Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a good paper that reviews and summarizes the application of Weibull accelerated failure time model to predict MTTF/MTBF. 

The authors have gave enough details about the Weibull distribution in the paper. To help the readers to get a better understanding of the model and help them to select the best model in their studies, it is better the authors could include a paragraph comparing Weibull model to other models, e.g. logistic, lognormal, etc. Only comparing it to the Cox model is not enough.

There might be many covariates that could be included in the model. Do the authors have any suggestions about feature selection in Weibull AFT model ?

Author Response

Please see the attachment.

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

“Using the Weibull accelerated failure time regression model to predict time to health events”

The authors showed how the Weibull AFT model, which is popular in the engineering world, can be utilized for statistical predictions in health science. They assessed the accuracy of the method in the medical context, with the help of records of 90 cancer patients.

The research is relevant to the field of statistical predictions, especially for diseases. The authors have illustrated the statistical application of the model, although this is still in the preliminary stage.

I pick on the recommendation for employment of larger datasets for more rigorous assessments.

 

The manuscript is well-written and the content is well elaborated. The conclusions are consistent with the insights provided in the methodology and results.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The abstract is too brief. Could you enter why you use this method by adding a minimum of background?

 

Even the introduction is too reductive.

 

In biostatistics, the Cox model is the most used model in survival analysis or, in any case, in time-dependent events. It is necessary to make the clinical reader understand this procedure's advantages.

 

The article is too technical; we need to include discussions and background that will encourage even a reader who is less expert in the method to read and then cite the paper.

 

I suggest the authors enrich the applied part with figures and output that capture attention.

 

Even if the dataset can be downloaded, I would avoid putting R output, especially if it doesn't describe the dataset variables: what is diagyr? instead, use Kaplan-Meier curves and tables of patients at risk.

 

Remove the R code from line 182 to line 198 and add a table for the model obtained using tab_model (sjPlot) reporting HR (95%CI) and p-value

Author Response

Please see the attachment.

Author Response File: Author Response.pdf