HotspotToSpots

HotspotToSpots converts a 2D hotspot image into spot dataframe rows. Connected nonzero regions above threshold become one spot each.

Keep this tool public because HotspotDetection produces an image-like hotspot output, while downstream analyst workflows usually need tabular spot centroids for per-cell counting, intensity summaries, QC plots, or export through explicit table writer tools.

Inputs

• hotspot_image: 2D hotspot score or mask image.

• threshold: minimum value included in connected components.

Outputs

• spot_id, y, x, intensity, score, area, and label.

• spot_count.

No spot CSV artifact is written because BioImageFlow records these rows in the output dataframe. If no connected components pass the threshold, direct process_row() returns an empty list and workflow execution records spot_count=0 as manifest-only scalar metadata.

Dependencies and Core Libraries

imageio, NumPy, and package-local connected-component traversal.

Minimal Example

from bioimageflow_core import Arguments
from bioimageflow_sairpico_tools import HotspotToSpots

HotspotToSpots().process_row(
    Arguments(hotspot_image="hotspot.tif", threshold=0.5)
)

Expected Results

Synthetic hotspot masks produce one dataframe row per connected component with stable centroid and area values. Blank hotspot masks produce no dataframe rows; the selected v1 record manifest and run node result expose spot_count=0 as scalar_output metadata.

Failure Modes

Unreadable images, unsupported dimensions, and invalid thresholds raise errors.