major limitation is the lack of clinical confirmation of the obtained results. In my opinion, the work lacks an important element, which is laboratory tests. At this stage of the research, we only have scans obtained in subsequent measurements, but there is no confirmation anywhere of the clinical condition of the animal that the authors were really dealing with. Was the increased temperature of the udder correlated with the ongoing inflammation, even subclinical? This is not known from this manuscript.

As a veterinarian, I see a significant lack of these correlations in this work.

I also do not agree with the statement that currently used methods of diagnosing mastitis are subjective or expensive. Currently, laboratory tests, such as bacteriological examination of milk from each quarter individually, are the most objective tests used widely throughout the world, and the cost of this test is not high.

Author Response

Response to Reviewer 3 Comments

 

Dear Editor,

 

We appreciate the time and effort that you and the reviewers dedicated to providing feedback on our manuscript and are grateful for the insightful comments on and valuable improvements to our paper. Thank you also for giving me the opportunity to submit a revised draft of my manuscript (agriengineering-3156166) titled A new way to identify mastitis in cows using artificial intelligence. All changes in the manuscript were highlighted in red for reviewer 3, to facilitate the understanding of the editor and reviewer. We have incorporated the suggestions made by the reviewer. Below is a point-by-point response to the reviewer's comments and concerns.

 

  1. Mastitis, or inflammation of the udder, is one of the most common and at the same time most costly ailments in dairy herds. It causes a decrease in production, a decrease in milk quality, a deterioration in the condition of the animals, and in extreme cases can lead to death. Mastitis usually develops as a result of the immune system's reaction to the penetration of bacteria (as well as fungi, viruses and algae) into the teat canal, but it can also be a consequence of chemical, mechanical or thermal trauma to the udder. Mastitis can take the form of an easily recognizable clinical inflammation of the udder, but it can also occur in a subclinical form with few or no symptoms. The authors of the manuscript entitled "A new way to identify mastitis in cows using artificial intelligence" describe the use of the latest thermal imaging techniques in mastitis diagnostics.

 

  1. The content of the manuscript is very interesting, but a major limitation is the lack of clinical confirmation of the obtained results. In my opinion, the work lacks an important element, which is laboratory tests. At this stage of the research, we only have scans obtained in subsequent measurements, but there is no confirmation anywhere of the clinical condition of the animal that the authors were really dealing with. Was the increased temperature of the udder correlated with the ongoing inflammation, even subclinical? This is not known from this manuscript.

 

  1. As a veterinarian, I see a significant lack of these correlations in this work.

 

  1. I also do not agree with the statement that currently used methods of diagnosing mastitis are subjective or expensive. Currently, laboratory tests, such as bacteriological examination of milk from each quarter individually, are the most objective tests used widely throughout the world, and the cost of this test is not high.

 

Author´s response: The California Mastitis Test (CMT) is a widespread technique for identifying cases of subclinical mastitis. The main advantages of the CMT are its low cost, the fact that it can be performed by the milker during milking, the speed with which the results can be interpreted and the indication of the level of infection in each mammary quarter. The option of more precise and modern techniques has not been adopted due to the technological level of the production unit, the distance from analysis centers and the financial resources associated with the need to collect samples for laboratory analysis. In this sense, precision livestock farming is an emerging field that includes computer technologies used to develop tools aimed at the real-time monitoring and management of the smallest production units (animals), improving production processes such as the early detection of diseases and the automation of tasks such as milking, grazing, feeding and efficient resource management.

The option of more precise and modern techniques has not been adopted due to the technological level of the production unit, the distance from analysis centers and the financial resources associated with the need to collect samples for laboratory analysis. In this sense, precision livestock farming is an emerging field that includes computer technologies used to develop tools aimed at real-time monitoring and management of the smallest production units (animals), improving production processes such as the early detection of diseases and the automation of tasks such as milking, grazing, feeding and efficient resource management.

 

Inserted text lines 12-16 “Mastitis is a disease that is considered an obstacle in dairy farming. Some methods of diagnosing mastitis have been used effectively over the years, but with an associated relative cost that reduces the producer's profit. In this context, this sector needs tools that offer an early, safe and non-invasive diagnosis and that direct the producer to apply resources to confirm the clinical picture, minimizing the cost of monitoring the herd”

 

Inserted text lines 50-64: “When subclinical mastitis occurs, it has been shown that the local inflammatory re-sponse causes fever in the animal and changes in tissue blood flow [32, 33], which can be the origin of an increase in the surface temperature of the udder skin [34]. The udder with mastitis has an elevated temperature even before clinical symptoms appear. In addition, [36] reported an increase of 2 to 3°C in the udder surface temperature of lactating cows af-ter inoculation with Escherichia coli in different parts of the udder. In a study carried out by the National Institute of Animal Health (NIAH), [35] observed that udder temperature measured using thermography can be a useful diagnostic tool for detecting mastitis in dairy cattle. Similarly, the authors [37] used thermographic images as a tool to detect mas-titis in sheep, especially subclinical mastitis.

