Learnable Total Variation with Lambda Mapping for Low-Dose CT Denoising

TL;DR

This paper introduces a Learnable Total Variation framework with a pixel-wise regularization map for low-dose CT denoising, improving adaptivity and performance over traditional methods.

Contribution

It proposes coupling an unrolled TV solver with a neural network to predict spatially adaptive regularization, trained end-to-end for enhanced denoising.

Findings

Achieves up to +3.7 dB PSNR improvement.

Provides an interpretable alternative to CNNs.

Demonstrates consistent gains over classical TV and U-Net.

Abstract

While Total Variation (TV) excels in noise reduction and edge preservation, its reliance on a scalar regularization parameter limits adaptivity. In this study, we present a Learnable Total Variation (LTV) framework coupling an unrolled TV solver with a LambdaNet that predicts a per-pixel regularization map. The proposed framework is trained end-to-end to optimize reconstruction and regularization jointly, yielding spatially adaptive smoothing. Experiments on the DeepLesion dataset, using realistic LoDoPaB-CT simulation, show consistent gains over classical TV and FBP+U-Net, achieving up to +3.7 dB PSNR and 8% relative SSIM improvement. LTV provides an interpretable alternative to black-box CNNs for low-dose CT denoising.