Image descriptor Tutorial

Image descriptor calculation web tool

Authors: Philip Doganis, Marios Kotsiandris, NTUA

Introduction

The images taken by Electron or Fluorescence microscopy are valuable sources of information that depict the actual structure and shape of the nanomaterial. To this end, we introduce a web tool, which provides the user with a systematic framework for the automated analysis of microscopy images of nanomaterials and the calculation of nanoparticle descriptors. The term Descriptors or factors has been adopted to refer to significant properties of nanoparticles, as they describe not only the morphology of the nanoparticle but also its behavior. The only data someone can extract from digital images are relevant to the shape, size and color of the nanoparticle. Nanoparticles come in a variety of sizes and shapes, ranging from simple to highly complex. As such attributes may have great influence on their functionality, some characteristic measurements of nanoparticle size should be defined in order to achieve proper characterization of size and shape.

The scope of this tutorial lies mainly in providing a guide to members of the nanotoxicology community that wish to derive nanomaterial descriptors from microscopy images of nanomaterials using the image descriptor calculation web tool presented here.

The web application has been developed in Java and is based on the open source ImageJ software. Its input consists of microscopy nanomaterial images, as well as user-­‐defined parameters regarding the choice of applied filter and descriptors to be calculated. The output comprises a table with the calculated descriptor values for each particle and the average value for all particles of the image, which can be produced in various easily transferable formats. It should be noted here that image descriptor calculations will also be made available as RESTful web services that can be accessed from systems of any architecture that follows the tool’s RESTful API. The user interface is hosted in the following address: http://app.jaqpot.org:8880/imageAnalysis/.The source code can be found at: https://github.com/enanomapper/imageAnalysis.

For the sake of brevity, technical details on the programming and additional remarks on the functionality of the tool were omitted but will be included in Deliverable 4.2 of eNanoMapper (which will be made publicly available at http://enanomapper.net/), alongside with a further analysis of features.

Currently, the applicability of the tool covers gold and metal oxide nanoparticles, but the application can be easily extended to cover other types of nanomaterials, such as carbon nanotubes. The tool is released with a user interface and has been made available as a web service, thus allowing its integration into larger applications, such as the eNanoMapper Computational infrastructure, and facilitating the utilization of the produced descriptors for nanoQSAR modelling.

The descriptors calculated will be described next, followed by a presentation of the application workflow.

You can download the tutorial >> here <<