Screening for subclinical mastitis by measuring udder surface temperature has a high predictive diagnostic capacity, similar to the California Mastitis Test (CMT). However, analysis of the reliability of surface temperature by thermography among cows with different body and physiological characteristics living in different environmental condi-tions must be determined in each case [7].”

 

Inserted text lines 362-369: “The results of this study indicate that image processing techniques applied to thermal images were able to extract characteristics that helped classify the images; however, more studies need to be carried out involving convolution networks and their techniques in order to obtain new results that help producers make decisions. In terms of detection time, the proposed predictive methodology is faster and more objective, directing animals to the CMT test. In relation to SCC and Electronic Somatic Cell Count and other laboratory methods (such as microbial culture), the proposed methodology can also support the identification of animals that will have their milk samples tested.”

 

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors, congratulations on your research. Quick and ingenious solutions are continuously needed and relevant in the diagnosis of mastitis. Tissue temperature evaluation is a technique increasingly used to assist farmers and for other purposes, such as heat detection. This study beautifully combines infrared thermography and artificial intelligence to provide a simple and fast method for diagnosing mastitis. To improve the article, I recommend a thorough literary review, as there are several sentences and phrases throughout the text that are completely out of context. Additionally, a few more targeted remarks are provided below.

Line 17 replace By means with “using”

Lines 17-19 Subclinical mastitis is an infection that typically does not show visible symptoms but can still affect the health of the cow. Therefore, an animal diagnosed with subclinical mastitis would generally not be considered entirely healthy. To clarify the sentence, you might say: "The image bank used in the research consisted of 165 images, each with a resolution of 360 x 360 pixels, sourced from a database of 55 animals diagnosed with subclinical mastitis, all of which were not exhibiting clinical symptoms at the time of imaging."

Lines 22-25 replace with this for clarity “The proposed predictive methodologies, based on knowledge transfer, were effective in accurately classifying the images. This significantly enhanced the automatic detection of both healthy animals and those diagnosed with subclinical mastitis using thermal images of the udders of dairy cows.”

Line 29 replace with “Early diagnosis”

Line 33 replace Asymptomatic with subclinic

Line 38 replace parlor with room

Lines 39-42 replace with this for clarity “In this context, developing diagnostic methods that integrate automation is crucial to minimize costs and losses, ensure accuracy and speed in diagnosis, and promote non-invasive techniques like thermography to enhance animal welfare.”

Line 45 use the journal format for cite the references, in this case use [1,4,7,8]

Lines 50-55 replace with this for clarity “In a study on dairy cows with mastitis, Zaninelli et al., 2018 [1] developed a software that automatically measures udder temperature. The research demonstrated a strong correlation between these temperatures and the somatic cell count, which significantly facilitates the diagnosis of clinical mastitis. The findings indicate that elevated udder temperature is a reliable early indicator of mastitis, allowing for quicker intervention and better management of udder health. In another study, Wilson et al., 2017 [9] compared automatic image recognition software with manual methods for detecting E. coli infections in cows. The results showed that the automated system provided comparable accuracy to manual methods, with the added benefits of increased speed and objectivity in identifying infections. This suggests that the implementation of image recognition technology could streamline the process of infection detection, improving both efficiency and reliability in herd management.”

Line 75 delete this Determining the convergence result

Lines 84-86 replace with this for clarity Bayesian optimization is a hyperparameter search technique that surpasses both grid search and random search. Unlike these methods, Bayesian optimization leverages knowledge from previous iterations to inform the search process. This approach enhances decision-making by more effectively identifying the optimal hyperparameter settings for evaluating a model.

Line 170  replace with this The clinical status of the animals was assessed using the California Mastitis Test (CMT) before milking, after discarding the first drop of milk.

Line 170-171 delete this sentence

Lines 290-292 references are needed to support this statement

Practically, sources need to be added throughout the entire discussion section.

Comments on the Quality of English Language

Moderate editing of English language required.

Author Response

Response to Reviewer 4 Comments

 

Dear Editor,

 

We appreciate the time and effort that you and the reviewers dedicated to providing feedback on our manuscript and are grateful for the insightful comments on and valuable improvements to our paper. Thank you also for giving me the opportunity to submit a revised draft of my manuscript (agriengineering-3156166) titled A new way to identify mastitis in cows using artificial intelligence. All changes in the manuscript were highlighted in red for reviewer 4, to facilitate the understanding of the editor and reviewer. We have incorporated the suggestions made by the reviewer. Below is a point-by-point response to the reviewer's comments and concerns.

