Ruminating, Eating, and Locomotion Behavior Registered by Innovative Technologies around Calving in Dairy Cows

Question  

Answers  

 L85:  Benaissa et al., number 19, but in References, number 19 is Fadul et al… (L410)

 Corrected

L89: “…visualbservations.”  ??  Typing error, in my opinion.

Corrected to – „visual observations”

Numbering of subsections is inaccurate (2.1, 2.2, 2.2.1, 2.4, 2.5), 2.3 is missing. This should be corrected.

Corrected

L145: “… measurements -  .”  Typing error?

Corrected

L155: “… (independent indicator) and  .”  Incomprehensible sentence.

Corrected

Regression equation on L166-167 does not agree with presented on Figure 1.

Corrected to. –„ ..increase after calving (y = -0,0537x + 21,453 R² = 0,0341),..”

Numbering of figures is inaccurate:  Fig. 1, 2, 4, 5… (number 3 is omitted).

Corrected

Question  

Answers  

The simple summary is poorly written and misleading;

We rewrote whole simple summary –

“Simple summary. The use of calving detection technology is currently increasing in animal reproduction farming. In our current study, we decided to test the hypothesis: can we predict the onset of calving using sensor data registered by innovative technologies. According our results we concluded that using network machine-learning algorithms to combine rumination time, eating time, drinking, activity and down time characteristics resulted in sensitive and particular alarms at the daily from ten days before calving. We found hat sensor data collected by new technologies for ruminating, eating, and moving can help us predict calving time. For calving prediction, we can use rumination time, eating time, drinking, activity and down time. Rumination time, eating time, drinking time, and activity all decreased ten days before calving; drinking gulps decreased on the ninth and second days before calving; and down time decreased two days before calving. These data could be used by machine learning-based calving prediction algorithms for effective retrospective calving prediction”

Abstract section: It lacks the main conclusion of the study. The way you present results is very hard to follow and read.

Please rewrite the entire abstract section, with aim, material and methods, results and discussion (presented clearer) and the main conclusions/implications of your study.

We rewrote whole abstract section –

“Abstract: Hypothesis for this study was that can we predict the onset of calving using sensor data registered by innovative technologies. Based on this the aim of this study was to identify indicators for calving prediction using sensor data, described rumination, eating, and locomotion behavior around calving as registered by innovative technologies.

54 cows were chosen from the entire herd without calving problems or other health problems. For registration of rumination, eating, and locomotion behaviors, the Rumiatch sensors (RWS) was used. RWS recorded rumination, eating, and locomotor behavior (rumination time, eating time, drinking time, drinking gulps, bolus, chews per minute, chews per bolus, activity up and down time, temp average, temp minimum, temp maximum, activity change, other chews, ruminate chews, eating chews). RWS sensors were placed one month before expected calving and removed ten days after calving. Data were registered 10 days before and 10 days after calving.

We found that rumination time had a tendency to decrease by 0.0537 min/h ten days before and ten days after calving. On calving day, we recorded a sudden drop, the eating time had a tendency to decrease by 0.0214 min/h ten days before and ten days after calving. Differences in drinking time were found between the third day after calving and the tenth day before calving, but in general drinking time was lower 10 days before calving day, compared to 10 days after calving.  Drinking gulps differed between the ninth and second days before calving compared to the calving day. Activity was found to be significant between the first, third and fith day after calving and the ninth and second day before calving. Atctivity had a tendency to decrease by 0.6337 10 days before and after calving. Down time between had a tendency to decease aprroximatlety 2 days before calving.

Based on the results of this study, we can say that sensor data collected by new technologies for ruminating, eating, and moving can help us predict calving time”

Introduction: As far as I know, PDT is PLF (precision livestock farming), please clarify and change throughout text;

Corrected to - PLF (precision livestock farming)

Lines 106-123: Please rewrite this section, you have made this subsection in a haste, repeating information and having sentences that are not grammatically correct.

Corrected to – „60 cows were chosen from the entire herd. Cows were selected according to the following factors: cows has to be within 30 days of calving of the Lithuanian black and white breed, with two or more lactations, 500 – 570 kg body weight, and an average productivity of previous lactation - 11000 kg milk per year, with dry cows consuming 14 kg DM/day and fresh cows consuming 26 kg DM/day. A general clinical examination revealed that none of the cows had clinical indications compatible with any disease. Cows having clinical symptoms of diseases (mastitis, metritis) (n = 6) were excluded from the research. The total number of cows in this group was 54. The cows in the study had no calving problems or other health problems”

Lines 128-144: Please be more careful when describing the behavioural parameters, do consult with other authors who better defined they behavioural patterns, e.g. ‘Total number of gulps eaten while drinking’

Corrected according Zehner et al.

Figure 3 is lacking;

Corrected

Figure 6: showing the number of chews per bolus, I find this averages very unlikely, usually cows chew each bolus 20-30 times. Please use some published studies to validate/back-up your findings.

We added information in methodology section - Table 3.Parameters recorded by RWS (Zehner et al. [17]).   

