Analyzing the Data Completeness of Patients’ Records Using a Random Variable Approach to Predict the Incompleteness of Electronic Health Records
, Paniz Abedin 2,‡, Sahar Hooshmand 3,‡ and Muhammed Shelleh 4,‡
Round 1
Reviewer 1 Report
The proposed algorithm and the corresponding results are fairly interested. and the subject is well written, although rather short in analysis.
However, some minor correction needed, like to rewrite in a clear manner all the formulations in the manuscript (e.g. using some third party equation editors like MathType the authors are writing in MS Word).
Moreover, when authors you use "Where..." after some formulas, they should continue with ",... where...".
The reviewer
Author Response
Reviewer 1
Comment 1.1: The proposed algorithm and the corresponding results are fairly interested. and the subject is well written, although rather short in analysis.
Response 1.1: We truly appreciate the comment from the reviewer.
Comment 1.2: However, some minor correction needed, like to rewrite in a clear manner all the formulations in the manuscript (e.g. using some third party equation editors like MathType the authors are writing in MS Word).
Response 1.2: All mathematical formulations are now re-written. We truly appreciate the reviewer pointing at this mistake.
Comment 1.3: Moreover, when authors you use "Where..." after some formulas, they should continue with ",... where...".
Response 1.3: This has now been fixed.
Reviewer 2 Report
This article presents a study dedicated to solving an urgent problem - analyzing the data completeness of patients’ records. The work was done at a fairly high level, but to improve the quality it needs to be slightly reworked.
1. In the Introduction, it is necessary to perform a more qualitative analysis of existing methods for analyzing the data completeness, to note the advantages and disadvantages of known methods.
2. The algorithm of data completeness analysis proposed in the article is some interest, however, its comparison with known algorithms, including neural networks algorithms, unfortunately, made only at a descriptive level. Thus, in the Discussion section it is necessary to present a critical analysis of the proposed algorithm and its objective comparison with known methods.
3. The quality of figures needs to be improved. Explanatory labels and numerical values on graphs are almost impossible to read.
4. Mathematical expressions for f (x, k, s) and Ccost are displayed incorrectly in the text.
5. References should be formatted as specified in Instructions for Authors. See https://www.mdpi.com/journal/applsci/instructions
Author Response
Reviewer 2
Comment 2.1: In the Introduction, it is necessary to perform a more qualitative analysis of existing methods for analyzing the data completeness, to note the advantages and disadvantages of known methods.
Response 2.1: This article is builds on the algorithm invented by the first author to analyzing data incompleteness of electronic health records. This has now been clearly stated in the introduction section.
Comment 2.2: The algorithm of data completeness analysis proposed in the article is some interest, however, its comparison with known algorithms, including neural networks algorithms, unfortunately, made only at a descriptive level. Thus, in the Discussion section it is necessary to present a critical analysis of the proposed algorithm and its objective comparison with known methods.
Response 2.2: This ties to our previous response where this type of discussion cannot be performed since the article builds on a unique method invented by the first author.
Comment 2.3: The quality of figures needs to be improved. Explanatory labels and numerical values on graphs are almost impossible to read.
Response 2.3: The authors have now fixed this problem to a large extent.
Comment 2.4: Mathematical expressions for f (x, k, s) and Ccost are displayed incorrectly in the text.
Response 2.4: We have now fixed this problem.
Comment 2.5: References should be formatted as specified in Instructions for Authors. See https://www.mdpi.com/journal/applsci/instructions
Response 2.5: The references are now properly formatted.
Reviewer 3 Report
This paper presents several algorithms developed by the authors for incompleteness analysis of data sets of health records. Based on these analyses, the authors also intend to predict what parts of health records are more likely to be incomplete.
The paper is well structured, and the essence of the research presented in the paper can be grasped with ease by the reader. The experiments presented by the authors have substance and provide a good basis for the analysis of data incompleteness and incompleteness prediction of patient records in the context described by the authors. The presentation of the algorithms’ design is easy to understand.
However, trusting the solidity of the experiments and understanding the paper are hampered by inappropriate use of English in a text that is dense in statistics terminology.
Some of the mathematical formulas presented in the paper are incorrectly laid out and some established names and methods in probabilities and statistics are repeatedly misspelled.
Citations become questionable when the bibliographical entries do not provide the names of the authors. References include online sources which are not formatted in accordance with an academic style.
The sources of the data samples used in the experiments are not provided. The data sources must be specified using a standard academic style even when these sources are considered common knowledge by the authors and are easy to access by specialists in their research area.
The figures presented in the paper need captions that summarize the meaning of the presented data even if this meaning appears self explanatory for the authors by just looking at the graphics.
A careful revision of the English language and academic style used in the paper is necessary. References must revised and correctly formatted in accordance with the chosen academic style.
Author Response
Reviewer 3
Comment 3.1: However, trusting the solidity of the experiments and understanding the paper are hampered by inappropriate use of English in a text that is dense in statistics terminology.
Response 3.1: We have now worked on the English and eliminated a few glaring mistakes.
Comment 3.2: Some of the mathematical formulas presented in the paper are incorrectly laid out and some established names and methods in probabilities and statistics are repeatedly misspelled.
Response 3.2: We have now fixed the problem.
Comment 3.3: Citations become questionable when the bibliographical entries do not provide the names of the authors. References include online sources which are not formatted in accordance with an academic style.
Response 3.3: The citations are now in the right format.
Comment 3.4: The sources of the data samples used in the experiments are not provided. The data sources must be specified using a standard academic style even when these sources are considered common knowledge by the authors and are easy to access by specialists in their research area.
Response 3.4: The beginning part of section 4 provides this information.
Comment 3.5: The figures presented in the paper need captions that summarize the meaning of the presented data even if this meaning appears self explanatory for the authors by just looking at the graphics.
Response 3.5: We thank the reviewer for this comment.
Comment 3.6: A careful revision of the English language and academic style used in the paper is necessary. References must revised and correctly formatted in accordance with the chosen academic style.
Response 3.6: We have now fixed these problems.
