get_patient_count.py

import os
import json
import pandas as pd
import numpy as np

patients_codes = {}
doac_patients_codes = {}

dates = [
    "input_2019-01-01.csv",
    "input_2019-02-01.csv",
    "input_2019-03-01.csv",
    "input_2019-04-01.csv",
    "input_2019-05-01.csv",
    "input_2019-06-01.csv",
    "input_2019-07-01.csv",
    "input_2019-08-01.csv",
    "input_2019-09-01.csv",
    "input_2019-10-01.csv",
    "input_2019-11-01.csv",
    "input_2019-12-01.csv",
    "input_2020-01-01.csv",
    "input_2020-02-01.csv",
    "input_2020-03-01.csv",
    "input_2020-04-01.csv",
    "input_2020-05-01.csv",
    "input_2020-06-01.csv",
    "input_2020-07-01.csv",
    "input_2020-08-01.csv",
    "input_2020-09-01.csv",
    "input_2020-10-01.csv",
    "input_2020-11-01.csv",
    "input_2020-12-01.csv",
    "input_2021-01-01.csv",
    "input_2021-02-01.csv",
    "input_2021-03-01.csv",
    "input_2021-04-01.csv",
    "input_2021-05-01.csv",
    "input_2021-06-01.csv",
    "input_2021-07-01.csv",
    "input_2021-08-01.csv",
    "input_2021-09-01.csv",
    "input_2021-10-01.csv",
    "input_2021-11-01.csv",
    "input_2021-12-01.csv",
]

patients_list = []
doac_patients_list = []
for file in os.listdir("output"):

    if file in dates:
        date = file.split("_")[-1][:-4]

        df = pd.read_csv(os.path.join("output", file))

        patients = np.unique(df["patient_id"])
        patients_list.extend(patients)

        doac_subset = df[df["doac"] == 1]
        doac_patients = patients = np.unique(doac_subset["patient_id"])
        doac_patients_list.extend(doac_patients)


unique_patients = len(np.unique(patients_list))
unique_patients_doac = len(np.unique(doac_patients_list))

count_df = pd.DataFrame(
    [
        ["mechanical_valve", unique_patients],
        ["mechanical_valve_doac", unique_patients_doac],
    ],
    columns=["group", "count"],
)
count_df.to_csv("output/patient_count.csv")