Image Panels¶

Use This When¶

Use image panels when you need to inspect the microscopy images behind the tables. plot_images browses imported source images. plot_representative_images draws a curated panel from images that have already been selected as representative examples.

These plots are visual checks and figure panels, not statistical tests.

Input Data¶

Image panels need an image table on the input object. A Batch or Experiment must have run image import, either through processData(import_images=True), importImages(), or getImageTable().

The image table must include image paths and labels such as AnimalName, Marker, ROI, ImageName, and ImagePath. Representative panels also need a representative_images table that records the selected subjects, groups, ROIs, and marker set.

Raw pandas.DataFrame summary tables are not enough for these plots because they do not carry image paths.

Main Functions¶

Function	Registry name	Use
`plot_images`	`images`	Browse or save grids of imported images, optionally filtered by marker, subject, or ROI.
`plot_representative_images`	`representative_images`	Draw saved representative selections and export the selected source-image files.

Common Options¶

Option	Meaning
`markers`	Marker names to display. Use a list for several markers.
`subject_filter` / `animal_filter`	Restrict the image rows to one or more subjects.
`roi_filter`	Restrict `plot_images` to matching ROI labels.
`merge`	Add merged marker panels when several markers are requested.
`draw_rois`	Draw ROI outlines when ROI geometry is available.
`scale_bar`	Add a scale bar using image width or pixel-size information.
`fast_loading`, `preview_max_dim`, `image_workers`	Speed controls for large images.
`edit_mode`, `image_adjustments`, `use_existing_edits`	Interactive or saved brightness/contrast adjustments.

Outputs¶

Both functions return a matplotlib.figure.Figure. The returned figure carries PyFLASH metadata attributes such as the filtered image table and saved path.

With save=True, plot_images writes a figure under the object's image figure folder. plot_representative_images writes a representative image figure, a representative_image.csv file, and copied source-image files in the representative export folder.

Examples¶

Browse one marker without saving:

from PyFLASH.plotting import plot_images

fig = plot_images(batch, markers=["DAPI"], save=False, show=False)
print(fig.PyFLASH_image_df[["AnimalName", "Marker", "ROI"]].head())

Render representative panels after selections have been stored:

from PyFLASH.plotting import plot_representative_images

fig = plot_representative_images(
    batch,
    markers=["DAPI", "GFAP"],
    merge=True,
    save=True,
    show=False,
)
print(fig.PyFLASH_save_path)

Interpretation¶

Use image panels to check whether the plotted measurements correspond to credible images, whether ROI labels match the expected anatomy, and whether representative examples are balanced across conditions. Do not treat a representative image panel as evidence of a group effect unless it is paired with a quantitative summary and an appropriate statistical analysis.

For very large image sets, start with fast_loading=True or preview_max_dim=1024, then rerun the final panel without preview downsampling if full-resolution output matters.