def slide_info(display_all_properties=False):
"""
Display information (such as properties) about training images.
Args:
display_all_properties: If True, display all available slide properties.
"""
t = Time()
num_train_images = get_num_training_slides()
obj_pow_20_list = []
obj_pow_40_list = []
obj_pow_other_list = []
for slide_num in range(1, num_train_images + 1):
slide_filepath = get_training_slide_path(slide_num)
print("\nOpening Slide #%d: %s" % (slide_num, slide_filepath))
slide = open_slide(slide_filepath)
print("Level count: %d" % slide.level_count)
print("Level dimensions: " + str(slide.level_dimensions))
print("Level downsamples: " + str(slide.level_downsamples))
print("Dimensions: " + str(slide.dimensions))
objective_power = int(slide.properties[openslide.PROPERTY_NAME_OBJECTIVE_POWER])
print("Objective power: " + str(objective_power))
if objective_power == 20:
obj_pow_20_list.append(slide_num)
elif objective_power == 40:
obj_pow_40_list.append(slide_num)
else:
obj_pow_other_list.append(slide_num)
print("Associated images:")
for ai_key in slide.associated_images.keys():
print(" " + str(ai_key) + ": " + str(slide.associated_images.get(ai_key)))
print("Format: " + str(slide.detect_format(slide_filepath)))
if display_all_properties:
print("Properties:")
for prop_key in slide.properties.keys():
print(" Property: " + str(prop_key) + ", value: " + str(slide.properties.get(prop_key)))
print("\n\nSlide Magnifications:")
print(" 20x Slides: " + str(obj_pow_20_list))
print(" 40x Slides: " + str(obj_pow_40_list))
print(" ??x Slides: " + str(obj_pow_other_list) + "\n")
t.elapsed_display()
# if __name__ == "__main__":
# show_slide(2)
# slide_info(display_all_properties=True)
# slide_stats()
# training_slide_to_image(4)
# img_path = get_training_image_path(4)
# img = open_image(img_path)
# img.show()
# slide_to_scaled_pil_image(5)[0].show()
# singleprocess_training_slides_to_images()
# multiprocess_training_slides_to_images()
def singleprocess_training_slides_to_images():
"""
Convert all WSI training slides to smaller images using a single process.
"""
t = Time()
num_train_images = get_num_training_slides()
training_slide_range_to_images(1, num_train_images)
t.elapsed_display()
def slide_info(datapath, display_all_properties=False): ## ap
"""
Display information (such as properties) about training images.
Args:
display_all_properties: If True, display all available slide properties.
"""
t = Time()
num_train_images = get_num_training_slides()
obj_pow_20_list = []
obj_pow_40_list = []
obj_pow_other_list = []
train_images = glob.glob(os.path.join(datapath, '*.svs')) ## ap
## for slide_num in range(1, num_train_images + 1):
for slide_filepath in train_images: ## ap
slide_num = int(
os.path.basename(slide_filepath).split('.svs')[0]) ## ap
## slide_filepath = get_training_slide_path(slide_num) ## ap
print("\nOpening Slide #%d: %s" % (slide_num, slide_filepath))
slide = open_slide(slide_filepath)
print("Level count: %d" % slide.level_count)
print("Level dimensions: " + str(slide.level_dimensions))
print("Level downsamples: " + str(slide.level_downsamples))
print("Dimensions: " + str(slide.dimensions))
objective_power = int(
slide.properties[openslide.PROPERTY_NAME_OBJECTIVE_POWER])
print("Objective power: " + str(objective_power))
if objective_power == 20:
obj_pow_20_list.append(slide_num)
elif objective_power == 40:
obj_pow_40_list.append(slide_num)
else:
obj_pow_other_list.append(slide_num)
print("Associated images:")
for ai_key in slide.associated_images.keys():
print(" " + str(ai_key) + ": " +
str(slide.associated_images.get(ai_key)))
print("Format: " + str(slide.detect_format(slide_filepath)))
if display_all_properties:
print("Properties:")
for prop_key in slide.properties.keys():
print(" Property: " + str(prop_key) + ", value: " +
str(slide.properties.get(prop_key)))
print("\n\nSlide Magnifications:")
print(" 20x Slides: " + str(obj_pow_20_list))
print(" 40x Slides: " + str(obj_pow_40_list))
print(" ??x Slides: " + str(obj_pow_other_list) + "\n")
t.elapsed_display()
def multiprocess_training_slides_to_images():
"""
Convert all WSI training slides to smaller images using multiple processes (one process per core).
Each process will process a range of slide numbers.
"""
timer = Time()
# how many processes to use
num_processes = multiprocessing.cpu_count()
num_processes = 12
pool = multiprocessing.Pool(num_processes)
num_train_images = get_num_training_slides()
if num_processes > num_train_images:
num_processes = num_train_images
images_per_process = num_train_images / num_processes
print("Number of processes: " + str(num_processes))
print("Number of training images: " + str(num_train_images))
# each task specifies a range of slides
tasks = []
for num_process in range(1, num_processes + 1):
start_index = (num_process - 1) * images_per_process + 1
end_index = num_process * images_per_process
start_index = int(start_index)
end_index = int(end_index)
tasks.append((start_index, end_index))
if start_index == end_index:
print("Task #" + str(num_process) + ": Process slide " +
str(start_index))
else:
print("Task #" + str(num_process) + ": Process slides " +
str(start_index) + " to " + str(end_index))
# start tasks
results = []
for t in tasks:
results.append(pool.apply_async(training_slide_range_to_images, t))
for result in results:
(start_ind, end_ind) = result.get()
if start_ind == end_ind:
print("Done converting slide %d" % start_ind)
else:
print("Done converting slides %d through %d" %
(start_ind, end_ind))
timer.elapsed_display()
def singleprocess_training_slides_to_images(slides_path,
n_slides: Optional[int] = None
): ## ap
"""
Convert all WSI training slides to smaller images using a single process.
