# ECG Segmentation by Neural Networks: Errors and Correction

## TL;DR

This paper investigates how deep convolutional network ensembles correct errors in ECG segmentation and explores using ensemble error information to evaluate data representation quality, highlighting the role of outlier distillation.

## Contribution

It introduces a method to analyze error correction in ECG segmentation ensembles and demonstrates the potential of ensemble errors for data quality assessment.

## Key findings

- Ensemble error correction improves ECG segmentation accuracy.
- Outlier distillation helps evaluate data representation quality.
- Ensemble errors can indicate data and model issues.

## Abstract

In this study we examined the question of how error correction occurs in an ensemble of deep convolutional networks, trained for an important applied problem: segmentation of Electrocardiograms(ECG). We also explore the possibility of using the information about ensemble errors to evaluate a quality of data representation, built by the network. This possibility arises from the effect of distillation of outliers, which was demonstarted for the ensemble, described in this paper.