Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Combining Hough Transform and Deep Learning Approaches to Reconstruct ECG Signals From Printouts

Felix Krones, Ben Walker, Terry Lyons, Adam Mahdi

TL;DR

This work presents the team's (SignalSavants) winning contribution to the 2024 George B. Moody PhysioNet Challenge, and shows the challenges of building robust, generalisable, digitisation approaches.

Abstract

This work presents our team's (SignalSavants) winning contribution to the 2024 George B. Moody PhysioNet Challenge. The Challenge had two goals: reconstruct ECG signals from printouts and classify them for cardiac diseases. Our focus was the first task. Despite many ECGs being digitally recorded today, paper ECGs remain common throughout the world. Digitising them could help build more diverse datasets and enable automated analyses. However, the presence of varying recording standards and poor image quality requires a data-centric approach for developing robust models that can generalise effectively. Our approach combines the creation of a diverse training set, Hough transform to rotate images, a U-Net based segmentation model to identify individual signals, and mask vectorisation to reconstruct the signals. We assessed the performance of our models using the 10-fold stratified cross-validation (CV) split of 21,799 recordings proposed by the PTB-XL dataset. On the digitisation task, our model achieved an average CV signal-to-noise ratio of 17.02 and an official Challenge score of 12.15 on the hidden set, securing first place in the competition. Our study shows the challenges of building robust, generalisable, digitisation approaches. Such models require large amounts of resources (data, time, and computational power) but have great potential in diversifying the data available.

Combining Hough Transform and Deep Learning Approaches to Reconstruct ECG Signals From Printouts

TL;DR

This work presents the team's (SignalSavants) winning contribution to the 2024 George B. Moody PhysioNet Challenge, and shows the challenges of building robust, generalisable, digitisation approaches.

Abstract

This work presents our team's (SignalSavants) winning contribution to the 2024 George B. Moody PhysioNet Challenge. The Challenge had two goals: reconstruct ECG signals from printouts and classify them for cardiac diseases. Our focus was the first task. Despite many ECGs being digitally recorded today, paper ECGs remain common throughout the world. Digitising them could help build more diverse datasets and enable automated analyses. However, the presence of varying recording standards and poor image quality requires a data-centric approach for developing robust models that can generalise effectively. Our approach combines the creation of a diverse training set, Hough transform to rotate images, a U-Net based segmentation model to identify individual signals, and mask vectorisation to reconstruct the signals. We assessed the performance of our models using the 10-fold stratified cross-validation (CV) split of 21,799 recordings proposed by the PTB-XL dataset. On the digitisation task, our model achieved an average CV signal-to-noise ratio of 17.02 and an official Challenge score of 12.15 on the hidden set, securing first place in the competition. Our study shows the challenges of building robust, generalisable, digitisation approaches. Such models require large amounts of resources (data, time, and computational power) but have great potential in diversifying the data available.

Paper Structure

This paper contains 7 sections, 1 equation, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Data
  4. Architecture
  5. Scoring
  6. Results
  7. Discussion

Figures (4)

  • Figure 1: Four example images of signal 20999. (a) rotated by one degree, (b) rotated by two degrees with wrinkles and shadows, (c) rotated by three degrees with wrinkles and shadows, (d) not rotated with wrinkles and shadows.
  • Figure 2: A schematic diagram of our model architecture. Blue: Data, Yellow: Rule-based engineering, Green: Deep learning.
  • Figure 3: Example of predicted masks for signal 20999.
  • Figure 4: Vectorisation example for signal 20999. Left: original and predicted signal. Right: signal difference.