This program analyses a series of creatinine laboratory values and calculates the number of AKI, and their dates. It also has the ability to plot the results.
The program is written in Python 3. It requires Python 3.x, and the following packages installed: numpy, scipy, matplotlib, and pandas. In order to use the program, you have to include patients creatinine values in the input folder as Labsxx.csv files. If you also want the estimated glomerular filtration rate, then a Demographics.csv file is needed with patients identifiers, age, gender, and race, as these variables are used in the GER equation.
Please see the current files in the input folder as an example of how the files are structured.
Demographics.csv
MRN Age Gender Race
1 50 MALE WHITE
2 55 FEMALE BLACK
3 44 FEMALE ASIAN
4 58 MALE BLACK
5 61 FEMALE WHITE
Labs01.csv
MRN Creatinine1 TestDate1
1 1 2001-09-12
1 1.2 2001-09-13
1 1.1 2001-09-14
1 1.2 2001-09-15
Upon running AKIPredictor.py, the information in the Input folder is proccessed, and the following files are written in the Output folder:
aki.csv: This files contains a list of all patients, their estimated baseline creatinine, and baseline GFR, number of AKI episodes based on the AKIN criteria.Datesfolder: This folder contains filed named by the patient's MRN, and lists the dates when AKI were detected for these patients listed.Graphsfolder: This folder contains a list of.pngnames by patients MRNs, and illustrate the patients creatinine trend. The points identified as AKI episodes are highlighted in red.
Here is an example of the output:
MRN baseCr eGFR numAKI anyAKI CKD
1 1.0 87.3 2 True 2.0
2 0.9 83.4 1 True 2.0
3 0.9 77.7 1 True 2.0
4 0.7 120.5 4 True
5 0.8 79.5 0 False 2.0
