# Copyright (c) Thomas Else 2023-25.
# License: MIT
from collections import defaultdict
from typing import TYPE_CHECKING, Tuple
import numpy as np
if TYPE_CHECKING:
from ..io.msot_data import PAData
from .rois.roi_type import ROI
import matplotlib
[docs]
def split_roi_left_right(data: "PAData", base_roi="", split_template="unnamed"):
from .rois.roi_type import ROI
# Initial implementation - only supports one "dividing" roi.
data_rois = data.get_rois()
template_data = None
names = None
if (split_template + "_left", "0") in data_rois:
template_data = data_rois[(split_template + "_left", "0")]
names = ["left", "right"]
elif (split_template + "_right", "0") in data_rois:
template_data = data_rois[(split_template + "_right", "0")]
names = ["right", "left"]
else:
return []
template_polygon = template_data.get_polygon()
output = []
for (name, number), roi in data_rois.items():
if roi.position != "" or roi.generated:
continue
if base_roi in name:
# Process
base_polygon = roi.get_polygon()
base_region_a = base_polygon & template_polygon
base_region_b = base_polygon - base_region_a
# Convert to numpy array
try:
x, y = base_region_a.exterior.coords.xy
region_a = np.array([x[:-1], y[:-1]]).T
x, y = base_region_b.exterior.coords.xy
region_b = np.array([x[:-1], y[:-1]]).T
except AttributeError:
continue
# Get the metadata.
z_position = roi.z
run = roi.run
repetition = roi.repetition
roi_class = roi.roi_class
roi_a = ROI(region_a, z_position, run, repetition, roi_class, names[0])
roi_b = ROI(region_b, z_position, run, repetition, roi_class, names[1])
output.append(roi_a)
output.append(roi_b)
return output
ROI_NAMES = [
"brain",
"body",
"reference",
"aorta",
"tumour",
"background",
"artery",
"vein",
"muscle",
"phantom",
"unnamed",
"kidney",
"spleen",
"spine",
"skin",
"fat",
]
REGION_COLOURS = matplotlib.cm.hsv(np.linspace(0, 1, len(ROI_NAMES)))
REGION_COLOUR_MAP = defaultdict(lambda: REGION_COLOURS[-1])
for x, y in zip(ROI_NAMES, REGION_COLOURS):
REGION_COLOUR_MAP[x] = y
[docs]
def close_loop(x):
return np.concatenate([x, x[0:1, :]])
[docs]
def get_rim_core_rois(roi_data: "ROI", distance: float, radius=False):
from ..io.msot_data import ROI
roi = roi_data.get_polygon()
if radius:
effective_radius = np.sqrt(roi.area / np.pi)
# distance is the desired radius
distance = -distance + effective_radius
roi_buffer = roi.buffer(-distance)
core_roi = ROI(
roi_buffer,
roi_data.z,
roi_data.run,
roi_data.repetition,
roi_data.roi_class + ".core",
roi_data.position,
generated=True,
ax0_index=roi_data.ax0_index,
)
rim_roi = ROI(
roi - roi_buffer,
roi_data.z,
roi_data.run,
roi_data.repetition,
roi_data.roi_class + ".rim",
roi_data.position,
generated=True,
ax0_index=roi_data.ax0_index,
)
return core_roi, rim_roi
[docs]
def add_rim_core_data(
data: "PAData", base_roi: Tuple[str, str], distance: float, radius=False
):
roi_data: "ROI" = data.get_rois()[base_roi]
return get_rim_core_rois(roi_data, distance, radius)
