{"id":{"repo_id":"buffalo","oai_identifier":"oai:ubir.buffalo.edu:10477/83832"},"canonical_url":"https://search.dev.ndltd.org/etd/buffalo/oai:ubir.buffalo.edu:10477/83832","repository":{"repo_id":"buffalo","name":"Buffalo","base_url":"https://ubir.buffalo.edu/oai/request"},"display":{"title":"Multi-Scale Deep Learning for Small Object Detection in Whole Slide H&E Stained Images: Tumor Bud Detection in Colorectal Cancer","abstract":"M.S.","abstract_html":"M.S.","abstract_has_math":false,"creators":["Sun, Chen-Yu; 0000-0002-5155-9555"],"institution":"State University of New York at Buffalo","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Doyle, Scott","Biomedical Engineering"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-06-17T19:54:32Z","date_published":"2022-06-17T19:54:32Z","updated_at":"2026-07-27T19:05:28Z","subjects":["biomedical engineering","pathology","bioinformatics"],"languages":["eng"],"rights":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/10477/83832","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Doyle, Scott","Biomedical Engineering"]},{"key":"dc:creator","label":"Author","values":["Sun, Chen-Yu; 0000-0002-5155-9555"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-06-17T19:54:32Z","2020"]},{"key":"dc:publisher","label":"Institution","values":["State University of New York at Buffalo"]},{"key":"dc:type","label":"Dc Type","values":["Text","Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["biomedical engineering","pathology","bioinformatics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/10477/83832"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["M.S.","Whole slide images of hematoxylin and eosin (H&E) stained tissues are widely used in clinical diagnoses as they are inexpensive and easier to obtain. Pathologists diagnosing Colorectal Cancer (CRC) use them to diagnose tumor progression by looking for tumor buds, which are a small group of tumor cells no more than five that exist in the invasive tumor front. Since such small objects cannot be spot at low magnification scale, the pathologists look for the invasive tumor fronts which are considered the ‘hotspots’, then further examine the regions at high magnification scale to manually count the tumor buds. This is a laborious process and is subject to the experience level of the observer. We aim to build a deep convolutional neural network (CNN) system that automates this process to provide more efficient and consistent detection results. The multi-tiered classifier is inspired by the workflow of pathologists that takes steps to target different biological characteristics at different magnification scale. This work is applicable to general small object detection in whole slide histological images, such as mitosis detection in breast cancer tissue or tumor satellite detection in oral cavity cancer.","**To request an accessible version of the file(s) associated with this item, contact library@buffalo.edu. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.**"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Multi-Scale Deep Learning for Small Object Detection in Whole Slide H&E Stained Images: Tumor Bud Detection in Colorectal Cancer"]}]}],"canonical_facts":{"dc:contributor":["Doyle, Scott","Biomedical Engineering"],"dc:creator":["Sun, Chen-Yu; 0000-0002-5155-9555"],"dc:date":["2022-06-17T19:54:32Z","2020"],"dc:description":["M.S.","Whole slide images of hematoxylin and eosin (H&E) stained tissues are widely used in clinical diagnoses as they are inexpensive and easier to obtain. Pathologists diagnosing Colorectal Cancer (CRC) use them to diagnose tumor progression by looking for tumor buds, which are a small group of tumor cells no more than five that exist in the invasive tumor front. Since such small objects cannot be spot at low magnification scale, the pathologists look for the invasive tumor fronts which are considered the ‘hotspots’, then further examine the regions at high magnification scale to manually count the tumor buds. This is a laborious process and is subject to the experience level of the observer. We aim to build a deep convolutional neural network (CNN) system that automates this process to provide more efficient and consistent detection results. The multi-tiered classifier is inspired by the workflow of pathologists that takes steps to target different biological characteristics at different magnification scale. This work is applicable to general small object detection in whole slide histological images, such as mitosis detection in breast cancer tissue or tumor satellite detection in oral cavity cancer.","**To request an accessible version of the file(s) associated with this item, contact library@buffalo.edu. Please include the item's persistent URL [http://hdl.handle.net/. . .] in your request.**"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/10477/83832"],"dc:language":["eng"],"dc:publisher":["State University of New York at Buffalo"],"dc:rights":["Users of works found in University at Buffalo Institutional Repository (UBIR) are responsible for identifying and contacting the copyright owner for permission to reuse. University at Buffalo Libraries do not manage rights for copyright-protected works and cannot assist with permissions.","Copyright retained by author."],"dc:subject":["biomedical engineering","pathology","bioinformatics"],"dc:title":["Multi-Scale Deep Learning for Small Object Detection in Whole Slide H&E Stained Images: Tumor Bud Detection in Colorectal Cancer"],"dc:type":["Text","Thesis"]},"updated_at":"2026-07-27T19:05:28Z"}