Dual-Energy CT Toolkit: Projection, Decomposition & Reconstruction (Python + FBP)

Build and understand a complete dual-energy CT imaging pipeline — from raw fan-beam projections, to material decomposition, to full water/bone image reconstruction.

This toolkit gives you clean, readable Python scripts that you can run, modify, and extend for your own research, demos, or learning.

🚀 What’s Included

✔ Projection Generator

• Poly-energetic fan-beam forward projector

• Realistic 80 kVp + 140 kVp spectra

• Water + bone phantom with controllable geometry

• Outputs low- and high-energy sinograms

✔ Material Decomposition Module

• Linear 2×2 inversion

• Newton-based iterative refinement

• Returns pure material sinograms (L_water, L_bone)

✔ Reconstruction Engine

• Fan-beam FBP with ramp filtering

• Full-scan, Parker weighting, and generalized redundancy support

• 512×512 water and bone reconstructions

• Raw float32 output compatible with ImageJ, Slicer, Fiji, etc.

✔ Example Phantom + Ready-to-Run Demo

• Generates projections → decomposition → reconstruction

• Perfect for teaching, research, or algorithm testing

🔧 Why This Toolkit?

Most CT examples online stop at projections or at reconstruction.

This package gives you the entire pipeline end-to-end, with clear physics, readable code, and modular functions that you can drop into your own projects.

Use it to:

• Learn dual-energy physics

• Build your own CT simulation tools

• Validate ML reconstruction algorithms

• Create teaching demos and YouTube explanations

• Experiment with material imaging workflows

🧠 Who Is This For?

✔ Medical physicists

✔ ML researchers (CT physics)

✔ Imaging scientists

✔ Students learning CT reconstruction

✔ Anyone building simulation-based pipelines