Filopodia Analysis

Cell Edge Mask

Generates a binary cell-body mask from a fluorescence image.

Details

Step 1 of the FiloQuant pipeline. Converts the input to grayscale, applies a lower-bound intensity threshold, optionally fills interior holes, then smooths the mask with morphological opening (n_open erosions followed by n_open dilations). Extra erode+dilate cycles can be added to further refine the boundary.

Parameters:

• threshold — intensity cutoff (0--255)
• n_open — number of opening iterations for smoothing
• n_erode_dilate — additional erode+dilate cycles
• fill_holes — fill interior holes before opening

Output port mask is a MaskData (white = cell body).

Direction	Port	Type
Input	image	image
Output	mask	mask

Properties: Open Iterations, Extra Erode+Dilate, ``

---

Filopodia Detect

Detects filopodia candidates as a binary mask.

Details

Step 2 of the FiloQuant pipeline. Optionally applies CLAHE for local contrast enhancement and a 5x5 centre-surround sharpening convolution to accentuate thin bright structures, followed by a 3x3 median despeckle (x2) and intensity thresholding. Small isolated blobs (<8 px) are discarded. If a cell_mask is connected, an exclusion zone is dilated around the cell body so candidates too close to the cell edge are removed.

Parameters:

• threshold — intensity cutoff (0--255)
• n_distance_from_edge — exclusion zone width in pixels around the cell body
• use_convolve — enable 5x5 sharpening kernel
• use_clahe — enable CLAHE local contrast pre-enhancement

Output port mask is a MaskData of filopodia candidate regions. Connect to FilopodiaAnalyzeNode together with the cell_mask.

Direction	Port	Type
Input	cell_mask	mask
Output	mask	mask

Properties: Distance from Edge (px), ,

---

Filopodia Analyze

Skeletonizes the filopodia mask and measures each branch.

Details

Step 3 (final step) of the FiloQuant pipeline.

Processing steps:

• Subtract the cell body (cell_mask) from the filopodia candidate mask to isolate protrusions only
• Remove objects smaller than min_size_px
• Optionally close small gaps with repair_cycles morphological close iterations (FiloQuant's "Filopodia repair")
• Skeletonize via skimage.morphology.skeletonize
• Measure each connected skeleton branch with diagonal-aware edge counting
• Measure total cell edge length via skimage.measure.perimeter

Outputs:

• table — TableData with columns: x, y, filopodia_length_px, edge_length_px (one row per detected filopodium skeleton branch)
• visualization — colour composite (dark background, green = cell body, cyan = isolated filopodia mask, magenta = skeleton)

Direction	Port	Type
Input	filopodia_mask	mask
Input	cell_mask	mask
Output	visualization	image

Properties: Min Size (px), Repair Cycles (close)

---