orthancInterface/pythonScripts/scanOrthancQuick.py

#labkey/orthanc interface

#scans the orthanc internal database and fills labkey table

#"Orthanc" section expected in the configuration file, with
#"queryName": query to be filled for each image 
#"demographicsQuery": query to be filled for every patient
#"schemaName": the schema both queries are part of
#"project": name of the project under labkey that collects Orthanc data,
#           typically named Orthanc or similar
#"participantField": sorting participant field label in labkey study


import os
import json
import re
import sys

def main(parameterFile):

    fhome=os.path.expanduser('~')
    fsetup=os.path.join(fhome,'.labkey','setup.json')

    with open(fsetup,'r') as f:
        setup=json.load(f)

    sys.path.insert(0,setup['paths']['nixWrapper'])
    import nixWrapper

    nixWrapper.loadLibrary("labkeyInterface")
    import labkeyInterface
    import labkeyDatabaseBrowser

    nixWrapper.loadLibrary('orthancInterface')
    import orthancInterface
    import orthancDatabaseBrowser

    fconfig=os.path.join(fhome,'.labkey','network.json')

    net=labkeyInterface.labkeyInterface()
    net.init(fconfig)
    db=labkeyDatabaseBrowser.labkeyDB(net)


    onet=orthancInterface.orthancInterface()
    onet.init(fconfig)
    odb=orthancDatabaseBrowser.orthancDB(onet)


    with open(parameterFile) as f:
        pars=json.load(f)

    opars=pars['Orthanc']
    project=opars['project']
    
    participantField=opars['participantField']

    patients=odb.getPatients()


    #equivalent for labkey side?
    dsDemo=db.selectRows(project,opars['schemaName'],\
            opars['demographicsQuery'],[])
    dsPatients=[row['OrthancId'] for row in dsDemo['rows']]
    pMissing=[p for p in patients if p not in dsPatients]
    print('Missing : {}'.format(len(pMissing)))

    #we need details for all patients (some might have just a study uploaded
    pdata={p:odb.getPatientData(p) for p in patients}
    
    #update patient data for missing patients
    prows=[]
    for p in pMissing:
        dicom=pdata[p]['MainDicomTags']
        row={}
        row[participantField]=dicom['PatientID']
        row['birthDate']=dicom['PatientBirthDate']
        row['PatientName']=dicom['PatientName']
        row['OrthancId']=p
        prows.append(row)

    db.modifyRows('insert',project,opars['schemaName'],\
                    opars['demographicsQuery'],prows)

    n=len(patients)
    i=0
    #download the full images dataset (~1000 rows)
    ds=db.selectRows(project,opars['schemaName'],opars['queryName'],[])
    for p in patients:
        dicom=pdata[p]['MainDicomTags']
        patientId=dicom['PatientID']

        #get all studies for pateint
        #qfilter={'variable':participantField,'value':patientId,'oper':'eq'}
        dsStudies=[row['orthancStudy'] for row in ds['rows'] if row[participantField]==patientId]
        #number of entries for patient (if new must be inserted)
        seqNum=len(dsStudies)
        
        #unique
        dsStudies=list(set(dsStudies))
        studies=pdata[p]['Studies']
        pHex=p.split('-')

        print("[{:>3d}/{:>3d}] ID: {:>10s} [{}] Studies: {:>2}/{:>2} Entries in DB: {:>4d}".format(i,n,patientId,pHex[-1],len(dsStudies),len(studies),seqNum))
        missingStudies=[s for s in studies if s not in dsStudies]
        #print('Missing studies: {}'.format(len(missingStudies)))

        srows=[]
        for s in missingStudies:

            sdata=odb.getStudyData(s)
            sdicom=sdata['MainDicomTags']
            sid=sdicom['StudyInstanceUID']
            #print('Data: {}'.format(sdata))
            #this is try/except protetcted in previous version...
            sdate="19700101"
            try:
                sdate=sdicom['StudyDate']
            except KeyError:
                pass
            #continue
        
            print('\tStudy[{}]: {}/{}'.format(sdate,s,sid))
        
            for se in sdata['Series']:

                sedata=odb.getSeriesData(se)
                sedicom=sedata['MainDicomTags']
                    
                seid=sedicom['SeriesInstanceUID']
                #print('Data: {}'.format(sedata))
                seDesc="NONE"
                try:
                    seDesc=sedicom['SeriesDescription']
                except KeyError:
                    pass

                #replace letters that might trip the database
                spanishOAcute=''.join([chr(3619),chr(3603)])
                spanishAAcute=''.join([chr(3619),chr(3585)])
                seDesc=re.sub(spanishOAcute,'o',seDesc)
                seDesc=re.sub(spanishAAcute,'a',seDesc)


                print('\t\tSeries[{}]: {}/{}'.format(seDesc,se,seid))

                #qfilter={'variable':'orthancSeries','value':se,'oper':'eq'}
                #ds=db.selectRows(project,opars['schemaName'],\
                #        opars['queryName'],[qfilter])
                
                #count existing entries for patient
                #qfilter={'variable':participantField,'value':patientId,'oper':'eq'}
                #ds=db.selectRows(project,opars['schemaName'],\
                #opars['queryName'],[qfilter])
                #    seqNum=len(ds['rows'])
                    
                #create new row
                row={}
                row[participantField]=patientId
                row['sequenceNum']=seqNum
                row['orthancSeries']=se
                
                
                row['dicomStudy']=sid
                row['orthancStudy']=s
                row['dicomSeries']=seid
                row['studyDate']=sdate
                row['seriesDescription']=seDesc
                srows.append(row)
                seqNum+=1

        #upload updates
        if len(srows)>0:
            db.modifyRows('insert',project,opars['schemaName'],\
                opars['queryName'],srows)


        i+=1
    print("Done")

if __name__=="__main__":
    main(sys.argv[1])