Code fragments of main.py

• import os
• import os.path as osp
• import numpy as np
• import pandas as pd
• import matplotlib.pyplot as plt
• import glob
• from fhirpy import SyncFHIRClient
• ## Define directory of output
• here = osp.dirname(osp.abspath(__file__))
• out_dir = osp.join(here, 'output')
• if not os.path.exists(out_dir):
• os.makedirs(out_dir)
• runs = sorted(glob.glob(os.path.join(out_dir, 'run_*')))
• run_id = int(runs[-1].split('_')[-1]) + 1 if runs else 0
• experiment_dir = os.path.join(out_dir, 'run_{}'.format(str(run_id)))
• if not os.path.exists(experiment_dir):
• os.makedirs(experiment_dir)
• ## Define (output) file formats
• if not osp.exists(osp.join(experiment_dir, 'report.txt')):
• with open(osp.join(experiment_dir, 'report.txt'), 'w') as f:
• pass
• ## Define (input) variables from Docker Container environment variables
• fhir_server = str(os.environ['FHIR_SERVER'])
• fhir_port = str(os.environ['FHIR_PORT'])
• ## Load aggregated results from previous stations
• # if run_id > 0:
• # prev_experiment_dir = osp.join(out_dir, 'run_{}'.format(str(run_id - 1)))
• # if osp.exists(osp.join(prev_experiment_dir, '')):
• # pass
• # else:
• # pass
• # else:
• # pass
• ## Collect Data Statistic
• # Create an instance
• client = SyncFHIRClient('http://{}:{}/fhir'.format(fhir_server, fhir_port))
• # Search for patients
• # Return lazy search set
• patients = client.resources('Patient')
• patients_data = []
• for patient in patients:
• patient_birthDate = None
• try:
• patient_birthDate = patient.birthDate
• except: pass
• # patinet_id, gender, birthDate
• patients_data.append([patient.id, patient.gender, patient_birthDate])
• patients_df = pd.DataFrame(patients_data, columns=["patient_id", "gender", "birthDate"])
• # Search for media
• media_list = client.resources('Media').include('Patient', 'subject')
• media_data = []
• for media in media_list:
• media_bodySite = None
• media_reasonCode = None
• media_note = None
• try:
• media_bodySite = media.bodySite.text
• except: pass
• try:
• media_reasonCode = media.reasonCode[0].text
• except: pass
• try:
• media_note = media.note[0].text
• except: pass
• media_data.append([media.subject.id, media.id, media_bodySite, media_reasonCode, media_note, media.content.url])
• media_df = pd.DataFrame(media_data, columns=["patient_id", "media_id", "bodySite", "reasonCode", "note", "image_url"])
• data_df = pd.merge(patients_df, media_df, on='patient_id', how='outer')
• # Python3 code to calculate age in years
• import datetime
• from datetime import date
• def calculateAge(born):
• if born is None:
• return None
• born = datetime.datetime.strptime(born, '%Y-%m-%d').date()
• today = date.today()
• try:
• birthday = born.replace(year=today.year)
• except ValueError: birthday = born.replace(year=today.year, month=born.month + 1, day=1)
• if birthday > today:
• return today.year - born.year - 1
• else:
• return today.year - born.year
• data_df['age'] = data_df['birthDate'].map(lambda x: calculateAge(x))
• with open(osp.join(experiment_dir, 'report.txt'), 'a') as f:
• f.write('Stations have {} instances with complete attributes \n'.format(data_df[(data_df['age'].notna()) & (data_df['note'].notna()) & (data_df['gender'].notna())].shape[0]))
• f.write('Station has {} instances whose attribute \"age\" is missing\n'.format(data_df['age'].isna().sum()))
• f.write('Station has {} instances whose attribute \"gender\" is missing\n'.format(data_df['gender'].isna().sum()))
• f.write('Station has {} instances whose attribute \"bodySite\" is missing\n'.format(data_df['bodySite'].isna().sum()))
• data_df['MEL'] = data_df['note'].map(lambda x: 1 if x == 'MEL' else 0)
• data_df['NV'] = data_df['note'].map(lambda x: 1 if x == 'NV' else 0)
• data_df['BKL'] = data_df['note'].map(lambda x: 1 if x == 'BKL' else 0)
• data_df['DF'] = data_df['note'].map(lambda x: 1 if x == 'DF' else 0)
• data_df['SCC'] = data_df['note'].map(lambda x: 1 if x == 'SCC' else 0)
• data_df['BCC'] = data_df['note'].map(lambda x: 1 if x == 'BCC' else 0)
• data_df['VASC'] = data_df['note'].map(lambda x: 1 if x == 'VASC' else 0)
• data_df['AK'] = data_df['note'].map(lambda x: 1 if x == 'AK' else 0)
• age_df = data_df[['MEL','NV','BCC','AK','BKL','DF','VASC','SCC','age']]
• def replace_age(row):
• if row['age'] == np.nan:
• return pd.Series([np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan])
• elif row['MEL'] == 1.0:
• return pd.Series([row['age'], np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan])
• elif row['NV'] == 1.0:
• return pd.Series([np.nan, row['age'], np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan])
• elif row['BCC'] == 1.0:
• return pd.Series([np.nan, np.nan, row['age'], np.nan, np.nan, np.nan, np.nan, np.nan, np.nan])
• elif row['AK'] == 1.0:
• return pd.Series([np.nan, np.nan, np.nan, row['age'], np.nan, np.nan, np.nan, np.nan, np.nan])
• elif row['BKL'] == 1.0:
• return pd.Series([np.nan, np.nan, np.nan, np.nan, row['age'], np.nan, np.nan, np.nan, np.nan])
• elif row['DF'] == 1.0:
• return pd.Series([np.nan, np.nan, np.nan, np.nan, np.nan, row['age'], np.nan, np.nan, np.nan])
• elif row['VASC'] == 1.0:
• return pd.Series([np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, row['age'], np.nan, np.nan])
• elif row['SCC'] == 1.0:
• return pd.Series([np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, row['age'], np.nan])
• age_df = age_df.apply(lambda row: replace_age(row), axis=1).rename(columns={0:'MEL', 1:'NV', 2:'BCC', 3:'AK', 4:'BKL', 5:'DF', 6:'VASC', 7:'SCC', 8:'age'}).drop(['age'], axis=1)
• age_df.boxplot()
• plt.savefig(osp.join(experiment_dir,'age_dist.jpg'))
• age_quantile = pd.DataFrame([age_df.quantile(0.25), age_df.quantile(0.5), age_df.quantile(0.75)], index=['lower quartile', 'median', 'upper quartile'])
• age_quantile.to_csv(osp.join(experiment_dir, 'age.csv'), index=True)
• sex_df = data_df[['MEL','NV','BCC','AK','BKL','DF','VASC','SCC','gender']]
• sex_sum_df = sex_df.groupby('gender').sum()
• sex_sum_df.T.plot(kind='bar')
• plt.savefig(osp.join(experiment_dir,'gender_mal.jpg'))
• sex_sum_df.to_csv(osp.join(experiment_dir, 'gender.csv'), index=True)
• anotom_df = data_df[['MEL','NV','BCC','AK','BKL','DF','VASC','SCC','bodySite']]
• anotom_sum_df = anotom_df.groupby('bodySite').sum()
• anotom_sum_df.T.plot(kind='bar')
• plt.savefig(osp.join(experiment_dir,'anotom_mal.jpg'))
• anotom_sum_df.to_csv(osp.join(experiment_dir, 'anotom.csv'), index=True)

Graph