DA-Capnet: Dual Attention Deep Learning Based on U-Net for Nailfold Capillary Segmentation

Yuli Sun Hariyani,Cheolsoo Park,Heesang Eom

doi:10.1109/access.2020.2965651

Yuli Sun Hariyani, Cheolsoo Park + Show 1 more

Open Access

https://doi.org/10.1109/access.2020.2965651

Copy DOI

Abstract

Automatic nailfold capillary segmentation is a challenging task owing to noise and large variabilities in images caused by insufficient focusing and low visibility of the capillaries. This task can be useful to detect and estimate the severity of autoimmune diseases of connective tissues or learning the status of white blood cells based on the cells’ blood flow on the nailfold capillary. Previous studies have addressed this task using manual, semi-automated, and automated segmentation method. However, further improvement is still required. With the recent progress of deep learning on medical imaging, we herein propose dual attention deep learning based on U-Net for nailfold capillary segmentation, named DA-CapNet. Our DA-CapNet improves the U-Net architecture by integrating a dual attention module that can capture a better representation of feature maps from input images. Furthermore, DA-CapNet is compared with three baselines: adaptive Gaussian algorithm, SegNet, the original U-Net. We experimentally demonstrate that our proposed method outperforms these baselines.

Highlights

Nailfold capillaroscopy is a non-invasive, inexpensive, and reproducible imaging technique to evaluate microcirculations under a nailfold, which is a small vessel under the nail
Nailfold capillary segmentation is a challenging task owing to sensitivity to external factors when capturing nailfold capillary images; large
Our contributions are as follows: 1) We propose a dual attention module that combines the benefits of previous attention modules, squeeze-excitation (SE) [19] and convolution block attention module (CBAM) [20], which is integrated into the U-Net architecture

Summary

Introduction

Nailfold capillaroscopy is a non-invasive, inexpensive, and reproducible imaging technique to evaluate microcirculations under a nailfold, which is a small vessel under the nail. The nailfold capillary could provide the status of white blood cells based on the cells’ blood flow in the microvessel using a light source of specific wavelength [1], [2]. This technique is typically used to monitor the microcirculations by analyzing the morphology of nailfold capillary such as shape, length, and width [3], [4]. Nailfold capillary segmentation is a challenging task owing to sensitivity to external factors when capturing nailfold capillary images (e.g., air bubbles trapped in the oil, or reflection owing to light source attached to the microscopy); large

Methods

Results

Conclusion