Skip to content

Commit

Permalink
_estimate_min_bead distance returns just the number of pixels of the …
Browse files Browse the repository at this point in the history 
…input.

blacked
  • Loading branch information
@juliomateoslangerak
juliomateoslangerak committed Feb 9, 2024
1 parent 763f4a8 commit 9c68803
Showing 1 changed file with 34 additions and 24 deletions.
58 changes: 34 additions & 24 deletions src/microscopemetrics/samples/psf_beads.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -59,9 +59,10 @@ def _calculate_bead_intensity_outliers(
bead_considered_intensity_outlier = {}

bead_max_intensities = [
bead.max() for im_bead_crops in bead_crops.values()
for ch_bead_crops in im_bead_crops for
bead in ch_bead_crops
bead.max()
for im_bead_crops in bead_crops.values()
for ch_bead_crops in im_bead_crops
for bead in ch_bead_crops
]

mean = np.mean(bead_max_intensities)
Expand Down Expand Up @@ -104,7 +105,9 @@ def _generate_key_values(
sum(len(ch) for ch in im) for im in discarded_positions_self_proximity.values()
),
"nr_of_beads_discarded_axial_edge": bead_properties_df["considered_axial_edge"].sum(),
"nr_of_beads_considered_intensity_outlier": bead_properties_df["considered_intensity_outlier"].sum(),
"nr_of_beads_considered_intensity_outlier": bead_properties_df[
"considered_intensity_outlier"
].sum(),
"nr_of_beads_considered_bad_z_fit": bead_properties_df["considered_bad_z_fit"].sum(),
"nr_of_beads_considered_bad_y_fit": bead_properties_df["considered_bad_y_fit"].sum(),
"nr_of_beads_considered_bad_x_fit": bead_properties_df["considered_bad_x_fit"].sum(),
Expand Down Expand Up @@ -135,9 +138,15 @@ def _generate_key_values(
"resolution_mean_fwhm_x_microns": bead_properties_df["x_fwhm_micron"].mean() or None,
"resolution_median_fwhm_x_microns": bead_properties_df["x_fwhm_micron"].median() or None,
"resolution_std_fwhm_x_microns": bead_properties_df["x_fwhm_micron"].std() or None,
"resolution_mean_fwhm_lateral_asymmetry_ratio": bead_properties_df["fwhm_lateral_asymmetry_ratio"].mean(),
"resolution_median_fwhm_lateral_asymmetry_ratio": bead_properties_df["fwhm_lateral_asymmetry_ratio"].median(),
"resolution_std_fwhm_lateral_asymmetry_ratio": bead_properties_df["fwhm_lateral_asymmetry_ratio"].std(),
"resolution_mean_fwhm_lateral_asymmetry_ratio": bead_properties_df[
"fwhm_lateral_asymmetry_ratio"
].mean(),
"resolution_median_fwhm_lateral_asymmetry_ratio": bead_properties_df[
"fwhm_lateral_asymmetry_ratio"
].median(),
"resolution_std_fwhm_lateral_asymmetry_ratio": bead_properties_df[
"fwhm_lateral_asymmetry_ratio"
].std(),
}


Expand Down Expand Up @@ -208,6 +217,7 @@ def _find_beads(channel: np.ndarray, sigma: Tuple[float, float, float], min_dist
threshold_rel=0.2,
min_distance=int(min_distance),
exclude_border=(1, int(min_distance // 2), int(min_distance // 2)),
p_norm=2,
)

# Convert arrays to sets for easier comparison
Expand Down Expand Up @@ -241,8 +251,8 @@ def _find_beads(channel: np.ndarray, sigma: Tuple[float, float, float], min_dist
bead_images = [
channel[
:,
(pos[1] - int(min_distance // 2)):(pos[1] + int(min_distance // 2)),
(pos[2] - int(min_distance // 2)):(pos[2] + int(min_distance // 2)),
(pos[1] - int(min_distance // 2)) : (pos[1] + int(min_distance // 2)),
(pos[2] - int(min_distance // 2)) : (pos[2] + int(min_distance // 2)),
]
for pos in valid_positions
]
Expand Down Expand Up @@ -384,7 +394,7 @@ class PSFBeadsAnalysis(mm_schema.PSFBeadsDataset, AnalysisMixin):

def _estimate_min_bead_distance(self):
# TODO: get the resolution somewhere or pass it as a metadata
return 1 * self.input.min_lateral_distance_factor
return self.input.min_lateral_distance_factor

def _generate_centroids_roi(
self, positions, root_name, color, stroke_width, positions_filter=None
Expand Down Expand Up @@ -594,9 +604,9 @@ def run(self) -> bool:
bead_properties["min_intensity_min"].append(bead.min())
bead_properties["intensity_std"].append(bead.std())
bead_properties["intensity_zscore"].append(bead_zscores[image_label][ch][i])
bead_properties[
"considered_intensity_outlier"
].append(bead_considered_intensity_outlier[image_label][ch][i])
bead_properties["considered_intensity_outlier"].append(
bead_considered_intensity_outlier[image_label][ch][i]
)
bead_properties["z_centroid"].append(bead_positions[image_label][ch][i][0])
bead_properties["y_centroid"].append(bead_positions[image_label][ch][i][1])
bead_properties["x_centroid"].append(bead_positions[image_label][ch][i][2])
Expand All @@ -616,17 +626,18 @@ def run(self) -> bool:
bead_properties["y_fwhm"].append(bead_fwhms[image_label][ch][i][1])
bead_properties["x_fwhm"].append(bead_fwhms[image_label][ch][i][2])
bead_properties["fwhm_lateral_asymmetry_ratio"].append(
max(
bead_fwhms[image_label][ch][i][1],
bead_fwhms[image_label][ch][i][2]
) / min(
bead_fwhms[image_label][ch][i][1],
bead_fwhms[image_label][ch][i][2]
)
max(bead_fwhms[image_label][ch][i][1], bead_fwhms[image_label][ch][i][2])
/ min(bead_fwhms[image_label][ch][i][1], bead_fwhms[image_label][ch][i][2])
)
bead_properties["z_fwhm_micron"].append(
bead_fwhms_micron[image_label][ch][i][0]
)
bead_properties["y_fwhm_micron"].append(
bead_fwhms_micron[image_label][ch][i][1]
)
bead_properties["x_fwhm_micron"].append(
bead_fwhms_micron[image_label][ch][i][2]
)
bead_properties["z_fwhm_micron"].append(bead_fwhms_micron[image_label][ch][i][0])
bead_properties["y_fwhm_micron"].append(bead_fwhms_micron[image_label][ch][i][1])
bead_properties["x_fwhm_micron"].append(bead_fwhms_micron[image_label][ch][i][2])
bead_properties["considered_axial_edge"].append(
bead_considered_axial_edge[image_label][ch][i]
)
Expand Down Expand Up @@ -686,7 +697,6 @@ def run(self) -> bool:
return True



# Calculate 2D FFT
# slice_2d = raw_img[17, ...].reshape([1, n_channels, x_size, y_size])
# fft_2D = fft_2d(slice_2d)
Expand Down

0 comments on commit 9c68803

Please sign in to comment.