JCAT-19-306-SDCT-UNET-test.txt

import os
import random
import matplotlib.pyplot as plt
import pylab
import numpy as np

import keras
from keras.models import Model
from keras import models
from keras.layers import *
from keras.optimizers import Adam
from keras.optimizers import RMSprop
from keras.optimizers import SGD
from keras.regularizers import l2
from keras.preprocessing.image import ImageDataGenerator
import keras.backend as K
from keras.callbacks import TensorBoard
from keras.callbacks import LearningRateScheduler, ModelCheckpoint
SEED = 145

import pickle

cvDict = pickle.load(open("excldCV5-data/cvDict.p","rb"))

listOfPtIds = cvDict['cv5']

listOfAllImages = os.listdir('pyOsiriXimageExports/compton')

model = models.load_model('190204_excldCV5-MADplot-800Epochs.h')

comptonDir = 'pyOsiriXimageExports/compton'
peDir = 'pyOsiriXimageExports/pe'
maskDir = 'pyOsiriXimageExports/mask'

listOfDSCs = []

for eachPt in listOfPtIds:
    thisPtImages = []
    for eachImage in listOfAllImages:
        if eachImage[:3] == eachPt:
            thisPtImages.append(eachImage)

    print(len(thisPtImages))

    sumMaskPixels = 0.0
    sumPredictedMaskPixels = 0.0
    sumOverlap = 0.0

    for eachImage in thisPtImages:

        maskPath = maskDir + '/' + eachImage
        mask = np.load(maskPath)

        comptonPath = comptonDir + '/' + eachImage
        compton = np.load(comptonPath)

        pePath = peDir + '/' + eachImage
        pe = np.load(pePath)
        image = np.stack((compton,pe))
        image = np.moveaxis(image, 0, -1)

        image = image[np.newaxis,:,:,:]
        yhat = model.predict(image)

        predictedMask = yhat[0,:,:,0]>0.5

        overlap = np.logical_and(mask, predictedMask)
        sumOverlap = sumOverlap+np.sum(overlap)
        sumMaskPixels = sumMaskPixels + np.sum(mask)
        sumPredictedMaskPixels = sumPredictedMaskPixels + np.sum(predictedMask)

    print(sumOverlap)
    print(sumMaskPixels)
    print(sumPredictedMaskPixels)

    DSC = float(2*sumOverlap)/float(sumMaskPixels+sumPredictedMaskPixels)
    listOfDSCs.append(DSC)

print(listOfDSCs)

sum = 0
for eachItem in listOfDSCs:
    sum+=eachItem

print(sum/6.0)