# Typicality- and instance-dependent label noise-combating: a novel framework for simulating and combating real-world noisy labels for endoscopic polyp classification

## TL;DR

This paper introduces a new framework to simulate and combat label noise in medical image classification, particularly for endoscopic polyp detection, by considering the typicality of samples.

## Contribution

The paper introduces TIDN, a novel label noise model that incorporates sample typicality for more realistic noise simulation and improved classification.

## Key findings

- TIDN simulates real-world label noise more accurately than existing IIN and IDN models.
- The TIDN-combating framework outperforms others in classification performance with simulated and real-world noisy labels.
- A TIDN-attention module and recursive algorithm enhance the network's ability to correct noisy labels.

## Abstract

Learning with noisy labels aims to train neural networks with noisy labels. Current models handle instance-independent label noise (IIN) well; however, they fall short with real-world noise. In medical image classification, atypical samples frequently receive incorrect labels, rendering instance-dependent label noise (IDN) an accurate representation of real-world scenarios. However, the current IDN approaches fail to consider the typicality of samples, which hampers their ability to address real-world label noise effectively. To alleviate the issues, we introduce typicality- and instance-dependent label noise (TIDN) to simulate real-world noise and establish a TIDN-combating framework to combat label noise. Specifically, we use the sample’s distance to decision boundaries in the feature space to represent typicality. The TIDN is then generated according to typicality. We establish a TIDN-attention module to combat label noise and learn the transition matrix from latent ground truth to the observed noisy labels. A recursive algorithm that enables the network to make correct predictions with corrections from the learned transition matrix is proposed. Our experiments demonstrate that the TIDN simulates real-world noise more closely than the existing IIN and IDN. Furthermore, the TIDN-combating framework demonstrates superior classification performance when training with simulated TIDN and actual real-world noise.