 

  1. Dear authors, congratulations on your research. Quick and ingenious solutions are continuously needed and relevant in the diagnosis of mastitis. Tissue temperature evaluation is a technique increasingly used to assist farmers and for other purposes, such as heat detection. This study beautifully combines infrared thermography and artificial intelligence to provide a simple and fast method for diagnosing mastitis. To improve the article, I recommend a thorough literary review, as there are several sentences and phrases throughout the text that are completely out of context. Additionally, a few more targeted remarks are provided below.

 

  1. Line 17 replace By means with “using”

Author´s response: Line 18 - the term “by means of” was replaced by “using”.

 

  1. Lines 17-19 Subclinical mastitis is an infection that typically does not show visible symptoms but can still affect the health of the cow. Therefore, an animal diagnosed with subclinical mastitis would generally not be considered entirely healthy. To clarify the sentence, you might say: "The image bank used in the research consisted of 165 images, each with a resolution of 360 x 360 pixels, sourced from a database of 55 animals diagnosed with subclinical mastitis, all of which were not exhibiting clinical symptoms at the time of imaging."

Author´s response: the sentence was replaced by the reviewer's suggestion “The image bank used in the research consisted of 165 images, each with a resolution of 360 x 360 pixels, sourced from a database of 55 animals diagnosed with subclinical mastitis, all of which were not exhibiting clinical symptoms at the time of imaging.” (Lines 18-20)

 

  1. Lines 22-25 replace with this for clarity “The proposed predictive methodologies, based on knowledge transfer, were effective in accurately classifying the images. This significantly enhanced the automatic detection of both healthy animals and those diagnosed with subclinical mastitis using thermal images of the udders of dairy cows.”

Author´s response: the sentence was replaced by the reviewer's suggestion “The proposed predictive methodologies, based on knowledge transfer, were effective in accurately classifying the images. This significantly enhanced the automatic detection of both healthy animals and those diagnosed with subclinical mastitis using thermal images of the udders of dairy cows.” (Lines 24-27)

 

  1. Line 29 replace with “Early diagnosis”

Author´s response: the term early diagnosis was inserted (Line 32).

 

  1. Line 33 replace Asymptomatic with subclinic

Author´s response: replaced "Asymptomatic" with "Subclinic" (Line 35).

 

  1. Line 38 replace parlor with room

Author´s response: replaced "parlor" with "room" (Line 40).

 

  1. Lines 39-42 replace with this for clarity “In this context, developing diagnostic methods that integrate automation is crucial to minimize costs and losses, ensure accuracy and speed in diagnosis, and promote non-invasive techniques like thermography to enhance animal welfare.”

Author´s response: the sentence was replaced by the reviewer's suggestion “In this context, developing diagnostic methods that integrate automation is crucial to minimize costs and losses, ensure accuracy and speed in diagnosis, and promote non-invasive techniques like thermography to enhance animal welfare.” (Lines 40-44)

 

  1. Line 45 use the journal format for cite the references, in this case use [1,4,7,8]

Author´s response: replaced by [1,4,7,8] (Line 47).

 

  1. Lines 50-55 replace with this for clarity “In a study on dairy cows with mastitis, Zaninelli et al., 2018 [1] developed a software that automatically measures udder temperature. The research demonstrated a strong correlation between these temperatures and the somatic cell count, which significantly facilitates the diagnosis of clinical mastitis. The findings indicate that elevated udder temperature is a reliable early indicator of mastitis, allowing for quicker intervention and better management of udder health. In another study, Wilson et al., 2017 [9] compared automatic image recognition software with manual methods for detecting E. coli infections in cows. The results showed that the automated system provided comparable accuracy to manual methods, with the added benefits of increased speed and objectivity in identifying infections. This suggests that the implementation of image recognition technology could streamline the process of infection detection, improving both efficiency and reliability in herd management.”

Author´s response: replaced by the reviewer's suggestion “In a study on dairy cows with mastitis, Zaninelli et al., 2018 [1] developed a software that automatically measures udder temperature. The research demonstrated a strong correlation between these temperatures and the somatic cell count, which significantly facilitates the diagnosis of clinical mastitis. The findings indicate that elevated udder temperature is a reliable early indicator of mastitis, allowing for quicker intervention and better management of udder health. In another study, Wilson et al., 2017 [9] compared automatic image recognition software with manual methods for detecting E. coli infections in cows. The results showed that the automated system provided comparable accuracy to manual methods, with the added benefits of increased speed and objectivity in identifying infections. This suggests that the implementation of image recognition technology could streamline the process of infection detection, improving both efficiency and reliability in herd management.” (Lines 65-75).

