Swell: intrinsic optical signal imaging analysis for spreading depolarization event identification and segmentation

Swell is an open-source desktop application for analysing intrinsic optical signal (IOS) imaging recordings of spreading depolarization (SD), cortical spreading depression, and spreading depression events.

The application is organised around a two-window workflow. The host window loads image stacks, catalogues events manually or from an automated two-stage detector, attaches electrophysiological DC traces, and exports results. The analysis window opens a single event for interactive wavefront segmentation, using prompt, box, brush, eraser and fill tools together with SAM 2 mask propagation, and saves reviewed masks back to the project.

Exported metrics include optical intensity (ΔI/I₀), propagation speed, absolute area recruited, and area recruited relative to a user-defined region of interest, with pixel-to-millimetre calibration and configurable baseline ranges.

Packaged macOS and Windows builds require no local Python installation. Projects are stored as portable .swell containers holding masks, prompts, regions, and optionally the source images.

This release corresponds to the version used in the associated methods manuscript (in review).