Volume-Rendering-In-iOS

Implement volume-rendering for patient-data on iOS. Direct volume rendering, projection modes, and multi-planar reconstruction on iOS using SceneKit + Metal.

This fork extends the original sample with enhanced projection pipelines, tri-planar multi-planar reconstruction (MPR), and optional DICOM loading through GDCM.

Tech

• Metal Shader (Graphic)
• SceneKit (Scene Graph)
• Metal fragment ray-marching for DVR/MIP/MinIP/AIP
• Optional GDCM bridge for native DICOM series loading
• Snapshot helpers for debugging transfer functions and volume slices

Source Project

Unity Volume Rendering

Data

• Sample volumes are included in the app bundle under VolumeRendering-iOS/Resource/Images/
  • chest.raw.zip (Int16 signed LE)
  • head.raw.zip (Int16 signed LE)
• Transfer Functions (TF) in VolumeRendering-iOS/Resource/TransferFunction/
  • ct_arteries.tf, ct_entire.tf, ct_lung.tf
• DICOM: when GDCM libraries are linked (see below), .dcm series and zipped folders can be imported at runtime

Features

• Rendering Modes: Direct Volume Rendering (DVR), Surface, Maximum Intensity Projection (MIP), Minimum Intensity Projection (MinIP), Average Intensity Projection (AIP)
• Projection Enhancements: optional transfer function application, HU windowing, and min/max/mean slab thickness controls
• Tri-planar MPR: simultaneous axial/coronal/sagittal reconstructions with draggable crosshairs, slab thickness control, oblique rotation, and shared 1D transfer function
• Transfer Function 1D: shared between DVR and MPR with presets for common CT windows
• Empty-space skipping (conservative) for DVR performance
• Snapshot helpers for exporting volume slices and transfer function textures during development

Contributors

• Thales Matheus Mendonça Santos — refined MinIP, AIP, and DVR pipelines, and implemented the tri-planar MPR workflow with optional GDCM-based DICOM integration.

How To Run in your Local

At First,

You should set git-lfs setting.
Because raw data file is so bigger than supported in git.

install git-lfs first,

brew install git-lfs

and set git lfs on in your local

git-lfs install

and pull lfs from server

git lfs pull

Run from Xcode

1. Open the Xcode project VolumeRendering-iOS.xcodeproj.
2. Select a device (prefer iPhone 15 Pro Max) and run.

DICOM via GDCM

1. Build GDCM for iOS (static libs or an XCFramework). A simple path is to use CMake with the iOS toolchain or to reuse prebuilt binaries from your toolchain.
2. Copy the resulting headers and libraries into Vendor/GDCM/include and Vendor/GDCM/lib respectively. The Xcode target already adds these locations to the header and library search paths.
3. Add the required static libraries (for example libgdcmCommon, libgdcmMSFF, libexpat, libz, libopenjp2, libcharls) to Link Binary With Libraries. If you use an .xcframework, drop it in the folder and drag it into Xcode.
4. Run the app, tap Import DICOM, and select a .zip, folder, or file representing a series. When GDCM is not linked, the importer gracefully reports that the native loader is unavailable.

Screenshots

Surface Rendering	Direct Volume Rendering	Maximum Intensity Projection

surface rendring

direct volume rendering

maximum intensity projection

Direct Volume Rendering

CT-Coronary-Arteries	CT-Chest-Entire	CT-Lung

CT-Coronary-Arteries

CT-Chest-Entire

CT-Lung

Lighting

Lighting Off	Lighting On

Lighting Off

Lighting On

Licenses

• This project is based on Unity Volume Rendering (mlavik1/UnityVolumeRendering). See upstream for original license.
• All new code in this repository is released under the Apache License 2.0 (LICENSE).

Known issues

• Performance metrics depend on device; baseline validation targets iPhone 15 Pro Max.
• Some presets may require TF shift adjustments for best visual results.
• Large RAW datasets require Git LFS; ensure git lfs pull completed successfully.