Skip to content

Latest commit

 

History

History
24 lines (18 loc) · 1.28 KB

README.md

File metadata and controls

24 lines (18 loc) · 1.28 KB

FinalProject

Our final project purpose to examine the impact of Surgeon and Hospital Procedural Volume on Outcomes after CABG procedure .

python files:

Bins.py - Divided each line of yearly experience to bins.

GLMM.py - GLMM python model.

Plots_trend.py - Generates our various plots depends on selected experience.

Reop SHAP output.py - Run the final models on Reop Data and generates confusion matrix and SHAP plots.

RetrainModelComplics.py - Retrain model for Complication models.

RetrainModels.py - Retrain model for Mortality models.

SHAP output.py - Run the final models and generates confusion matrix and SHAP plots.

SelectedYears.py - generates summaries tables of the experience of selected target variable.

tables.py - create six tables with analysis. each type of mortality : 'total_surgery_count', 'total_CABG', 'Reop'

with each outcome: 'Mortalty', 'Complics'. The data for creating the tables is in Tables directory.

and its name is HospID/surgid+'_allyears_expec_surgid_STSRCOM.csv'


Directories:

Tables dir - Contains all the data we used on the various scripts.

ClusterFiles - Script that we run on cluster, for grid search and SMOTE SHAP outputs.