Mass spectrometry (MS) imaging is a powerful investigation technique for a wide range of biological applications such as molecular histology of tissue, whole body sections, and bacterial films , and biomedical applications such as cancer diagnosis. MS imaging visualizes the spatial distribution of molecular ions in a sample by repeatedly collecting mass spectra across its surface, resulting in complex, high-dimensional imaging datasets. Two of the primary goals of statistical analysis of MS imaging experiments are classification (for supervised experiments), i.e. assigning pixels to pre-defined classes based on their spectral profiles, and segmentation (for unsupervised experiments), i.e. assigning pixels to newly discovered segments with relatively homogenous and distinct spectral profiles. To accomplish these goals, this research provides both statistical methods and statistical computing tools. First, we propose a novel spatial shrunken centroids framework for performing classification and segmentation of MS imaging experiments with feature selection. Spatial shrunken centroids combines spatial smoothing with statistical regularization in a model-based framework appropriate for both supervised and unsupervised settings. Second, we provide Cardinal, a free and open-source R package for processing, visualization, and statistical analysis of MS imaging experiments. Cardinal is the first R package designed specifically for MS imaging, and the first software for MS imaging that focuses specifically on experiments and statistical analysis. In addition to providing tools for statistical analysis, it also provides infrastructure to enable other statisticians to more easily develop new methods for MS imaging experiments. Lastly, to enable scalability of Cardinal to larger-than-memory datasets, we provide matter, a free and open-source R package for statistical computing with structured datasets-on-disk, such as MS imaging data files. Together, spatial shrunken centroids, Cardinal, and matter aim to allow scalable statistical analysis for high-resolution, high-throughput MS imaging experiments.
","abstract_html":"<p>Mass spectrometry (MS) imaging is a powerful investigation technique for a wide range of biological applications such as molecular histology of tissue, whole body sections, and bacterial films , and biomedical applications such as cancer diagnosis. MS imaging visualizes the spatial distribution of molecular ions in a sample by repeatedly collecting mass spectra across its surface, resulting in complex, high-dimensional imaging datasets. Two of the primary goals of statistical analysis of MS imaging experiments are classification (for supervised experiments), i.e. assigning pixels to pre-defined classes based on their spectral profiles, and segmentation (for unsupervised experiments), i.e. assigning pixels to newly discovered segments with relatively homogenous and distinct spectral profiles. To accomplish these goals, this research provides both statistical methods and statistical computing tools. First, we propose a novel spatial shrunken centroids framework for performing classification and segmentation of MS imaging experiments with feature selection. Spatial shrunken centroids combines spatial smoothing with statistical regularization in a model-based framework appropriate for both supervised and unsupervised settings. Second, we provide <em>Cardinal</em>, a free and open-source R package for processing, visualization, and statistical analysis of MS imaging experiments. <em>Cardinal</em> is the first R package designed specifically for MS imaging, and the first software for MS imaging that focuses specifically on experiments and statistical analysis. In addition to providing tools for statistical analysis, it also provides infrastructure to enable other statisticians to more easily develop new methods for MS imaging experiments. Lastly, to enable scalability of <em>Cardinal</em> to larger-than-memory datasets, we provide <em>matter</em>, a free and open-source R package for statistical computing with structured datasets-on-disk, such as MS imaging data files. Together, spatial shrunken centroids, <em>Cardinal</em>, and <em>matter</em> aim to allow scalable statistical analysis for high-resolution, high-throughput MS imaging experiments.</p>","abstract_has_math":false,"creators":["Bemis, Kyle"],"institution":null,"degree_name":"Doctor of Philosophy (PhD)","degree_level":"Dissertation","degree_discipline":"Statistics","degree_department":null,"school":null,"contributors":["Olga Vitek","Hyonho Chun","R. Graham Cooks","Hao Zhang"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-12-01T08:00:00Z","date_published":"2016-12-01T08:00:00Z","updated_at":"2026-07-24T03:54:09Z","subjects":["Pure sciences","Biological sciences","Applied sciences","Cardinal","Classification","Clustering","Imaging","Mass spectrometry","Software","Bioinformatics","Computer Sciences","Statistics and Probability"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://docs.lib.purdue.edu/open_access_dissertations/905","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Olga Vitek","Hyonho Chun","R. Graham Cooks","Hao Zhang"]},{"key":"dc:creator","label":"Author","values":["Bemis, Kyle"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"thesis:degree_discipline","label":"Discipline","values":["Statistics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Pure