Abstract

Mass spectrometry imaging is a powerful tool for investigating the spatial distribution of chemical compounds in a biological sample such as tissue. Two common goals of these experiments are unsupervised segmentation of images into newly discovered homogeneous segments and supervised classification of images into predefined classes. In both cases, the important secondary goals are to characterize the uncertainty associated with the segmentation and with the classification and to characterize the spectral features that define each segment or class. Recent analysis methods have focused on the spatial structure of the data to improve results. However, they either do not address these secondary goals or do this with separate post hoc procedures.We introduce spatial shrunken centroids, a statistical model-based framework for both supervised classification and unsupervised segmentation. It takes as input sets of previously detected, aligned, quantified, and normalized spectral features and expresses both spatial and multivariate nature of the data using probabilistic modeling. It selects informative subsets of spectral features that define each unsupervised segment or supervised class and quantifies and visualizes the uncertainty in spatial segmentations and in tissue classification. In the unsupervised setting, it also guides the choice of an appropriate number of segments. We demonstrate the usefulness of this framework in a supervised human renal cell carcinoma experimental dataset and several unsupervised experimental datasets, including a pig fetus cross-section, three rodent brains, and a controlled image with known ground truth. This framework is available for use within the open-source R package Cardinal as part of a full pipeline for the processing, visualization, and statistical analysis of mass spectrometry imaging experiments.

Highlights

  • We introduce spatial shrunken centroids, a statistical model-based framework for both supervised classification and unsupervised segmentation

  • Brains, and a controlled image with known ground truth. This framework is available for use within the open-source R package Cardinal as part of a full pipeline for the processing, visualization, and statistical analysis of mass spectrometry imaging experiments

  • We evaluate the performance of supervised classification on a human renal cell carcinoma experiment, demonstrating its utility for diagnosing pixels as cancer or normal tissue

Read more

Summary

EXPERIMENTAL PROCEDURES

Unsupervised Segmentation: Pig Fetus Cross-Section—The primary goal of this experiment was to discover morphological features of the pig fetus, such as major organs, through unsupervised analysis of the mass spectra. Mass spectra were collected using a Thermo Finnigan LTQ linear ion trap mass spectrometer with a DESI ion source over the 150 – 1,000 m/z range. The cropped dataset consisted of 4,959 mass spectra with 10,200 spectral features. We will use this dataset to demonstrate unsupervised statistical analysis using all the mass spectral peaks to recover the major morphological features. Unsupervised Segmentation: Cardinal painting with Known Segmentation—The goal of this experiment was to use a controlled sample to evaluate the quality of data acquisition and statistical analysis. The mass spectra were acquired using a Thermo Finnigan LTQ linear ion trap mass spectrometer with a DESI ion source over the 100 –1,000 m/z range. Mass spectra were normalized to a common total ion current, and peak picking was performed to reduce the dataset to 51 peaks.

RESULTS
K Nk kϭ1
DISCUSSION

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.