Next Article in Journal
The Effect of the Incorporation of a 3D-Printed Titanium Framework on the Mechanical Properties CAD/CAM Denture Base Materials
Next Article in Special Issue
Mortality Rate in Periprosthetic Proximal Femoral Fractures: Impact of Time to Surgery
Previous Article in Journal
Parametric Design of an Advanced Multi-Axial Energy-Storing-and-Releasing Ankle–Foot Prosthesis
Previous Article in Special Issue
Comparative Analysis of Cemented and Cementless Straight-Stem Prostheses in Hip Replacement Surgery for Elderly Patients: A Mid-Term Follow-up Study
 
 
Article
Peer-Review Record

Can Machine Learning Algorithms Contribute to the Initial Screening of Hip Prostheses and Early Identification of Outliers?

Prosthesis 2024, 6(4), 744-752; https://doi.org/10.3390/prosthesis6040052
by Khashayar Ghadirinejad *, Stephen Graves, Richard de Steiger, Nicole Pratt, Lucian B. Solomon, Mark Taylor and Reza Hashemi
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Prosthesis 2024, 6(4), 744-752; https://doi.org/10.3390/prosthesis6040052
Submission received: 3 May 2024 / Revised: 13 June 2024 / Accepted: 19 June 2024 / Published: 26 June 2024
(This article belongs to the Special Issue State of Art in Hip, Knee and Shoulder Replacement (Volume 2))

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript aimed to compare the effectiveness of using either Random Survival Forest (RSF) or regularised/unregularised Cox regression to account for patient and associated device confounding factors to current standard techniques. This study evaluated RSF and regularised/unregularised Cox regression using data from the Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) to detect outlier devices among 213 individual primary total hip components performed in 163,356 procedures from 1st January 2015 to the end of 2019. Device components and patient characteristics were the inputs, and time to first revision surgery was the primary outcome treated as a censored case for death. The effectiveness of the ML approaches was assessed based on the ability to detect the outliers identified  by the AOANJRR standard approach.

 

I read the article with interest

 

A)   The abstract is sufficiently developed, but a few concerns are present:

 

Comment 1: A sub-paragraph should be subdivided, rather than use the dark font at each beginning of section.

 

Comment 2: Talk about "principal objective", what are the other aims of the study? please also specify them in the introduction section.

 

B)    In the introduction, the characteristics of the Total hip arthroplasty have been accurately described, even if a little too synthetic.

 

Comment 3: You should clarify if you talk about first implant or revision from the beginning, it is not at all clear to the reader, it would be appropriate to specify it both in the abstract and in the introduction.

 

 

C)    In materials and methods, the evaluation methods have been adequately developed.

 

D)   In the results some numeric data are unclear:

 

Comment 4: It would be appropriate to make a graph of the demographic distribution of patients, talk only about implants could be confounding.

 

E)    The discussion is sufficiently developed, even if a little too synthetic.

 

Comment 5: It would be appropriate to refer to the key features of your study

Comment 6: It would be worth referring to the limitations of your study.

 

Finally, English language editing is needed.

 

Nevertheless, some major changes are needed to be considered suitable for publication.

Comments on the Quality of English Language

English language editing is needed.

Author Response

Please see the attached file.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The authors present an interesting study on the potential of artificial intelligence, using various types of machine learning analyses to find outliers in the acetabular and femoral components of total hip prostheses in the Australian registry, in comparison to traditional methods.

 

In the abstract and results, I would elaborate more on the characteristics of the analyzed population (gender, BMI, types of acetabular and femoral prosthetic components, etc.). This would be interesting for the reader beyond the results on the outliers.

 

The font is not uniform throughout the manuscript, please correct this. In the abstract, there are words in bold. In Table 3, "patient­and" the two words are joined together.

 

I would recommend clearly adding the exclusion criteria in the materials and methods section. Were both cemented and uncemented components analyzed? Prostheses for fractures and inflammatory arthritis? Post-traumatic cases?

 

The statistical analysis is explained in a thorough and easily understandable way, but I would recommend submitting the manuscript for review by a statistician experienced in machine learning algorithms.


It would also be interesting to have more details on the types of outliers (types of femoral stems, design, length, fixation philosophy, types of acetabular cups).

 

I would expand the discussion to focus more on explaining the results. Which types of femoral stems and acetabular cups are outliers, and if a particular design presented a higher risk of revision. Information that, beyond the interesting evaluation of the effectiveness of machine learning, can capture the reader's attention, especially that of the orthopedic surgeon.

Author Response

Please see the attached file.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Now the article is suitable for publication.

Back to TopTop