The Active Segmentation Platform for Microscopic Image Classification and Segmentation.

Sumit K Vohra,Dimiter Prodanov

doi:10.3390/brainsci11121645

Abstract

Image segmentation still represents an active area of research since no universal solution can be identified. Traditional image segmentation algorithms are problem-specific and limited in scope. On the other hand, machine learning offers an alternative paradigm where predefined features are combined into different classifiers, providing pixel-level classification and segmentation. However, machine learning only can not address the question as to which features are appropriate for a certain classification problem. The article presents an automated image segmentation and classification platform, called Active Segmentation, which is based on ImageJ. The platform integrates expert domain knowledge, providing partial ground truth, with geometrical feature extraction based on multi-scale signal processing combined with machine learning. The approach in image segmentation is exemplified on the ISBI 2012 image segmentation challenge data set. As a second application we demonstrate whole image classification functionality based on the same principles. The approach is exemplified using the HeLa and HEp-2 data sets. Obtained results indicate that feature space enrichment properly balanced with feature selection functionality can achieve performance comparable to deep learning approaches. In summary, differential geometry can substantially improve the outcome of machine learning since it can enrich the underlying feature space with new geometrical invariant objects.

Highlights

Segmentation of cells and sub-cellular structures is a frequent intermediate step in cell biology and pathology [1,2]
Differential geometry can substantially improve the outcome of machine learning since it can enrich the underlying feature space with new geometrical invariant objects
For the purposes of comparison we have selected filters, which were implemented in Ilastik—Laplacian of Gaussian (LoG), Hessian of Gaussian, Structure Tensor—except for the ALoG filter

Summary

Introduction

Segmentation of cells and sub-cellular structures (e.g., organelles) is a frequent intermediate step in cell biology and pathology [1,2]. Sample preparation can be quite different, depending on the particular experimental question—i.e., light microscopy using colored histological stains, which is still the standard in pathology; vs fluorescent microscopy using various fluorescent stains or genetic fluorescent markers, such as the green fluorescent protein GFP; vs transmission electron microscopy (TEM) using heavy metal contrasts (for example osmium) for sub cellular structures; vs phase contrast microscopy for live cell imaging. Classification Model b= w2 w x + w b −1 w ·x+ w Both segmentation and classification results can be fine-tuned by presenting additional ROIs or by refining and editing the existing ROIs. Both segmentation and classification results can be fine-tuned by presenting additional ROIs or by refining and editing the existing ROIs This allows a high degree of tunability of the outcome, which corresponds with practice. This feature has been borrowed from the TWS platform, which served as an inspiration for our work

Objectives

Methods

Results

Discussion

Conclusion