Fix: Segmentation "Hide All" fails when switching from 2D to 3D four-…

[nithin-trenser]

Context

Fix issue: #5948

After loading a segmentation file in the 2D viewport, switching to the 3D four-up viewport causes the "Hide All" functionality to stop working. When the user selects "Hide All", the segmentations remain visible instead of being hidden.

Changes & Results

Refactor segmentation type selection to use explicit viewport-based logic.
For 3D viewports, enforce Surface; for 2D, allow Contour or default to Labelmap.
This change will ensure consistent behavior and prevent unsupported types from propagating

Seg-3d-four-up-hide-error.mp4

Testing

• Load a study with a segmentation file : https://viewer.ohif.org/segmentation?StudyInstanceUIDs=1.3.6.1.4.1.14519.5.2.1.256467663913010332776401703474716742458
• Display the segmentation in a 2D viewport
• Switch the layout to 3D four-up viewport
• Click on "Hide All"

Checklist

PR

• My Pull Request title is descriptive, accurate and follows the
  semantic-release format and guidelines.

Code

• My code has been well-documented (function documentation, inline comments,
  etc.)

Public Documentation Updates

• The documentation page has been updated as necessary for any public API
  additions or removals.

Tested Environment

• OS: Ubuntu 24.04
• Node version: 22
• Browser: Chrome 147.0.7727.138

Greptile Summary

Fixes the "Hide All" regression that occurred when switching from a 2D viewport (with a Labelmap/Contour segmentation) to a 3D four-up layout. The root cause was that adding a Surface representation to the 3D viewport fired a SEGMENTATION_REPRESENTATION_MODIFIED event, and the old ?? Labelmap fallback would propagate Surface to fresh 2D viewports instead of defaulting to Labelmap.

• The new ternary chain for non-3D viewports explicitly guards against Surface being assigned to a 2D viewport: Contour is kept as-is, any other non-Surface truthy value is passed through, and Surface (or null/undefined) falls back to Labelmap.
• The 3D path is unchanged — Surface is always enforced there.

Confidence Score: 5/5

The change is narrow and well-targeted — it only affects the type-selection step inside the synchroniser callback, and the new logic correctly handles all current representation values (Labelmap, Contour, Surface) as well as null/undefined.

The fix is a single ternary chain that guards 2D viewports against receiving a Surface representation when the synchroniser fires after the 3D viewport is initialised. All existing cases (Contour preserved, Labelmap preserved, Surface → Labelmap, null → Labelmap) are handled correctly, and the 3D path is untouched.

No files require special attention.

Important Files Changed

Filename	Overview
extensions/cornerstone/src/services/SyncGroupService/createHydrateSegmentationSynchronizer.ts	Refactors the representation-type selection so a 2D viewport can no longer inherit a Surface representation propagated from the newly-added 3D viewport, fixing the "Hide All" regression.

_{Reviews (3): Last reviewed commit: "Merge branch 'master' into fix-seg-load-..." | Re-trigger Greptile}

[netlify]

❌ Deploy Preview for ohif-dev failed. Why did it fail? →

Name	Link
🔨 Latest commit	498b11c
🔍 Latest deploy log	https://app.netlify.com/projects/ohif-dev/deploys/6a0ec662bb5c400008a9643a

[sen-trenser]

@sedghi Could you please take a look at this PR and provide your feedback?
Thanks!

[Belbin-GK]

@wayfarer3130 Could you please review this PR?