“Chews per bolus (CB) - Chews performed during rumination between the

regurgitation and swallowing of 1 bolus”

Discussions: Please be careful when stating ‘These investigations began with the goal of regulating parturient cows and preventing dystocia and metritis.’. How one regulates calving? And how predicting the time of calving throughout the use of PLFs is preventing dystocia and metritis in cattle?! Dear authors, with all the respect for your work, this sort of statements are unacceptable.

We corrected and deleted this sentence

Rumination time it is not a biomarker as stated by the authors, please consult the available literature.

We corrected to – “Rumination time is a sensitive indicator of dairy cow health that is used in automated systems to detect early disease onset”

Line 313: Longer drinking durations around calving indicate that cows are hungrier? It makes absolutely no sense, are cows known to drink more water when they are hungry? The answer is no, water intake in the species it is corelated with dry matter intake, milk yield and outside temperature mainly.

We agree with you and deleted this sentence

I struggled the most with the discussion section to be honest, it is very poorly written and it makes little sense.

We corrected all discussion section with more concentration in our explanation of our results.

The conclusion is poorly written and makes absolutely no sense, and it sounds like ‘According to our hypothesis which was that changes in rumination, eating, and locomotion behavior before and after calving can be registered with innovative technologies, and the aim - to test this hypothesis, we planned to use innovative technologies to assess rumination, eating, and locomotion behavior 10 days before and 10 days after calving; we found from 9 to 3 days before calving reduction in rumination, eating behavior, such as rumination time, eating time, other chew, ruminate chew, drinking time and drinking gulp.

We rewrote the entire conclusion section – 

“Precision livestock farming systems, which are used to generate health and estrus alerts, effectively quantify behavioral changes around calving. Based on the results of this study, we can say that sensor data collected by new technologies for ruminating, eating, and moving can help us predict calving time. For calving prediction, we can use rumination time, eating time, drinking, activity and down time. Rumination time, eating time, drinking time, and activity all decreased ten days before calving; drinking gulps decreased on the ninth and second days before calving; and down time decreased two days before calving. These data could be used by machine learning-based calving prediction algorithms for effective retrospective calving prediction.

Using network machine-learning algorithms to combine rumination time, eating time, drinking, activity and down time characteristics resulted in sensitive and particular alarms at the daily from ten days before calving. Future research on calving event prediction should focus on longer period before calving and with more clinically healthy cows, using this dataset to make a prediction model using machine learning techniques. Also, calving events need to be tracked over shorter periods of time that farmers can get notifications they can use to make smart management decisions. Another use for calving prediction tools would be to distinguish between eutocial and dystocial calvings”

In my opinion, the authors made this manuscript in a haste, without investing enough of their time to write a proper quality article, with clear findings and implications.

Strongly advise for this manuscript to be either completely rewritten in a more professional manner and resubmitted for revision, with authors investing time and efforts, or to be rejected.

We corrected all parts of manuscript –

We changed title of manuscript – “Can We Predict Calving By Using Innovative Technologies Registered For Ruminating, Eating, And Locomotion Behavior?”

Hypothesis and aim – “Hypothesis for this study was that can wepredict the onset of calving using sensor data registered byinnovative technologies. Based on the hypothesis, the aim of this study was to identify indicators for calving prediction using sensor data, described rumination, eating, and locomotion behavior around calving as registered by innovative technologies”

Also, according that we made correction in whole manuscript.

Question  

Answers  

The hypothesis and study is not new: we already know that behaviour changes before calving, and that we can measure this with sensors. This has been shown in several studies. An interesting hypothesis would be whether we can predict the onset of calving with sensor data.

We corrected the title, hyphotesis and aim of this study. Also, we rewrote the discussion and conclusion sections.

The data that is gathered is interesting, and probably enough to answer the question above. But, the authors have only used descriptive analyses to show the behaviours before and after calving. Differences between days are calculated but those are not very interesting, if you do not use these results to make a prediction. With the use of machine learning techniques, you could develop a prediction model for the onset of calving. That would be interesting, and worth a publication, if you get nice results with these data - otherwise you need to gather more data of calving cows wearing sensors.  The authors even refer to studies where predictions have been done, e.g. Miller et al. 2020, Titler et al., 2015, and Ouellet et al., 2016.

Prediction equations or models are based on correlations. In our research work correlations between  trait from RumiWatch noseband sensor readings and a day before and after calving were  low and not statistically significant. As You can see in the figure bellow.

Regression analysis was used  for determination of the relationship between a single dependent ( trait from RumiWatch noseband sensor readings) variable and one  independent /predictor (a day before and after calving) variables. The analysis yields a predicted value for the criterion resulting from a linear combination of the predictors.

We think for such a modelling analysis a higher number of cows should be investigated.

I would advise the authors to change their hypothesis and the title to 'can we predict the onset of calving using sensor data?' and to use this dataset to make a prediction model using machine learning techniques. 

We changed title to – “Can We Predict Calving By Using Innovative Technologies Registered For Ruminating, Eating, And Locomotion Behavior?”

Hypothesis and aim to – “Hypothesis for this study was that can we predict the onset of calving using sensor dataregistered by innovative technologies. Based on the hypothesis, the aim of this study was to identify indicators for calving prediction using sensor data, described rumination, eating, and locomotion behavior around calving as registered by innovative technologies”