Skip to content

Fura Imaging Lab

Incomplete — under development

Fura Imaging Lab is incomplete. Its available tools are intended for evaluation and continued development; the plugin should not be treated as a complete, validated imaging-analysis workflow. Check results against known inputs and an independently validated method before relying on them for research.

Fura Imaging Lab provides a standalone Fura-2 imaging workspace within IntegraPose. Available tools include TIFF-stack import, frame alignment, ROI tracking, trace analysis, and workbook export. The sections below describe the implemented tools and their scientific conventions.

When to use it

Evaluation use cases Outside this role
Fura-2 ratiometric calcium imaging from upright or inverted scopes Wide-field behavior video - that's the main app's job
Quick alignment + ROI tracking workflows that don't need a full ImageJ pipeline Highly customized post-processing already handled by lab-specific MATLAB tools
Testing ROI traces and workbook exports against known imaging inputs Studies requiring a complete, validated analysis workflow

What it does

  1. Stack import. Reads multi-page TIFFs directly when the optional tifffile package is installed.
  2. Frame alignment. Corrects drift between frames so an ROI drawn once stays on the cell across the recording.
  3. ROI tracking. Draw, label, and (optionally) track ROIs across the stack.
  4. Trace analysis. Computes per-ROI mean 340/380 nm intensities, background-subtracted signals, ratios, baseline correction, and event metrics.
  5. Trace export. Writes raw signals, corrected signals, metrics, ROI definitions, and analysis parameters to a multi-sheet workbook.

Scientific conventions

  • AXI channel frames are paired one-to-one by acquisition timestamp in temporal order, using a tolerance of 75% of the median within-channel interval; unmatched acquisitions are excluded. Each pair is assigned the midpoint of its 340 and 380 timestamps. The export retains both source timestamps, source indices, and pair separation.
  • Ratios are mean(340) / mean(380). A non-finite or nonpositive 380 denominator, including after background subtraction, yields a missing ratio instead of an artificial large value.
  • Without a background ROI, raw traces remain available but background-subtracted columns are missing; raw values are never duplicated under a background-subtracted label.
  • ROI traces are mean intensities and are therefore already area normalized. Dividing those means by ROI area is not offered.
  • Each baseline interval contributes one mean baseline anchor. Multiple anchors are linearly interpolated, with constant values outside the first and last anchor; a single interval produces constant baseline subtraction.
  • Requested baseline normalization fails visibly when its baseline samples, mean, or variance are insufficient. Raw values are never silently relabeled as normalized values.

Required dependencies

tifffile is optional but strongly recommended for direct TIFF support. Install with:

pip install tifffile

The plugin runs on standard image arrays even without tifffile, but multi-page TIFFs will need a manual conversion step otherwise.

Output layout

The selected .xlsx workbook contains tracking coordinates, raw and background-subtracted signals, ratios, normalized analysis input, smoothed and baseline-corrected traces, event metrics, and the parameters needed to interpret the run.

Practical advice

  • Run alignment before drawing ROIs; otherwise drift smears the trace.
  • Keep ROI labels short and meaningful - they become column names in the workbook.
  • Review the timestamp-matched pair count after import; a lower count than either source channel indicates unmatched acquisitions.

Where this fits

Fura Imaging Lab operates on imaging inputs and exports its own workbook. It does not synchronize these recordings with pose or behavior outputs.