mit-lcp / critical-data-book Goto Github PK

Looks like some encoding issues with the chapter04_code.Rfile. I noticed it only because github is detecting it as rebol.

The proofs seem to have different chapter numbers than what is listed here.

They should obviously match.

I found more than 1500 patients with Afib RVR and are treated by the three drugs mentioned in chapter 23. But there are only about 370 of them having recorded heartrates in the chartevents table (MIMIC-III). I checked that only about 20000 patients have heart rate measurements in chartevents. So do I need to use the ECG data for heartrate or do you guys have other suggestion for dealing with those missing heartrates records?

An error occurs in propensity_score_matching.r

tryCatch( .jcall(m, "V", "run"),

• NumberFormatException = function(e) {
• e$jobj$printStackTrace()
• }
• );
  다음에 오류가 있습니다.jcall(m, "V", "run") : java.lang.StackOverflowError

How can we solve it?

The code for Section 2, Chapter 3c needs a fair amount of work to make it easy to follow and reuse:
https://github.com/MIT-LCP/critical-data-book/tree/master/section2/chapter3c/code

One issue is that the code includes several files with the .m~ extension, which I assume are temporary files. In some cases these were duplicates of other files, so I removed them, but there are still some remaining.

Should the .m~ files be renamed with a .m extension or can they be removed?

I was wondering if the Python script used in Chapter 28. Improving Patient Cohort Identification Using Natural Language Processing can be made available.
Thanks.

As noted in, for example, section2/chapter3c

Note: the dataset (aline.mat) was not committed.

The dataset should be made easily available to the reader.

I have uploaded the aline dataset used for chapter 5 to physiobank. We need to determine if this is the same dataset used in chapter 3, and if the csv file can be imported, converted to the matlab data structure.

See: http://physionet.org/physiobank/database/mimic2-iaccd/

Several of the chapters (for example, https://github.com/MIT-LCP/critical-data-book/tree/master/section2/chapter3c and https://github.com/MIT-LCP/critical-data-book/tree/master/section2/chapter3d) require associated data (in these cases, an arterial line dataset).

The data is sensitive and so cannot be included as part of the public repository. Instead we could provide:

• code for generating the dataset from MIMIC and/or
• sample data

In Ch18, trend analysis required the exact year for each record, which has been anonymized. Could you please provide the data you used for analysis in the Chapter?

We should set up a testing framework for the code in this repo.

Originally posted by @sanyaa2 in #8 (comment)

critical-data-book/LICENSE

License is currently MIT license. Need to clarify what license the code will be under, especially since most of the code also appears in some form in the book.

In the chartevents.csv table we downloaded,

one CHARTTIME, is 5/12/34 12:00

Can someone help with what does 34 mean? How to translate to 34 to a correct calendar year.

Thanks

Jin

Impute has been spelt as 'inpute' throughout the code in chapter 3c:
https://github.com/MIT-LCP/critical-data-book/tree/master/section2/chapter3c/code