Detect fluid_shipment_anomalies

fluid_shipment_anomalies.py

import pandas as pd

# cd /data/predict1/to_nda/nda-submissions
df=pd.read_csv('blood_saliva_rack_Pronet.csv')

index=[]
for i,row in df.iterrows():


    rc=row['Rack Code']
    pr=row['Position on Rack']
    ic=row['Inventory Code']

    flag=False

    # Rack Code must contain 2 numbers
    if pd.isna(rc) or rc in [' ',-3,-9] or \
        sum(c.isdigit() for c in rc)<2:
        flag=True

    elif pd.isna(pr) or len(pr)>3 or len(pr)<2:
        flag=True

    # length of Inventory Code is 10-12 characters
    elif pd.isna(ic) or len(ic)<10:
        flag=True

    if flag:
        index.append(i)


df1=df.loc[index]

df1.to_csv('anomalous_fluid_shipment.csv',index=False)



#######################################################

# blood rack miscount detection
rack_code='9500011256'
files=glob('PronetYA/raw/*/surveys/*.Pronet.json')
subjects=[]
for file in files:
    with open(file) as f:
        data=json.load(f)

    # print(file)
    for d in data:
        if 'baseline_arm_' in d['redcap_event_name'] or 'month_2_arm_' in d['redcap_event_name']:
            if d['chrblood_rack_barcode'].strip()==rack_code:
                subjects.append(d['chric_record_id'])
                print(d['chric_record_id'],d['chrblood_drawdate'])



# saliva rack miscount detection
rack_code='ProNET-0812'
files=glob('PronetYA/raw/*/surveys/*.Pronet.json')
subjects=[]
for file in files:
    with open(file) as f:
        data=json.load(f)

    # print(file)
    for d in data:
        if 'baseline_arm_' in d['redcap_event_name'] or 'month_2_arm_' in d['redcap_event_name']:
            # if d['chrsaliva_box1b'].strip()==rack_code:
            # if d['chrsaliva_box2b'].strip()==rack_code:
            if d['chrsaliva_box3b'].strip()==rack_code:
                subjects.append(d['chric_record_id'])
                print(d['chric_record_id'],d['chrsaliva_coldate'])