import ocsv
import sys
# argv[1] - input file
# argv[2] - name of the column to be flipped
# argv[3] - output file
fin = open("C:\\Users\\AadarshSam\\Desktop\\CAPSTONE PROJECT_SCRIPTS\\PID.csv", 'r')
col = ocsv.getColumns(fin.readline())
flipCol = col['daystonext']
flipCol2=col['agyradm']
flipcol3=col['Gender']
currentpid = ''
# list of list of list
seqs = []
def func(line):
global currentpid, seqs
row = line.strip().split(',')
#print row[flipCol2]
pid = row[col['PID']]
if pid == currentpid:
seqs[-1][0].append(row[flipCol])
seqs[-1][1].append(row[col['cost\n']])
else:
seqs.append([[row[flipCol]], [row[col['cost\n']]]])
currentpid = pid
ocsv.runFunc(fin, func)
newline = line.strip()
row = newline.split(',')
if skipLast and row[col['nextLOS']] == '': return
newline += ',' + bucketLOS(row[col['LOS']])
newline += ',' + bucketLOS(row[col['nextLOS']])
newline += ',' + bucketCost(row[col['cost']])
newline += ',' + bucketCost(row[col['nextCost']])
newline += ',' + dxmap[row[col['diag_p']]]
newline += ',' + prmap[row[col['proc_p']]]
fout.write(newline + '\n')
skipLast = sys.argv[3] == 'True'
fin = open(sys.argv[1], 'r')
fout = open(sys.argv[2], 'w')
chf = 'chf' in sys.argv[1].lower()
colline = fin.readline().strip()
col = ocsv.getColumns(colline)
dxmap = parseICD9Mapping('AppendixASingleDX.txt')
prmap = parseICD9Mapping('AppendixBSinglePR.txt')
# add new columns
newcols = ['LOS_b', 'nextLOS_b', 'cost_b', 'nextCost_b', 'diag_p_ccs', 'proc_p_ccs']
for newcol in newcols:
colline += ',' + newcol
col[newcol] = len(col)
fout.write(colline + '\n')
ocsv.runFunc(fin, convert)
fin.close()
fout.close()
import ocsv
import sys
import random
# argv[1] input file
# argv[2] first output file
# argv[3] second output file
f1 = open(sys.argv[2], 'w')
f2 = open(sys.argv[3], 'w')
fin = open(sys.argv[1], 'r')
line = fin.readline()
col = ocsv.getColumns(line)
f1.write(line)
f2.write(line)
pid1, pid2 = '', ''
def func(line):
global pid1, pid2
row = line.strip().split(',')
pid = row[col['PID']]
if pid != pid1 and pid != pid2:
if int(row[col['gender']]) == 0: # 0 is female
pid1 = pid
else:
pid2 = pid
if pid == pid1:
f1.write(line)
else:
f2.write(line)
import ocsv
import sys
if len(sys.argv) > 2:
path = sys.argv[1] # input path
out = sys.argv[2] # output path
fin = open(path, 'r')
fout = open(out, 'w')
line = fin.readline()
fout.write(line)
col = ocsv.getColumns(line.strip())
samePtnLines = []
pid = ''
skipPtn = False
def func(line):
global samePtnLines, pid, skipPtn
row = line.strip().split(',')
rowpid = row[col['PID']]
if pid != '' and pid != rowpid:
if not skipPtn:
for samePLine in samePtnLines:
fout.write(samePLine)
skipPtn = False
samePtnLines = []
pid = rowpid
if skipPtn:
return
if float(row[col['cost']]) <= 1:
truepos = truepos + 1
else:
falsepos = falsepos + 1
elif row[col['thirtyday']] == '0':
trueneg = trueneg + 1
else:
falseneg = falseneg + 1
seqRange = range(int(sys.argv[5]), int(sys.argv[6]) + 1) if len(sys.argv) > 4 else [0]
fout = open(sys.argv[4], 'w')
msg = 'SeqLength,T+,T-,F+,F-,rowcount,accuracy,precision,baseline'
fout.write(msg + '\n')
print(msg)
for seqLength in seqRange:
ftest = open(sys.argv[1], 'r')
col = ocsv.getColumns(ftest.readline())
currentPID = ''
currentSeq = []
truepos = 0
trueneg = 0
falsepos = 0
falseneg = 0
rowcount = 0
ocsv.runFunc(ftest, nb)
ftest.close()
result = [seqLength, truepos, trueneg, falsepos, falseneg, rowcount, (truepos + trueneg) / rowcount, truepos / (truepos + falsepos), (trueneg + falsepos) / rowcount]
msg = ','.join([str(item) for item in result])
fout.write(msg + '\n')
print(msg)
fout.close()
import ocsv
import sys
fin = open(sys.argv[1], 'r')
fout = open(sys.argv[2], 'w')
CCS = sys.argv[3]
pids = set()
def saveCohortPID(line):
row = line.strip('\n').split(',')
if row[col['DXCCS_' + CCS]] == '1':
pids.add(row[col['PID']])
def outputCohort(line):
row = line.strip('\n').split(',')
if row[col['PID']] in pids:
fout.write(line)
line = fin.readline()
fout.write(line)
col = ocsv.getColumns(line.strip('\n'))
print('Finding all PID in this cohort')
ocsv.runFunc(fin, saveCohortPID)
print('There are totally ' + str(len(pids)) + ' PIDs in this cohort')
fin.close()
fin = open(sys.argv[1], 'r')
line = fin.readline()
print('Writing cohort to output')
ocsv.runFunc(fin, outputCohort)
