-
Notifications
You must be signed in to change notification settings - Fork 2
/
e03_4_qinProcessNLSQ.jl
119 lines (98 loc) · 3.25 KB
/
e03_4_qinProcessNLSQ.jl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
# Process QIN Breast Data using NRRM and CNRRM
# Run this script by launching julia then entering: include("path/to/e03_4_qinProcessNLSQ.jl")
# Estimated runtime: ~500 seconds, on 4 cores
# Choice of parameters for the reference tissue. Should match e03_05
# It's best to leave this as it is
refName = "refY"
# First, figure out the current file path
curFilePath = Base.source_path() # Path to current file
workDir = dirname(curFilePath) # Our working directory (directory of current file)
## If the above doesn't work, then user should enter the location of the CLRRM directory
## by manually modifying this script
# Load pre-requisite packages
using DCEMRI
using MAT
# Get the directories and files of interest
matDir = workDir * "/data/QINBreast/ForJulia"
warmupFile = workDir * "/data/QIBA-ToftsV6/QIBAv6-Mini4Jl-warmup.mat"
auxCode = workDir * "/jlfiles/auxCode.jl"
refRegCode = workDir * "/jlfiles/refRegionFunction.jl"
outDir = workDir * "/data/QINBreast/FromJulia"
if !(isdir(outDir))
mkdir(outDir)
end
############### Warmup run
results = fitdata(datafile=warmupFile, models=[2])
rm("output.mat") # Delete output since it isn't needed
##############
# This will load the functions for fitting the models
include(auxCode)
include(refRegCode)
# Define the reference tissue parameters
refKTrans = 0.1;
refVe = 0.1;
if (refName == "refWS")
refKTrans = 0.07;
refVe = 0.14;
elseif (refName == "refP")
refKTrans = 0.137;
refVe = 0.115;
end
refKep = refKTrans/refVe
############## Main run
matFiles = readdir(matDir)
numFiles = length(matFiles)
# Loop through the .mat files
for q=1:numFiles
curName = matFiles[q]
curFile = matDir * "/$curName"
# Load data
matData = matread(curFile)
Ct = matData["Ct"]
(nT, nV) = size(Ct)
Ct = reshape(Ct, nT, nV, 1)
Cp = vec(matData["Cp"])
t = vec(matData["t"])
tResFactor = round(Int64,matData["tResFactor"])
(nT, nV) = size(Ct)
mask = ones(nV,1) .> 0
# Simulate reference tissue curve
Crr = DCEMRI.toftskety(t, [refKTrans, refKep], Cp)
Ct = reshape(Ct,nT,nV)
# Downsample so that Crr, Cp, and t have same temporal resolution as Ct
Crr = downsample(Crr,tResFactor)
Cp = downsample(Cp,tResFactor)
t = downsample(t,tResFactor)
# Non-Linear Tofts Model
pkParamsT, residT = DCEMRI.fitdce(Ct, mask, t, Cp, models=[2])
# Re-arrange the output from Tofts Model fit
kt = pkParamsT[1,:,1]
ve = pkParamsT[2,:,1]
kep = kt./ve
# Parameter mask using threshold on fitted Tofts Model parameters
pMask = (kt.>0.001) & (kt.<1) & (ve.>0.001) & (ve.<1) & (kep.<35)
kt = kt[pMask]
ve = ve[pMask]
kep = kep[pMask]
# NRRM and CNRRM
(pkParamsCN, residCN, estKepRR, pkParamsN, residN, runtimeN, runtimeCN)=fitCNRRM(Ct, Crr, t, estKepRR=0.0, doTime=true)
# Output results
results = Dict()
results["pkParamsN"] = pkParamsN
results["residN"] = residN
results["pkParamsCN"] = pkParamsCN
results["residCN"] = residCN
results["runtimeN"] = runtimeN
results["runtimeC"] = runtimeCN
results["estKepRRN"] = estKepRR
results["pkParamsT"] = pkParamsT
results["residT"] = residT
results["kt"] = kt
results["ve"] = ve
results["kep"] = kep
results["Crr"] = Crr
results["Cp"] = Cp
results["t"] = t
results["pMask"] = convert(Array{Bool,1}, pMask)
matwrite(outDir * "/$curName", results)
end