基于自适应分区演化水平集的肿瘤分割算法，你能告诉我代码流程吗

基于自适应分区演化水平集的肿瘤分割算法涉及多个步骤，以下是一个可能的代码流程：

1. 导入所需的Python库和模块：

import numpy as np
import cv2
from scipy.ndimage import distance_transform_edt

2. 定义一个函数来加载并预处理肿瘤图像：

def load_image(image_path):
    img = cv2.imread(image_path, 0)
    img = cv2.medianBlur(img, 5)
    img = cv2.equalizeHist(img)
    img = cv2.normalize(img, None, 0, 255, cv2.NORM_MINMAX)
    return img

3. 定义一个函数来初始化演化水平集变量，并计算图像的梯度值：

def init_level_set(img_shape, radius):
    level_set = np.zeros(img_shape)
    center = (img_shape[0] // 2, img_shape[1] // 2)
    y, x = np.ogrid[:img_shape[0], :img_shape[1]]
    dist_from_center = np.sqrt((x - center[0])**2 + (y - center[1])**2)
    level_set[dist_from_center <= radius] = 1
    level_set[dist_from_center > radius] = -1
    return level_set

def get_gradient(img):
    grad_x = cv2.Sobel(img, cv2.CV_32F, 1, 0, ksize=3)
    grad_y = cv2.Sobel(img, cv2.CV_32F, 0, 1, ksize=3)
    gradient = np.sqrt(grad_x**2 + grad_y**2)
    return gradient

4. 定义一个函数来计算自适应分区：

def get_partition(level_set, num_partitions):
    num_pixels = level_set.size
    pixels_per_partition = num_pixels // num_partitions
    sorted_levels = np.sort(level_set.flatten())
    partitions = []
    start_index = 0
    for i in range(num_partitions - 1):
        end_index = start_index
        while end_index < num_pixels and end_index - start_index < pixels_per_partition:
            end_index += 1
        threshold = sorted_levels[end_index]
        partition = np.zeros(level_set.shape)
        partition[level_set >= threshold] = 1
        partitions.append(partition)
        start_index = end_index
    partition = np.zeros(level_set.shape)
    partition[level_set >= sorted_levels[start_index]] = 1
    partitions.append(partition)
    return partitions

5. 定义一个函数来计算演化速度：

def get_speed(img_grad, beta, c_0, c_1, level_set):
    inside = level_set > 0
    outside = level_set < 0
    c = np.zeros(img_grad.shape)
    c[inside] = c_0
    c[outside] = c_1
    speed = c * img_grad
    dirac = 1 / (2 * beta)
    dirac_level_set = np.zeros(level_set.shape)
    dirac_level_set[level_set >= 0] = 1
    speed = dirac * dirac_level_set * speed
    return speed

6. 定义一个函数来更新演化水平集：

def