{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_52840"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_52840","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"STRUCTURE-AWARE STATISTICAL LEARNING : ORDINAL CHEST RADIOGRAPHY AND MULTI-LAYER COMMUNITY DETECTION","abstract":"Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2025","abstract_html":"Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2025","abstract_has_math":false,"creators":["Liu, Meiqi"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Weng, Haolei","Wang, Ling","Cui, Yuehua","Feng, Yang","Sung, Chih-Li"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T18:59:28Z","subjects":["Statistics"],"languages":["eng"],"rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52840","local:Liu_grad.msu_0128D_21819"],"render_values":[{"text":"etd:52840","href":null,"code":true},{"text":"local:Liu_grad.msu_0128D_21819","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/8xbg-av77","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/8xbg-av77","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_52840","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Weng, Haolei","Wang, Ling","Cui, Yuehua","Feng, Yang","Sung, Chih-Li"]},{"key":"dc:creator","label":"Author","values":["Liu, Meiqi"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2025"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Statistics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial-NoDerivatives 4.0 International"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:52840","local:Liu_grad.msu_0128D_21819","https://doi.org/doi:10.25335/8xbg-av77"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2025","Modern statistical learning requires methods that effectively exploit the inherent structure of complex data. This dissertation presents three contributions that address this need in two domains: ordinal medical image analysis and multi-layer community detection. The common thread is to encode domain structure directly into learning objectives and inference procedures so that the resulting methods are data-efficient, stable, and theoretically grounded.First, we develop a deep learning model with ordinal outcomes for grading pneumoconiosis in chest X-rays, which significantly improves diagnostic accuracy via a hierarchical loss function that explicitly models disease progression. Second, based on a careful asymptotic analysis of spectral embeddings, we introduce a spectral algorithm with adaptive layer aggregation for efficient community detection in multi-layer network data. Third, to further address layer heterogeneity and data contamination in multi-layer community detection, we leverage recent advances in multi-task learning to develop a regularized spectral method with provable guarantees, which not only can adaptively explore the similarity of community structure among related layers but also is robust against contaminated layers. Validated through comprehensive numerical experiments and expert-annotated radiographs in the medical imaging study, these methods deliver effective solutions for ordinal classification and multi-layer community detection. Collectively, the contributions demonstrate how aligning modeling and optimization with domain structure -- ordinality in medical imaging and cross-layer coherence in networks -- yields interpretable, accurate, and robust learning procedures.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (116 p.) : ill."]},{"key":"dc:title","label":"Title","values":["STRUCTURE-AWARE STATISTICAL LEARNING : ORDINAL CHEST RADIOGRAPHY AND MULTI-LAYER COMMUNITY DETECTION"]}]}],"canonical_facts":{"dc:contributor":["Weng, Haolei","Wang, Ling","Cui, Yuehua","Feng, Yang","Sung, Chih-Li"],"dc:creator":["Liu, Meiqi"],"dc:date":["2025"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2025","Modern statistical learning requires methods that effectively exploit the inherent structure of complex data. This dissertation presents three contributions that address this need in two domains: ordinal medical image analysis and multi-layer community detection. The common thread is to encode domain structure directly into learning objectives and inference procedures so that the resulting methods are data-efficient, stable, and theoretically grounded.First, we develop a deep learning model with ordinal outcomes for grading pneumoconiosis in chest X-rays, which significantly improves diagnostic accuracy via a hierarchical loss function that explicitly models disease progression. Second, based on a careful asymptotic analysis of spectral embeddings, we introduce a spectral algorithm with adaptive layer aggregation for efficient community detection in multi-layer network data. Third, to further address layer heterogeneity and data contamination in multi-layer community detection, we leverage recent advances in multi-task learning to develop a regularized spectral method with provable guarantees, which not only can adaptively explore the similarity of community structure among related layers but also is robust against contaminated layers. Validated through comprehensive numerical experiments and expert-annotated radiographs in the medical imaging study, these methods deliver effective solutions for ordinal classification and multi-layer community detection. Collectively, the contributions demonstrate how aligning modeling and optimization with domain structure -- ordinality in medical imaging and cross-layer coherence in networks -- yields interpretable, accurate, and robust learning procedures.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (116 p.) : ill."],"dc:identifier":["etd:52840","local:Liu_grad.msu_0128D_21819","https://doi.org/doi:10.25335/8xbg-av77"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["Attribution-NonCommercial-NoDerivatives 4.0 International"],"dc:subject":["Statistics"],"dc:title":["STRUCTURE-AWARE STATISTICAL LEARNING : ORDINAL CHEST RADIOGRAPHY AND MULTI-LAYER COMMUNITY DETECTION"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:28Z"}