{"id":{"repo_id":"bu","oai_identifier":"oai:open.bu.edu:2144/53409"},"canonical_url":"https://search.ndltd.org/etd/bu/oai:open.bu.edu:2144/53409","repository":{"repo_id":"bu","name":"Boston University","base_url":"https://open.bu.edu/oai/request"},"display":{"title":"Fusion model using unsupervised and self-supervised learning for lung ultrasound video classification of consolidations","abstract":"Lung ultrasound (LUS) has emerged as a promising imaging modality for pediatric pneumonia diagnosis due to its portability and safety. Automating LUS interpretation is challenging due to the variability of the data and limited availability of expert annotations. This thesis explores approaches to automate LUS interpretation with minimal reliance on labeled datasets. First, unsupervised methods using hand-crafted and deep features were explored for clustering frame-level patterns. Building on the insights of this study, the thesis then contributes a fusion-based framework that combines self-supervised contrastive learning with unsupervised anomaly detection. The proposed model learns discriminative representations from unlabeled video frames and integrates them with latent-space deviation scores obtained via dimensionality reduction. This hybrid architecture enables patient-independent, explainable classification of LUS videos. Systematic validation experiments demonstrate that the proposed approach has the potential to support real-world clinical decision-making, especially in resource-limited settings where expert interpretation is scarce.","abstract_html":"Lung ultrasound (LUS) has emerged as a promising imaging modality for pediatric pneumonia diagnosis due to its portability and safety. Automating LUS interpretation is challenging due to the variability of the data and limited availability of expert annotations. This thesis explores approaches to automate LUS interpretation with minimal reliance on labeled datasets. First, unsupervised methods using hand-crafted and deep features were explored for clustering frame-level patterns. Building on the insights of this study, the thesis then contributes a fusion-based framework that combines self-supervised contrastive learning with unsupervised anomaly detection. The proposed model learns discriminative representations from unlabeled video frames and integrates them with latent-space deviation scores obtained via dimensionality reduction. This hybrid architecture enables patient-independent, explainable classification of LUS videos. Systematic validation experiments demonstrate that the proposed approach has the potential to support real-world clinical decision-making, especially in resource-limited settings where expert interpretation is scarce.","abstract_has_math":false,"creators":["Bhattacharjee, Saunak"],"institution":"Boston University","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":[],"advisors":["Betke, Margrit"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-11-18T20:25:14Z","subjects":["Artificial intelligence","Medical imaging","Lung ultrasound","Self-supervised learning","Unsupervised learning","Video frame classification"],"languages":["en_US"],"rights":["Attribution-NonCommercial 4.0 International"],"rights_urls":["http://creativecommons.org/licenses/by-nc/4.0/"],"identifier_entries":[],"person_identifier_entries":[{"key":"dc:identifier.orcid","label":"ORCID","values":["0009-0004-4233-5156"],"render_values":[{"text":"0009-0004-4233-5156","href":"https://orcid.org/0009-0004-4233-5156","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/2144/53409","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"https://hdl.handle.net/2144/53409","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://open.bu.edu/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aopen.bu.edu%3A2144%2F53409","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Betke, Margrit"]},{"key":"dc:creator","label":"Author","values":["Bhattacharjee, Saunak"]},{"key":"dc:identifier.orcid","label":"ORCID","values":["0009-0004-4233-5156"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-07-30T18:11:42Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-07-30T18:11:42Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:date.updated","label":"Dc Date Updated","values":["2026-07-22T16:15:46Z"]},{"key":"dc:type","label":"Dc Type","values":["Thesis/Dissertation"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Boston University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial intelligence","Medical imaging","Lung ultrasound","Self-supervised learning","Unsupervised learning","Video frame classification"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en_US"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution-NonCommercial 4.0 International"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://creativecommons.org/licenses/by-nc/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/2144/53409"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["2025"]},{"key":"dc:description.abstract","label":"Abstract","values":["Lung ultrasound (LUS) has emerged as a promising imaging modality for pediatric pneumonia diagnosis due to its portability and safety. Automating LUS interpretation is challenging due to the variability of the data and limited availability of expert annotations. This thesis explores approaches to automate LUS interpretation with minimal reliance on labeled datasets. First, unsupervised methods using hand-crafted and deep features were explored for clustering frame-level patterns. Building on the insights of this study, the thesis then contributes a fusion-based framework that combines self-supervised contrastive learning with unsupervised anomaly detection. The proposed model learns discriminative representations from unlabeled video frames and integrates them with latent-space deviation scores obtained via dimensionality reduction. This hybrid architecture enables patient-independent, explainable classification of LUS videos. Systematic validation experiments demonstrate that the proposed approach has the potential to support real-world clinical decision-making, especially in resource-limited settings where expert interpretation is scarce."]},{"key":"dc:title","label":"Title","values":["Fusion model using unsupervised and self-supervised learning for lung ultrasound video classification of consolidations"]}]}],"canonical_facts":{"dc:contributor.advisor":["Betke, Margrit"],"dc:creator":["Bhattacharjee, Saunak"],"dc:date.accessioned":["2026-07-30T18:11:42Z"],"dc:date.available":["2026-07-30T18:11:42Z"],"dc:date.issued":["2025"],"dc:date.updated":["2026-07-22T16:15:46Z"],"dc:description":["2025"],"dc:description.abstract":["Lung ultrasound (LUS) has emerged as a promising imaging modality for pediatric pneumonia diagnosis due to its portability and safety. Automating LUS interpretation is challenging due to the variability of the data and limited availability of expert annotations. This thesis explores approaches to automate LUS interpretation with minimal reliance on labeled datasets. First, unsupervised methods using hand-crafted and deep features were explored for clustering frame-level patterns. Building on the insights of this study, the thesis then contributes a fusion-based framework that combines self-supervised contrastive learning with unsupervised anomaly detection. The proposed model learns discriminative representations from unlabeled video frames and integrates them with latent-space deviation scores obtained via dimensionality reduction. This hybrid architecture enables patient-independent, explainable classification of LUS videos. Systematic validation experiments demonstrate that the proposed approach has the potential to support real-world clinical decision-making, especially in resource-limited settings where expert interpretation is scarce."],"dc:identifier.orcid":["0009-0004-4233-5156"],"dc:identifier.uri":["https://hdl.handle.net/2144/53409"],"dc:language.iso":["en_US"],"dc:rights":["Attribution-NonCommercial 4.0 International"],"dc:rights.uri":["http://creativecommons.org/licenses/by-nc/4.0/"],"dc:subject":["Artificial intelligence","Medical imaging","Lung ultrasound","Self-supervised learning","Unsupervised learning","Video frame classification"],"dc:title":["Fusion model using unsupervised and self-supervised learning for lung ultrasound video classification of consolidations"],"dc:type":["Thesis/Dissertation"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Boston University"]},"updated_at":"2026-11-18T20:25:14Z"}