Compressibility Analysis for the differentiable shift-variant Filtered Backprojection Model

TL;DR

This paper introduces a PCA-based compression method for the differentiable shift-variant FBP model in CBCT, drastically reducing parameters and computational complexity while maintaining accuracy, thus enhancing practical applicability.

Contribution

The paper presents a novel PCA-based compression technique that reduces the model's trainable parameters by over 97%, improving efficiency without sacrificing reconstruction quality.

Findings

97.25% reduction in trainable parameters

Maintained reconstruction accuracy after compression

Significantly faster training process

Abstract

The differentiable shift-variant filtered backprojection (FBP) model enables the reconstruction of cone-beam computed tomography (CBCT) data for any non-circular trajectories. This method employs deep learning technique to estimate the redundancy weights required for reconstruction, given knowledge of the specific trajectory at optimization time. However, computing the redundancy weight for each projection remains computationally intensive. This paper presents a novel approach to compress and optimize the differentiable shift-variant FBP model based on Principal Component Analysis (PCA). We apply PCA to the redundancy weights learned from sinusoidal trajectory projection data, revealing significant parameter redundancy in the original model. By integrating PCA directly into the differentiable shift-variant FBP reconstruction pipeline, we develop a method that decomposes the redundancy weight layer parameters into a trainable eigenvector matrix, compressed weights, and a mean vector. This innovative technique achieves a remarkable 97.25% reduction in trainable parameters without compromising reconstruction accuracy. As a result, our algorithm significantly decreases the complexity of the differentiable shift-variant FBP model and greatly improves training speed. These improvements make the model substantially more practical for real-world applications.