White Light Specular Reflection Data Augmentation for Deep Learning Polyp Detection

TL;DR

This paper introduces a novel data augmentation technique that adds artificial white light reflections to colonoscopy images, aiming to improve deep learning models' accuracy in polyp detection by making training scenarios more challenging.

Contribution

The paper presents a new data augmentation method that simulates white light reflections, enhancing deep learning models' robustness in polyp detection during colonoscopies.

Findings

Improved polyp detection accuracy with the new augmentation.

Enhanced model robustness against false positives.

Effective in creating more challenging training scenarios.

Abstract

Colorectal cancer is one of the deadliest cancers today, but it can be prevented through early detection of malignant polyps in the colon, primarily via colonoscopies. While this method has saved many lives, human error remains a significant challenge, as missing a polyp could have fatal consequences for the patient. Deep learning (DL) polyp detectors offer a promising solution. However, existing DL polyp detectors often mistake white light reflections from the endoscope for polyps, which can lead to false positives.To address this challenge, in this paper, we propose a novel data augmentation approach that artificially adds more white light reflections to create harder training scenarios. Specifically, we first generate a bank of artificial lights using the training dataset. Then we find the regions of the training images that we should not add these artificial lights on. Finally, we propose a sliding window method to add the artificial light to the areas that fit of the training images, resulting in augmented images. By providing the model with more opportunities to make mistakes, we hypothesize that it will also have more chances to learn from those mistakes, ultimately improving its performance in polyp detection. Experimental results demonstrate the effectiveness of our new data augmentation method.