model.py

import os
import nibabel as nib
import numpy as np
import matplotlib.pyplot as plt
import cv2
import math
from tqdm import tqdm
from tensorflow.python.framework import ops


IMG_PX_SIZE = 80

cwd = os.getcwd()
data_dir = '\\examples'
dataset  = os.listdir(cwd+data_dir)

for data in dataset:
    ops.reset_default_graph()
    def chunks(l, n):
        for i in range(0, len(l), n):
            yield l[i:i + n]

    def mean(l):
        return sum(l) / len(l)

    def process_data(data):

        path = cwd+'/'+ data_dir +'/'+ str(data)
        img3d = nib.load(path)
        a = np.array(img3d.get_data())

        slice_img=[]
        if(img3d.header['dim'][4]==1):
            b = cv2.resize(np.array(a[:,:,100]),(200,200))
            for i in range(a.shape[2]):
                slice_img.append(cv2.resize(np.array(a[:,:,i]),(IMG_PX_SIZE,IMG_PX_SIZE)))
        else:
            b = cv2.resize(np.array(a[:,:,12,1]),(200,200))
            for i in range(a.shape[2]):
                slice_img.append(cv2.resize(np.array(a[:,:,i,1]),(IMG_PX_SIZE,IMG_PX_SIZE)))


        new_slices = []

        chunk_sizes = math.ceil(len(slice_img) / 16)
        for slice_chunk in chunks(slice_img, chunk_sizes):
            slice_chunk = list(map(mean, zip(*slice_chunk)))
            new_slices.append(slice_chunk)

        return np.array(new_slices),b

    img_data,imagetoshow = process_data(data)

    import tflearn
    LR = 0.001

    from tflearn.layers.conv import conv_2d, max_pool_2d
    from tflearn.layers.core import input_data, dropout, fully_connected
    from tflearn.layers.estimator import regression

    convnet = input_data(shape=[None,IMG_PX_SIZE,IMG_PX_SIZE,1], name='input')

    convnet = conv_2d(convnet, 32, 2, activation='relu')
    convnet = max_pool_2d(convnet,2)

    convnet = conv_2d(convnet, 64, 2, activation='relu')
    convnet = max_pool_2d(convnet,2)

    convnet = conv_2d(convnet, 32, 2, activation='relu')
    convnet = max_pool_2d(convnet,2)

    convnet = conv_2d(convnet, 64, 2, activation='relu')
    convnet = max_pool_2d(convnet,2)

    convnet = fully_connected(convnet, 1024, activation='relu')
    convnet = dropout(convnet, 0.8)

    convnet = fully_connected(convnet, 2, activation='softmax')
    #convnet = regression(convnet, optimizer='adam', learning_rate=LR,loss='categorical_crossentropy', name='targets')

    model = tflearn.DNN(convnet)

    model.load('model.tflearn')
    model_out=model.predict(img_data[8].reshape(-1,IMG_PX_SIZE,IMG_PX_SIZE,1))
    if np.argmax(model_out[0]) == 0:
        print('BOLD MRI')
    else:
        print('T1w MRI')

    import matplotlib.pyplot as plt

    plt.imshow(imagetoshow)
    plt.show()