Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Radiologist-in-the-Loop Self-Training for Generalizable CT Metal Artifact Reduction

Chenglong Ma, Zilong Li, Yuanlin Li, Jing Han, Junping Zhang, Yi Zhang, Jiannan Liu, Hongming Shan

TL;DR

A novel radiologist-in-the-loop self-training framework for MAR, termed RISE-MAR, which can integrate radiologists’ feedback into the semi-supervised learning process, progressively improving the quality and quantity of pseudo ground-truths for enhanced generalization on real clinical CT images.

Abstract

Metal artifacts in computed tomography (CT) images can significantly degrade image quality and impede accurate diagnosis. Supervised metal artifact reduction (MAR) methods, trained using simulated datasets, often struggle to perform well on real clinical CT images due to a substantial domain gap. Although state-of-the-art semi-supervised methods use pseudo ground-truths generated by a prior network to mitigate this issue, their reliance on a fixed prior limits both the quality and quantity of these pseudo ground-truths, introducing confirmation bias and reducing clinical applicability. To address these limitations, we propose a novel Radiologist-In-the-loop SElf-training framework for MAR, termed RISE-MAR, which can integrate radiologists' feedback into the semi-supervised learning process, progressively improving the quality and quantity of pseudo ground-truths for enhanced generalization on real clinical CT images. For quality assurance, we introduce a clinical quality assessor model that emulates radiologist evaluations, effectively selecting high-quality pseudo ground-truths for semi-supervised training. For quantity assurance, our self-training framework iteratively generates additional high-quality pseudo ground-truths, expanding the clinical dataset and further improving model generalization. Extensive experimental results on multiple clinical datasets demonstrate the superior generalization performance of our RISE-MAR over state-of-the-art methods, advancing the development of MAR models for practical application. Code is available at https://github.com/Masaaki-75/rise-mar.

Radiologist-in-the-Loop Self-Training for Generalizable CT Metal Artifact Reduction

TL;DR

A novel radiologist-in-the-loop self-training framework for MAR, termed RISE-MAR, which can integrate radiologists’ feedback into the semi-supervised learning process, progressively improving the quality and quantity of pseudo ground-truths for enhanced generalization on real clinical CT images.

Abstract

Metal artifacts in computed tomography (CT) images can significantly degrade image quality and impede accurate diagnosis. Supervised metal artifact reduction (MAR) methods, trained using simulated datasets, often struggle to perform well on real clinical CT images due to a substantial domain gap. Although state-of-the-art semi-supervised methods use pseudo ground-truths generated by a prior network to mitigate this issue, their reliance on a fixed prior limits both the quality and quantity of these pseudo ground-truths, introducing confirmation bias and reducing clinical applicability. To address these limitations, we propose a novel Radiologist-In-the-loop SElf-training framework for MAR, termed RISE-MAR, which can integrate radiologists' feedback into the semi-supervised learning process, progressively improving the quality and quantity of pseudo ground-truths for enhanced generalization on real clinical CT images. For quality assurance, we introduce a clinical quality assessor model that emulates radiologist evaluations, effectively selecting high-quality pseudo ground-truths for semi-supervised training. For quantity assurance, our self-training framework iteratively generates additional high-quality pseudo ground-truths, expanding the clinical dataset and further improving model generalization. Extensive experimental results on multiple clinical datasets demonstrate the superior generalization performance of our RISE-MAR over state-of-the-art methods, advancing the development of MAR models for practical application. Code is available at https://github.com/Masaaki-75/rise-mar.
Paper Structure (28 sections, 9 equations, 10 figures, 4 tables)

This paper contains 28 sections, 9 equations, 10 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Clinical Quality Assessor
  4. Clinical Quality Assessment Collection
  5. Diverse Quality Augmentation
  6. Aligning CQA to Radiologists' Feedback
  7. Radiologist-in-the-Loop Self-Training
  8. Simulated-Domain Supervised Learning
  9. Clinical-Domain Unsupervised Learning
  10. Self-Training Objective Function
  11. Experiments
  12. Datasets
  13. DeepLesion Dataset
  14. CTPelvic1K Dataset
  15. Dental Dataset
  16. ...and 13 more sections

Figures (10)

  • Figure 1: Overview of RISE-MAR. (a) Radiologist-in-the-loop framework with a pretrained clinical quality assessor (CQA). (b) Overall architecture of CQA. "MA": metal artifact-affected, "GT": ground-truth. "sim.": simulated, "cli.": clinical.
  • Figure 2: Training of CQA. The predicted probability vector $\boldsymbol{q}_\psi$ is omitted for more intuitive presentation.
  • Figure 3: Visual results of different MAR methods on two typical slices from simulated CTPelvic1K test set. Sub-figures (a) and (c) are CT images, (b) and (d) are corresponding absolute error maps. For better visualization, metal pixels are filled with red color. Regions of interest are enlarged. The display window is [-300, 300] HU.
  • Figure 4: Profile line plot comparison of MAR methods. (a) CT image with a profile line (cyan colored); (b) HU values along the profile line.
  • Figure 5: Visual results of different MAR methods on four slices from the clinical test sets with real artifacts, where (a) and (b) are from DeepLesion dataset; (c) and (d) are from Dental dataset. The display window is [-300, 300] HU.
  • ...and 5 more figures