Topological image modification for object detection and topological image processing of skin lesions

Robin Vandaele,Guillaume Adrien Nervo,Olivier Gevaert

doi:10.1038/s41598-020-77933-y

Robin Vandaele, Guillaume Adrien Nervo + Show 1 more

Open Access

https://doi.org/10.1038/s41598-020-77933-y

Copy DOI

Abstract

We propose a new method based on Topological Data Analysis (TDA) consisting of Topological Image Modification (TIM) and Topological Image Processing (TIP) for object detection. Through this newly introduced method, we artificially destruct irrelevant objects, and construct new objects with known topological properties in irrelevant regions of an image. This ensures that we are able to identify the important objects in relevant regions of the image. We do this by means of persistent homology, which allows us to simultaneously select appropriate thresholds, as well as the objects corresponding to these thresholds, and separate them from the noisy background of an image. This leads to a new image, processed in a completely unsupervised manner, from which one may more efficiently extract important objects. We demonstrate the usefulness of this proposed method for topological image processing through a case-study of unsupervised segmentation of the ISIC 2018 skin lesion images. Code for this project is available on https://bitbucket.org/ghentdatascience/topimgprocess.

Highlights

We propose a new method based on Topological Data Analysis (TDA) consisting of Topological Image Modification (TIM) and Topological Image Processing (TIP) for object detection
Any existing method for segmentation or object detection[10,18,19,20,21,22] could lead to a possibly generic method for TIP as a replacement of Algorithm 1 which we introduce in this work, we will make clear that these are unfit for this purpose in our experiments
We evaluated our work in an unsupervised context to show that TDA using TIM and TIP improves the task of skin lesion segmentation using generic unsupervised segmentation algorithms

Summary

Introduction

We propose a new method based on Topological Data Analysis (TDA) consisting of Topological Image Modification (TIM) and Topological Image Processing (TIP) for object detection. Through this newly introduced method, we artificially destruct irrelevant objects, and construct new objects with known topological properties in irrelevant regions of an image. We demonstrate the usefulness of this proposed method for topological image processing through a case-study of unsupervised segmentation of the ISIC 2018 skin lesion images. Many real-world images contain outliers, as well as irrelevant objects, which complicate the use of persistent homology for this purpose We enclose this gap by introducing Topological Image Modification (TIM). To the best of our knowledge, neither TIM, the concept of TIP, nor the flow ‘TIM → TIP → Segmentation’ has been introduced or studied before

Methods

Results

Discussion

Conclusion