sciences","Biological sciences","Applied sciences","Cardinal","Classification","Clustering","Imaging","Mass spectrometry","Software","Bioinformatics","Computer Sciences","Statistics and Probability"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://docs.lib.purdue.edu/open_access_dissertations/905"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Mass spectrometry (MS) imaging is a powerful investigation technique for a wide range of biological applications such as molecular histology of tissue, whole body sections, and bacterial films , and biomedical applications such as cancer diagnosis. MS imaging visualizes the spatial distribution of molecular ions in a sample by repeatedly collecting mass spectra across its surface, resulting in complex, high-dimensional imaging datasets. Two of the primary goals of statistical analysis of MS imaging experiments are classification (for supervised experiments), i.e. assigning pixels to pre-defined classes based on their spectral profiles, and segmentation (for unsupervised experiments), i.e. assigning pixels to newly discovered segments with relatively homogenous and distinct spectral profiles. To accomplish these goals, this research provides both statistical methods and statistical computing tools. First, we propose a novel spatial shrunken centroids framework for performing classification and segmentation of MS imaging experiments with feature selection. Spatial shrunken centroids combines spatial smoothing with statistical regularization in a model-based framework appropriate for both supervised and unsupervised settings. Second, we provide Cardinal, a free and open-source R package for processing, visualization, and statistical analysis of MS imaging experiments. Cardinal is the first R package designed specifically for MS imaging, and the first software for MS imaging that focuses specifically on experiments and statistical analysis. In addition to providing tools for statistical analysis, it also provides infrastructure to enable other statisticians to more easily develop new methods for MS imaging experiments. Lastly, to enable scalability of Cardinal to larger-than-memory datasets, we provide matter, a free and open-source R package for statistical computing with structured datasets-on-disk, such as MS imaging data files. Together, spatial shrunken centroids, Cardinal, and matter aim to allow scalable statistical analysis for high-resolution, high-throughput MS imaging experiments.
"]},{"key":"dc:title","label":"Title","values":["A framework for the statistical analysis of mass spectrometry imaging experiments"]}]}],"canonical_facts":{"dc:contributor":["Olga Vitek","Hyonho Chun","R. Graham Cooks","Hao Zhang"],"dc:creator":["Bemis, Kyle"],"dc:description.abstract":["Mass spectrometry (MS) imaging is a powerful investigation technique for a wide range of biological applications such as molecular histology of tissue, whole body sections, and bacterial films , and biomedical applications such as cancer diagnosis. MS imaging visualizes the spatial distribution of molecular ions in a sample by repeatedly collecting mass spectra across its surface, resulting in complex, high-dimensional imaging datasets. Two of the primary goals of statistical analysis of MS imaging experiments are classification (for supervised experiments), i.e. assigning pixels to pre-defined classes based on their spectral profiles, and segmentation (for unsupervised experiments), i.e. assigning pixels to newly discovered segments with relatively homogenous and distinct spectral profiles. To accomplish these goals, this research provides both statistical methods and statistical computing tools. First, we propose a novel spatial shrunken centroids framework for performing classification and segmentation of MS imaging experiments with feature selection. Spatial shrunken centroids combines spatial smoothing with statistical regularization in a model-based framework appropriate for both supervised and unsupervised settings. Second, we provide Cardinal, a free and open-source R package for processing, visualization, and statistical analysis of MS imaging experiments. Cardinal is the first R package designed specifically for MS imaging, and the first software for MS imaging that focuses specifically on experiments and statistical analysis. In addition to providing tools for statistical analysis, it also provides infrastructure to enable other statisticians to more easily develop new methods for MS imaging experiments. Lastly, to enable scalability of Cardinal to larger-than-memory datasets, we provide matter, a free and open-source R package for statistical computing with structured datasets-on-disk, such as MS imaging data files. Together, spatial shrunken centroids, Cardinal, and matter aim to allow scalable statistical analysis for high-resolution, high-throughput MS imaging experiments.
"],"dc:identifier":["https://docs.lib.purdue.edu/open_access_dissertations/905"],"dc:subject":["Pure sciences","Biological sciences","Applied sciences","Cardinal","Classification","Clustering","Imaging","Mass spectrometry","Software","Bioinformatics","Computer Sciences","Statistics and Probability"],"dc:title":["A framework for the statistical analysis of mass spectrometry imaging experiments"],"thesis:degree_discipline":["Statistics"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Doctor of Philosophy (PhD)"]},"updated_at":"2026-07-24T03:54:09Z"}