Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CREATE-FFPE: Cross-Resolution Compensated and Multi-Frequency Enhanced FS-to-FFPE Stain Transfer for Intraoperative IHC Images

Yiyang Lin, Danling Jiang, Xinyu Liu, Yun Miao, Yixuan Yuan

TL;DR

The paper addresses the gap between rapid FS IHC during surgery and the higher quality FFPE staining by proposing CREATE-FFPE, a cross-resolution compensated and multi-frequency FS-to-FFPE stain transfer framework. It introduces two novel modules, CRCM to provide broader contextual information and WDGM to enhance high-frequency details, integrated within a GAN-based translation. The approach achieves state-of-the-art quantitative gains (notably substantial reductions in FID and KID×100) and improves downstream microsatellite instability prediction, demonstrating practical potential for improving intraoperative decision-making. This work enables high-quality intraoperative visualization and invites further extensions to other histopathology staining transfers.

Abstract

In the immunohistochemical (IHC) analysis during surgery, frozen-section (FS) images are used to determine the benignity or malignancy of the tumor. However, FS image faces problems such as image contamination and poor nuclear detail, which may disturb the pathologist's diagnosis. In contrast, formalin-fixed and paraffin-embedded (FFPE) image has a higher staining quality, but it requires quite a long time to prepare and thus is not feasible during surgery. To help pathologists observe IHC images with high quality in surgery, this paper proposes a Cross-REsolution compensATed and multi-frequency Enhanced FS-to-FFPE (CREATE-FFPE) stain transfer framework, which is the first FS-to-FFPE method for the intraoperative IHC images. To solve the slide contamination and poor nuclear detail mentioned above, we propose the cross-resolution compensation module (CRCM) and the wavelet detail guidance module (WDGM). Specifically, CRCM compensates for information loss due to contamination by providing more tissue information across multiple resolutions, while WDGM produces the desirable details in a wavelet way, and the details can be used to guide the stain transfer to be more precise. Experiments show our method can beat all the competing methods on our dataset. In addition, the FID has decreased by 44.4%, and KID*100 has decreased by 71.2% by adding the proposed CRCM and WDGM in ablation studies, and the performance of a downstream microsatellite instability prediction task with public dataset can be greatly improved by performing our FS-to-FFPE stain transfer.

CREATE-FFPE: Cross-Resolution Compensated and Multi-Frequency Enhanced FS-to-FFPE Stain Transfer for Intraoperative IHC Images

TL;DR

The paper addresses the gap between rapid FS IHC during surgery and the higher quality FFPE staining by proposing CREATE-FFPE, a cross-resolution compensated and multi-frequency FS-to-FFPE stain transfer framework. It introduces two novel modules, CRCM to provide broader contextual information and WDGM to enhance high-frequency details, integrated within a GAN-based translation. The approach achieves state-of-the-art quantitative gains (notably substantial reductions in FID and KID×100) and improves downstream microsatellite instability prediction, demonstrating practical potential for improving intraoperative decision-making. This work enables high-quality intraoperative visualization and invites further extensions to other histopathology staining transfers.

Abstract

In the immunohistochemical (IHC) analysis during surgery, frozen-section (FS) images are used to determine the benignity or malignancy of the tumor. However, FS image faces problems such as image contamination and poor nuclear detail, which may disturb the pathologist's diagnosis. In contrast, formalin-fixed and paraffin-embedded (FFPE) image has a higher staining quality, but it requires quite a long time to prepare and thus is not feasible during surgery. To help pathologists observe IHC images with high quality in surgery, this paper proposes a Cross-REsolution compensATed and multi-frequency Enhanced FS-to-FFPE (CREATE-FFPE) stain transfer framework, which is the first FS-to-FFPE method for the intraoperative IHC images. To solve the slide contamination and poor nuclear detail mentioned above, we propose the cross-resolution compensation module (CRCM) and the wavelet detail guidance module (WDGM). Specifically, CRCM compensates for information loss due to contamination by providing more tissue information across multiple resolutions, while WDGM produces the desirable details in a wavelet way, and the details can be used to guide the stain transfer to be more precise. Experiments show our method can beat all the competing methods on our dataset. In addition, the FID has decreased by 44.4%, and KID*100 has decreased by 71.2% by adding the proposed CRCM and WDGM in ablation studies, and the performance of a downstream microsatellite instability prediction task with public dataset can be greatly improved by performing our FS-to-FFPE stain transfer.

Paper Structure

This paper contains 12 sections, 9 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Method
  3. The Overview of CREATE-FFPE
  4. Cross-Resolution Compensation Module (CRCM)
  5. Wavelet Detail Guidance Module (WDGM)
  6. Loss Functions
  7. Experiments
  8. Dataset and Experimental Details
  9. Comparative Experiments
  10. Ablation Experiments
  11. Downstream Tasks
  12. Conclusion

Figures (4)

  • Figure 1: Examples of defects in FS images and our virtual FFPE. The defects mainly include background staining, weak positive staining, and poor nuclear detail.
  • Figure 2: The overview of our method. The proposed framework is composed of a main task, which can transfer the FS image to the FFPE one, as well as two auxiliary modules, CRCM in the orange box and WDGM in the green box.
  • Figure 3: The visualization results of our method and the competing methods. It is noted that since there are no paired FS and FFPE images in clinical practice, the ground truth is not shown here.
  • Figure 4: The visualization results of our ablation experiments. To better observe the details of the nucleus, we zoom in on some areas in the second row.