UMLS_mapping

This project's purpose is to convert the ICD9CM code to CUI. Since there is no way to request the CUI using the IDC9CM, we do it in 2 steps. First, get the name using the ICD9CM. Then use the name to get the CUI.

This project has the following files:

Authentication.py

The Authentication code is to generate a ticket number that we will use to connect to the API

UMLS_assignment.py

Same as UMLS_assignment.ipynb but in python code formate

UMLS_assignment.ipynb

The code does the following:

• Read the source file that contains the ICD9 codes (df)
• Clean the data by adding a decimal to ICD9 codes based on the rules in https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/HospitalQualityInits/Downloads/HospitalAppendix_F.pdf
• Put the unique ICD9 codes in a new table (ICD9_CUI_mapping)
• Connect to the API and get the name and CUI code. Also, add the name and CUI into ICD9_CUI_mapping table
• Save this table into ICD9_CUI_mapping.csv
• Merge the mapping and the original (ICD9_CUI_mapping & df)
• Save the result as CSV file (FinalResult.csv)

source_data.csv

• The patient data with ICD9 codes

ICD9_CUI_mapping.csv

• A table contains the unique ICD9 codes with its name and CUI code

FinalResult.csv

• The original file with CUI codes

Contributors

• Wejdan Bagais
• Naren Durbha

GitHub Link

• https://github.com/wbagais/UMLS_mapping