Artificial Intelligence Applied in Early Prediction of Lower Limb Fracture Complications

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Restructure the introduction to separately for immediate and less immediate complications

 

Standardize citations throughout the manuscript

 

incidence rates could be enhanced by using consistent units and providing contextual significance

 

Strengthen the introduction of AI by linking it to the detection and management of the previously discussed complications

 

Define specialized terms and concepts(Virchows triad and specific fracture types)

 

Why did authors choose not to implement any exclusion criteria in your study?

 

Can the authors elaborate on the process used to extract data from the medical charts?

 

Did you obtain ethical approval from a review board for this study?

 

What specific statistical methods were applied using IBM SPSS in your analysis?

 

How were the neural networks developed and validated in your AI algorithm?

 

All figures must be scientifically edited, charts also.

Remove screenshots of tables/software and create the tables according to journals recommendation in Instructions for Authors

Author Response

My answer.

Thank you for your collaboration.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

This  study analyze medical charts from the Clinical Emergency Hospital in Galați, Romania, found a higher prevalence of female patients and the highest incidence of deep venous thrombosis (DVT) among patients aged 71-90 years. Patients with a high BMI accounted for 15.69% of DVT patients. Elevated values of direct bilirubin and prothrombin activity were identified as significant predictors of DVT development.

However, some Recommendations for Improvement should be addressed before publishing.

1-    The abstract is poorly written, it should describe the method that is used in the experiments and summery of the critical results that is obtained

2-    In the introduction section, please explain the importance and innovation of the proposed solution. In addition, please clearly state the main contribution of the manuscript.

3-    The author should discuss in more detail the neural model configuration, used in the experiments detailing the range of hyper-parameters such as number of hidden neurons, learning rate and other parameters considered or method to select the best hyper-parameter values.

4-    Table3 can be converted to many feature engineering graphs, will be better for more understanding the nature of features and its coloration to the target.

5-    Adding correlation matrix between features, will enhance understanding the relationship between features.

6-    The authors say “statistical analysis was performed using IBM SPSS 29.0.2.0.”, they should clarify what are the statistical analysis they used,  None of the mathematical formula were developed by the authors; they presented generic principles.

7-    It’s better for the authors to employee different models rather than neural network such as logistic Regression model,  KNN, NB, DT and SVM, etc and compare the results for each model with proposed model

8—It is recommended that the authors expand the conclusion section to include a detailed discussion of the proposed model's limitations and future research directions.

 

 

Author Response

My answer.

Thank you for your collaboration.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All my comments were answered correspondingly. The work is ready to be published now.

Reviewer 2 Report

Comments and Suggestions for Authors

- ِAll of my comments were considered by the authors, Thanks to authors