Improving the Image Quality of Time-of-flight Secondary Ion Mass spectrometry Images

Open Access
Author:
Tarolli, Jay Gage
Graduate Program:
Chemistry
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
August 24, 2015
Committee Members:
  • Nicholas Winograd, Dissertation Advisor
  • Barbara Jane Garrison, Committee Member
  • Miriam Arak Freedman, Committee Member
  • James Z Wang, Committee Chair
Keywords:
  • Image Fusion
  • Secondary Ion Mass Spectrometry
  • Data Analysis
  • Data Fusion
Abstract:
As the boundaries of secondary ion mass spectrometry (SIMS) are pushed to chemically image biological systems with even greater spatial resolution, an inherent lack of information to analyze becomes more pronounced. A smaller sampling volume, coupled with the desire for molecule specific information, limits the detection sensitivity in these complex systems. To improve the visual quality of biological SIMS images, information from higher resolution imaging sources is combined with the chemical information using image fusion. Pan-sharpening, a subset of image fusion, was adapted to combine SIMS images with secondary electron microscopy (SEM), optical microscopy, and fluorescence microscopy images to improve the spatial resolution of the images. A synthetic model data set and experimentally obtained SIMS images of copper mesh grids prove the efficacy of the pan-sharpening algorithm for fusion with an SEM image. Colonies of algal cells were imaged with SIMS and fused with SEM images to determine the distributions of a wax monoester outside of the colonies and hydrocarbon containing oil bodies within the colonies with greater detail than possible before. The desire to implement image fusion as a universal data processing technique for SIMS imaging requires registration of the two images before performing pan-sharpening. To accomplish this, the Insight Segmentation and Registration Toolkit (ITK) was implemented for registering a pair of images acquired with two separate imaging modalities. Optical microscopy and fluorescence microscopy images were registered and fused with SIMS images to demonstrate the applicability of image fusion using virtually any conceivable source of higher resolution data.