Model-Based Correction of Segmentation Errors in Digitised Histological Images

Abstract

Abstract. This paper describes an application of topological, modelbased methods for the algorithmic correction of segmentation errors in digitised histological images. The topological analysis is provided by the spatial logic Discrete Mereotopology and integrates qualitative spatial reasoning and constraint satisfaction methods with classical image processing methods. A set of eight topological relations defined on binary segmented regions are factored out and reworked as nodes of a set of directed graphs. The graphs encode and constrain a set of set-theoretic and topological segmentation operations on regions, so that the interpreted images and any proposed changes made to the regions can be made to conform to a valid histological model. Worked examples are given using images of H&E stained H400 cell line cultures.

Mereotopological Correction of Segmentation Errors in Histological Imaging

“…This paper extends the method that was originally reported in [9] and developed in [1]. Specifically, eight graphs are added to the twelve defined in [1] to cover the set theoretic operators: sum, prod, diff, compl, xor (denoting union, intersection, relative difference, complement, and symmetric difference) and the discrete topological operators int D , cl D , ext D , and bndry D (denoting the discrete interior, closure, exterior, and boundary).…”

“…Specifically, eight graphs are added to the twelve defined in [1] to cover the set theoretic operators: sum, prod, diff, compl, xor (denoting union, intersection, relative difference, complement, and symmetric difference) and the discrete topological operators int D , cl D , ext D , and bndry D (denoting the discrete interior, closure, exterior, and boundary). The method is generic and does not favour any one particular image segmentation method over another, and all that is assumed is that pairs of regions can be represented as instances of binary relations.…”

“…This paper is an extended version of the work presented at the 21st Conference on Medical Image Understanding and Analysis (MIUA 2017) [1] where we describe an application of mereotopological model-based methods for the algorithmic correction of segmentation errors. In this approach, qualitative spatial reasoning (QSR) and constraint-satisfaction methods are integrated into image processing routines to create context-based histological imaging algorithms.…”