"""
t = Time()
## ap
# num_train_images = get_num_training_slides()
# training_slide_range_to_images(1, num_train_images)
## ap
trainig_slides_to_images(slides_path, n_slides=n_slides)
t.elapsed_display()
def slide_stats():
"""
Display statistics/graphs about training slides.
"""
t = Time()
if not os.path.exists(STATS_DIR):
os.makedirs(STATS_DIR)
num_train_images = get_num_training_slides()
slide_stats = []
for slide_num in range(1, num_train_images + 1):
slide_filepath = get_training_slide_path(slide_num)
print("Opening Slide #%d: %s" % (slide_num, slide_filepath))
slide = open_slide(slide_filepath)
(width, height) = slide.dimensions
print(" Dimensions: {:,d} x {:,d}".format(width, height))
slide_stats.append((width, height))
max_width = 0
max_height = 0
min_width = sys.maxsize
min_height = sys.maxsize
total_width = 0
total_height = 0
total_size = 0
which_max_width = 0
which_max_height = 0
which_min_width = 0
which_min_height = 0
max_size = 0
min_size = sys.maxsize
which_max_size = 0
which_min_size = 0
for z in range(0, num_train_images):
(width, height) = slide_stats[z]
if width > max_width:
max_width = width
which_max_width = z + 1
if width < min_width:
min_width = width
which_min_width = z + 1
if height > max_height:
max_height = height
which_max_height = z + 1
if height < min_height:
min_height = height
which_min_height = z + 1
size = width * height
if size > max_size:
max_size = size
which_max_size = z + 1
if size < min_size:
min_size = size
which_min_size = z + 1
total_width = total_width + width
total_height = total_height + height
total_size = total_size + size
avg_width = total_width / num_train_images
avg_height = total_height / num_train_images
avg_size = total_size / num_train_images
stats_string = ""
stats_string += "%-11s {:14,d} pixels (slide #%d)".format(max_width) % (
"Max width:", which_max_width)
stats_string += "\n%-11s {:14,d} pixels (slide #%d)".format(max_height) % (
"Max height:", which_max_height)
stats_string += "\n%-11s {:14,d} pixels (slide #%d)".format(max_size) % (
"Max size:", which_max_size)
stats_string += "\n%-11s {:14,d} pixels (slide #%d)".format(min_width) % (
"Min width:", which_min_width)
stats_string += "\n%-11s {:14,d} pixels (slide #%d)".format(min_height) % (
"Min height:", which_min_height)
stats_string += "\n%-11s {:14,d} pixels (slide #%d)".format(min_size) % (
"Min size:", which_min_size)
stats_string += "\n%-11s {:14,d} pixels".format(
round(avg_width)) % "Avg width:"
stats_string += "\n%-11s {:14,d} pixels".format(
round(avg_height)) % "Avg height:"
stats_string += "\n%-11s {:14,d} pixels".format(
round(avg_size)) % "Avg size:"
stats_string += "\n"
print(stats_string)
stats_string += "\nslide number,width,height"
for i in range(0, len(slide_stats)):
(width, height) = slide_stats[i]
stats_string += "\n%d,%d,%d" % (i + 1, width, height)
stats_string += "\n"
stats_file = open(os.path.join(STATS_DIR, "stats.txt"), "w")
stats_file.write(stats_string)
stats_file.close()
t.elapsed_display()
x, y = zip(*slide_stats)
colors = np.random.rand(num_train_images)
sizes = [10 for n in range(num_train_images)]
plt.scatter(x, y, s=sizes, c=colors, alpha=0.7)
plt.xlabel("width (pixels)")
plt.ylabel("height (pixels)")
plt.title("SVS Image Sizes")
plt.set_cmap("prism")
plt.tight_layout()
plt.savefig(os.path.join(STATS_DIR, "svs-image-sizes.png"))
plt.show()
plt.clf()
plt.scatter(x, y, s=sizes, c=colors, alpha=0.7)
plt.xlabel("width (pixels)")
plt.ylabel("height (pixels)")
plt.title("SVS Image Sizes (Labeled with slide numbers)")
plt.set_cmap("prism")
for i in range(num_train_images):
snum = i + 1
plt.annotate(str(snum), (x[i], y[i]))
plt.tight_layout()
plt.savefig(os.path.join(STATS_DIR, "svs-image-sizes-slide-numbers.png"))
plt.show()
plt.clf()
area = [w * h / 1000000 for (w, h) in slide_stats]
plt.hist(area, bins=64)
plt.xlabel("width x height (M of pixels)")
plt.ylabel("# images")
plt.title("Distribution of image sizes in millions of pixels")
plt.tight_layout()
plt.savefig(os.path.join(STATS_DIR, "distribution-of-svs-image-sizes.png"))
plt.show()
plt.clf()
whratio = [w / h for (w, h) in slide_stats]
plt.hist(whratio, bins=64)
plt.xlabel("width to height ratio")
plt.ylabel("# images")
plt.title("Image shapes (width to height)")
plt.tight_layout()
plt.savefig(os.path.join(STATS_DIR, "w-to-h.png"))
plt.show()
plt.clf()
hwratio = [h / w for (w, h) in slide_stats]
plt.hist(hwratio, bins=64)
plt.xlabel("height to width ratio")
plt.ylabel("# images")
plt.title("Image shapes (height to width)")
plt.tight_layout()
plt.savefig(os.path.join(STATS_DIR, "h-to-w.png"))
plt.show()