Sk-Unet Model with Fourier Domain for Mitosis Detection

TL;DR

This paper introduces a Fourier domain-based segmentation model, Sk-Unet, to improve mitosis detection in breast cancer images, effectively addressing domain shift issues across different scanners.

Contribution

The novel Sk-Unet model leverages Fourier domain techniques to mitigate domain shift, enhancing mitosis detection accuracy in histopathology images.

Findings

Achieved F1 score of 0.7456 on test set.

Effective domain shift mitigation through Fourier spectrum swapping.

Improved robustness across different scanner types.

Abstract

Mitotic count is the most important morphological feature of breast cancer grading. Many deep learning-based methods have been proposed but suffer from domain shift. In this work, we construct a Fourier-based segmentation model for mitosis detection to address the problem. Swapping the low-frequency spectrum of source and target images is shown effective to alleviate the discrepancy between different scanners. Our Fourier-based segmentation method can achieve F1 with 0.7456 on the preliminary test set.