|
a |
|
b/README.md |
|
|
1 |
 |
|
|
2 |
|
|
|
3 |
Modeling Clinical Trial Attrition Using Machine Intelligence: |
|
|
4 |
A driver analytics case study using 1,325 trials representing one million patients |
|
|
5 |
--------------------------------------------------------------------------------------------------------------- |
|
|
6 |
This file is the readme.txt file for the code folder that contains the R code that was used in the model |
|
|
7 |
|
|
|
8 |
|
|
|
9 |
TABLE OF CONTENTS |
|
|
10 |
----------------- |
|
|
11 |
|
|
|
12 |
* Full Author List |
|
|
13 |
|
|
|
14 |
* Introduction |
|
|
15 |
|
|
|
16 |
* Requirements |
|
|
17 |
|
|
|
18 |
* Folder Contents |
|
|
19 |
|
|
|
20 |
* Contact |
|
|
21 |
|
|
|
22 |
Full Author List |
|
|
23 |
---------------- |
|
|
24 |
Emmette Hutchison, Youyi Zhang, Sreenath Nampally, Imran Khan Neelufer, Vlad Malkov, Jim Weatherall, Faisal Khan and Khader Shameer |
|
|
25 |
|
|
|
26 |
|
|
|
27 |
Introduction |
|
|
28 |
------------ |
|
|
29 |
Patient attrition, also referred to as dropout or patient withdrawal, occurs when patients enrolled |
|
|
30 |
in a clinical trial either withdraw or are lost to follow-up by the clinical site and trial sponsor. |
|
|
31 |
|
|
|
32 |
|
|
|
33 |
Requirements |
|
|
34 |
------------ |
|
|
35 |
This was performed using RStudio and R. The versions of RStudio and R are listed below: |
|
|
36 |
RStudio 1.0.44, which can be downloaded from https://rstudio.com/products/rstudio/download/ |
|
|
37 |
R 3.5.2 (2018-12-20). RStudio can be installed from https://cran.r-project.org/mirrors.html |
|
|
38 |
|
|
|
39 |
The project requires the following R packages. The version numbers indicate the version of the packages |
|
|
40 |
that were used in the analysis. Please install the following packages using the command below in your R |
|
|
41 |
Environment |
|
|
42 |
|
|
|
43 |
SuperLearner==2.0-26 |
|
|
44 |
MASS==7.3-51.5 |
|
|
45 |
ranger==0.12.1 |
|
|
46 |
ipred==0.9-9 |
|
|
47 |
kernlab==0.9-29 |
|
|
48 |
arm==1.10-1 |
|
|
49 |
dplyr==1.4.2 |
|
|
50 |
caret==6.0-84 |
|
|
51 |
parallel==3.5.2 |
|
|
52 |
|
|
|
53 |
Folder Contents |
|
|
54 |
---------------- |
|
|
55 |
This folder contains the data files that was used in the analysis. The file descriptions are listed below: |
|
|
56 |
|
|
|
57 |
``` |
|
|
58 |
|---- readme.md : readme file for the data folder |
|
|
59 |
| |
|
|
60 |
|---- code : This folder contains the code required for the pre-processing of raw data and running the predictive model results, |
|
|
61 |
| further explanation is available in a readme file within this folder |
|
|
62 |
| |
|
|
63 |
|---- data: This folder contains two sub-folders, another readme file in the folder will explain the overview on each folder content. |
|
|
64 |
|---- analysis_ready |
|
|
65 |
|---- requirments.txt : is a file which lists the libraries used in the model building exercise, this file can be used to install the packages needed. |
|
|
66 |
| |
|
|
67 |
| Using command "Install.Packages("package name") one can easily install the required packages |
|
|
68 |
``` |
|
|
69 |
|
|
|
70 |
Contact |
|
|
71 |
-------- |
|
|
72 |
shameer.khader@astrazeneca.com |