Tammy M Milillo oral presentation (OB3-Thu3-1-2)
Novel image and data fusion method to improve spatial resolution and spectral content of biological images using ToF-SIMS
1 State University of New York, University at Buffalo, 465 Natural Science Complex, NY 14260-3000 Buffalo, United States
2 CUBRC, 4455 Genesee St Ste 106,, NY 14225 Buffalo, United States
Image fusion refers to a class of algorithms that allow images with complementary data and different resolutions to be combined into a single image. The fused image has improved spatial resolution and retains the information from both parent (original) images. Pixel based fusion provides the most detail, with fusion occurring at each pixel.
Segment based fusion uses features, shapes or objects to characterize or group pixels with similar characteristics for fusion. The resulting fused image shows texture and morphology as a result of increased homogeneity in the image. Multiresolution fusion uses feature shapes in the parent image to create objects which are then fused.
Time of Flight Secondary Ion Mass Spectrometry (ToF-SIMS) can detect trace elemental concentrations and ions representative of molecular and polymeric structures producing high spatial resolution images. This capability can be used to gain an understanding of the uptake mechanism and state of silver nanoparticles into Arabdopsis thaliana.
In previous published work, both pixel and segment based image fusion has been reported to successfully combine the chemical information from imaging mass spectrometry and optical spectroscopy with electron and optical microscopy that does not contain chemical information. In addition these techniques have been used to study the uptake mechanism of heavy metals and nanoparticles in soil and the environment. Both types of image fusion techniques have been applied to A. thaliana, to improve the spatial resolution and content of ToF-SIMS images obtained from the two operational modes.
Determining the type of transport (apoplastic or symplastic) within the plant, helps characterize the uptake mechanism of silver nanoparticles from the environment. Multiresolution fusion achieves classification the plant tissue types, providing information about the distribution of silver nanoparticles in the sample.
In the present study a novel algorithm is introduced which increases the spatial resolution of the fused image beyond the two orders of magnitude seen previously. In addition, it decreases edge effects and artifacts. By increasing the accuracy in the classification and identification of tissue types within the sample the spatial distribution of the silver nanoparticles can be observed. The fusion results clearly indicate that silver nanoparticles are concentrated in the plant stem’s phloem, indicating symplastic transport.