Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Comprehensive Dataset and Automated Pipeline for Nailfold Capillary Analysis

Linxi Zhao, Jiankai Tang, Dongyu Chen, Xiaohong Liu, Yong Zhou, Yuanchun Shi, Guangyu Wang, Yuntao Wang

TL;DR

A pioneering effort in constructing a comprehensive nailfold capillary dataset that serves as a crucial resource for training deep-learning models and integrated them into a novel end-to-end nailfold capillary analysis pipeline that excels in automatically detecting and measuring a wide range of size factors, morphological features, and dynamic aspects of nailfold capillaries.

Abstract

Nailfold capillaroscopy is widely used in assessing health conditions, highlighting the pressing need for an automated nailfold capillary analysis system. In this study, we present a pioneering effort in constructing a comprehensive nailfold capillary dataset-321 images, 219 videos from 68 subjects, with clinic reports and expert annotations-that serves as a crucial resource for training deep-learning models. Leveraging this dataset, we finetuned three deep learning models with expert annotations as supervised labels and integrated them into a novel end-to-end nailfold capillary analysis pipeline. This pipeline excels in automatically detecting and measuring a wide range of size factors, morphological features, and dynamic aspects of nailfold capillaries. We compared our outcomes with clinical reports. Experiment results showed that our automated pipeline achieves an average of sub-pixel level precision in measurements and 89.9% accuracy in identifying morphological abnormalities. These results underscore its potential for advancing quantitative medical research and enabling pervasive computing in healthcare. Our data and code are available at https://github.com/THU-CS-PI-LAB/ANFC-Automated-Nailfold-Capillary.

A Comprehensive Dataset and Automated Pipeline for Nailfold Capillary Analysis

TL;DR

A pioneering effort in constructing a comprehensive nailfold capillary dataset that serves as a crucial resource for training deep-learning models and integrated them into a novel end-to-end nailfold capillary analysis pipeline that excels in automatically detecting and measuring a wide range of size factors, morphological features, and dynamic aspects of nailfold capillaries.

Abstract

Nailfold capillaroscopy is widely used in assessing health conditions, highlighting the pressing need for an automated nailfold capillary analysis system. In this study, we present a pioneering effort in constructing a comprehensive nailfold capillary dataset-321 images, 219 videos from 68 subjects, with clinic reports and expert annotations-that serves as a crucial resource for training deep-learning models. Leveraging this dataset, we finetuned three deep learning models with expert annotations as supervised labels and integrated them into a novel end-to-end nailfold capillary analysis pipeline. This pipeline excels in automatically detecting and measuring a wide range of size factors, morphological features, and dynamic aspects of nailfold capillaries. We compared our outcomes with clinical reports. Experiment results showed that our automated pipeline achieves an average of sub-pixel level precision in measurements and 89.9% accuracy in identifying morphological abnormalities. These results underscore its potential for advancing quantitative medical research and enabling pervasive computing in healthcare. Our data and code are available at https://github.com/THU-CS-PI-LAB/ANFC-Automated-Nailfold-Capillary.
Paper Structure (12 sections, 5 figures, 3 tables)

This paper contains 12 sections, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Dataset
  3. Data collection
  4. Data annotation
  5. Method and Results
  6. End-to-end nailfold capillary image analysis pipeline
  7. End-to-end nailfold capillary video analysis pipeline
  8. Experimental results
  9. Discussion
  10. Conclusion
  11. Compliance with Ethical Standards
  12. Acknowledgments

Figures (5)

  • Figure 1: Data Collection and Annotation. (b) Segmentation with green, red, and yellow annotations for normal, enlarged, and blurred capillaries. (c) Keypoint annotation.
  • Figure 2: Nailfold Capillary Image End-to-end Analysis Pipeline
  • Figure 3: Nailfold Capillary Video End-to-end Analysis Pipeline
  • Figure 5: Exemplifying the Stages of Our Nailfold Capillary Image Analysis System. Excluded capillary refers to those initially proposed and subsequently excluded in the pipeline.
  • Figure 6: Detection of White Blood Cells (WBCs) in Our Nailfold Capillary Video Analysis System. Visualization of WBC traversal through a given capillary as bright bands in (a), with corresponding detected local maxima annotated in a black circle in (b).