Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Assessing the performance of CT image denoisers using Laguerre-Gauss Channelized Hotelling Observer for lesion detection

Prabhat Kc, Rongping Zeng

TL;DR

The paper addresses whether deep-learning CT denoisers that improve perceptual image quality necessarily enhance diagnostic detectability in CT. It combines perceptual metrics ($PSNR$, $SSIM$) with a task-based LCD assessment using the Laguerre-Gauss Channelized Hotelling Observer ($LG$-$CHO) on a $CCT$-$189$ phantom and LDCT data. Findings show that most DL denoisers raise $PSNR$ by $2.4$–$3.8$ dB and $SSIM$ by $0.05$–$0.11$, but these gains do not consistently translate into better LCD performance, with quarter-dose denoised outputs often inferior to normal-dose FBP. The work underscores the need for task-based validation when deploying denoisers clinically and points to future 3D studies and CAD-based lung nodule detection to further assess clinical impact.

Abstract

The remarkable success of deep learning methods in solving computer vision problems, such as image classification, object detection, scene understanding, image segmentation, etc., has paved the way for their application in biomedical imaging. One such application is in the field of CT image denoising, whereby deep learning methods are proposed to recover denoised images from noisy images acquired at low radiation. Outputs derived from applying deep learning denoising algorithms may appear clean and visually pleasing; however, the underlying diagnostic image quality may not be on par with their normal-dose CT counterparts. In this work, we assessed the image quality of deep learning denoising algorithms by making use of visual perception- and data fidelity-based task-agnostic metrics (like the PSNR and the SSIM) - commonly used in the computer vision - and a task-based detectability assessment (the LCD) - extensively used in the CT imaging. When compared against normal-dose CT images, the deep learning denoisers outperformed low-dose CT based on metrics like the PSNR (by 2.4 to 3.8 dB) and SSIM (by 0.05 to 0.11). However, based on the LCD performance, the detectability using quarter-dose denoised outputs was inferior to that obtained using normal-dose CT scans.

Assessing the performance of CT image denoisers using Laguerre-Gauss Channelized Hotelling Observer for lesion detection

TL;DR

The paper addresses whether deep-learning CT denoisers that improve perceptual image quality necessarily enhance diagnostic detectability in CT. It combines perceptual metrics (, ) with a task-based LCD assessment using the Laguerre-Gauss Channelized Hotelling Observer (-CCT189PSNR2.43.8SSIM0.050.11$, but these gains do not consistently translate into better LCD performance, with quarter-dose denoised outputs often inferior to normal-dose FBP. The work underscores the need for task-based validation when deploying denoisers clinically and points to future 3D studies and CAD-based lung nodule detection to further assess clinical impact.

Abstract

The remarkable success of deep learning methods in solving computer vision problems, such as image classification, object detection, scene understanding, image segmentation, etc., has paved the way for their application in biomedical imaging. One such application is in the field of CT image denoising, whereby deep learning methods are proposed to recover denoised images from noisy images acquired at low radiation. Outputs derived from applying deep learning denoising algorithms may appear clean and visually pleasing; however, the underlying diagnostic image quality may not be on par with their normal-dose CT counterparts. In this work, we assessed the image quality of deep learning denoising algorithms by making use of visual perception- and data fidelity-based task-agnostic metrics (like the PSNR and the SSIM) - commonly used in the computer vision - and a task-based detectability assessment (the LCD) - extensively used in the CT imaging. When compared against normal-dose CT images, the deep learning denoisers outperformed low-dose CT based on metrics like the PSNR (by 2.4 to 3.8 dB) and SSIM (by 0.05 to 0.11). However, based on the LCD performance, the detectability using quarter-dose denoised outputs was inferior to that obtained using normal-dose CT scans.

Paper Structure

This paper contains 6 sections, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Denoisers
  4. Image Quality Assessment
  5. Results
  6. Conclusion

Figures (1)

  • Figure 1: (a) CCT$189$ low contrast body phantom for measuring the low contrast detectability. (b,c) LG-CHO-based detectability performance for the $3$mm/$14$ HU insert for different denoisers.