 

  1. Line 75 delete this Determining the convergence result

Author´s response: deleted "Determining the outcome of convergence" (Line 78)

 

  1. Lines 84-86 replace with this for clarity Bayesian optimization is a hyperparameter search technique that surpasses both grid search and random search. Unlike these methods, Bayesian optimization leverages knowledge from previous iterations to inform the search process. This approach enhances decision-making by more effectively identifying the optimal hyperparameter settings for evaluating a model.

Author´s response: replaced by reviewer's suggestion “Bayesian optimization is a hyperparameter search technique that surpasses both grid search and random search. Unlike these methods, Bayesian optimization leverages knowledge from previous iterations to inform the search process. This approach enhances decision-making by more effectively identifying the optimal hyperparameter settings for evaluating a model.” (Lines 105-109).

 

  1. Line 170 replace with this The clinical status of the animals was assessed using the California Mastitis Test (CMT) before milking, after discarding the first drop of milk.

Author´s response: replaced by reviewer's suggestion “The clinical status of the animals was assessed using the California Mastitis Test (CMT) before milking, after discarding the first drop of milk.” (Lines 191-192)

 

  1. Line 170-171 delete this sentence

Author´s response: deleted sentence

 

  1. Lines 290-292 references are needed to support this statement

 

  1. Practically, sources need to be added throughout the entire discussion section.

Author´s response:

Lines 306-313: Text adjustment.

Lines 322-324: Text adjustment.

Lines 330-332: inserted literature support “The results of this study showed that the ability to detect subclinical mastitis was superior to the research carried out by [2], where the accuracy of the mastitis classification algorithm was 83.33%.”

Lines 335-339: inserted from literature “Similar results were achieved by the study carried out by [31] to detect subclinical mastitis using a deep learning model based on Convolutional Neural Networks (CNN) using 7615 udder thermograms from 40 Murrah buffaloes and had training accuracy and validation accuracy of 0.970 and 0.943, respectively. Thus, the improved deep learning CNN models efficiently predicted cases of subclinical mastitis.”

Lines 348-370: Text adjustment.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I understand and support the idea, but CMT is still insufficient.  

Author Response

Response to Reviewer 1 Comments – Round 2

 

Dear Editor,

 

We appreciate the time and effort that you and the reviewers dedicated to providing feedback on our manuscript and are grateful for the insightful comments on and valuable improvements to our paper. Thank you also for giving me the opportunity to submit a revised draft of my manuscript (agriengineering-3156166) titled A new way to identify mastitis in cows using artificial intelligence. All changes in the manuscript were highlighted in green for reviewer 1, to facilitate the understanding of the editor and reviewer. We have incorporated the suggestions made by the reviewer. Below is a point-by-point response to the reviewer's comments and concerns.

 

I understand and support the idea, but CMT is still insufficient. 

 

Lines 192-201: The test was carried out for each mammary quarter, with scores ranging from 0 to 5 being assigned, whereby at score zero there was no precipitate formation (healthy), at score 1 there was slight precipitation (trace of infection), score 2 showed moderate precipitation (subclinical mastitis), score 3 showed clear precipitation but no gel formation (subclinical mastitis), score 4 showed clear gel formation (subclinical mastitis) and score 5 showed marked gel formation (subclinical mastitis). To limit the subjectivity in interpreting the results, only those with scores between 2 and 5 were considered for the selection of animals with subclinical mastitis and the sensitivity range for detecting sick animals was 93% for SCC > 500,000 cells/ml [38].

 

Lines 209-212: A total of 165 labeled images, each with a resolution of 360 x 360 pixels, were extracted from a database comprising 55 cattle. These images were classified into two distinct groups: "Healthy" and "Subclinical Mastitis," based on the CMT which were used as the training, testing, and validation data sets.

 

  1. Brito, J. R. F.; Caldeira, G. A. V.; Verneque, R. S.; Brito, M. A. V. P. Sensitivity and specificity of the California Mastitis Test as a diagnostic tool for subclinical mastitis in quarter somatic cell count estimation. Brazilian Journal of Veterinary Research, 1997, 17(2):49-53. https://doi.org/10.1590/S0100-736X1997000200002

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors made significant corrections to the manuscript. Although I still have reservations about the content of the manuscript, I consent to its publication.

Author Response

Response to Reviewer 3 Comments – Round 2

 

Dear Editor,

 

We appreciate the time and effort that you and the reviewers dedicated to providing feedback on our manuscript and are grateful for the insightful comments on and valuable improvements to our paper. Thank you also for giving me the opportunity to submit a revised draft of my manuscript (agriengineering-3156166) titled A new way to identify mastitis in cows using artificial intelligence.

 

The authors made significant corrections to the manuscript. Although I still have reservations about the content of the manuscript, I consent to its publication.

 

We would like to thank the reviewer for all his valuable contributions in helping the article reach the quality required for publication.

Author Response File: Author Response.pdf

Back to TopTop