{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T10:13:15Z","timestamp":1788430395029,"version":"build-2803163510"},"reference-count":0,"publisher":"Bentham Science Publishers Ltd.","content-domain":{"domain":["eurekaselect.com"],"crossmark-restriction":true},"short-container-title":["CMIR"],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:sec>\n                    <jats:title>Introduction\/ Objective:<\/jats:title>\n                    <jats:p>MRI-based Alzheimer\u2019s staging is often modeled as separate tasks, although clinical assessment progresses from screening to more refined staging.\nThis study developed a continual-learning framework for retaining earlier diagnostic knowledge.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods:<\/jats:title>\n                    <jats:p>A nested multi-head framework was implemented for sequential learning of related binary classification tasks from structural MRI slices. Because\nverified subject identifiers were unavailable in the Mendeley dataset, the primary experiments were kept at the level of slice-based methodological\nanalysis. Subject-level external validation was additionally carried out on OASIS-VBM with subject-wise three-fold cross-validation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>results:<\/jats:title>\n                    <jats:p>In hold-out testing, area under the curve values were 0.937, 1.000, and 0.998 for the screening, early-stage, and advanced-stage tasks, respectively. The corresponding final accuracies were 89.22%, 98.93%, and 96.71%. Screening performance was higher than that obtained with independent training and sequential training without Elastic Weight Consolidation, while forgetting in the early-stage task decreased from 8.02 to 0.54 percentage points.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results:<\/jats:title>\n                    <jats:p>In the Mendeley slice-level evaluation, the proposed EWC + replay model retained high post-sequence discrimination, with AUC values of 0.976,\n0.985, and 1.000 for T1, T2, and T3. The ECE values were 0.034, 0.057, and 0.187. When the model was tested at the subject level on OASIS,\nAUC values decreased to 0.753, 0.748, and 0.749. This difference shows that the slice-level results should be interpreted cautiously.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion:<\/jats:title>\n                    <jats:p>The findings support the use of the proposed framework for studying sequential retention and forgetting in hierarchical AD staging. However,\nMendeley results are methodological findings at the slice level and should not be interpreted as patient-level clinical performance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion:<\/jats:title>\n                    <jats:p>This study presents a nested continual-learning framework for hierarchical Alzheimer\u2019s disease staging. The framework was tested with ablation,\ncalibration, baseline, and subject-level validation analyses. The results show that it can be used to study retention across related AD staging tasks.\nFuture studies using subject-wise longitudinal cohorts, 3D models, and multimodal data may further improve its clinical relevance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2174\/0115734056512153260820074645","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T09:16:52Z","timestamp":1788427012000},"update-policy":"https:\/\/doi.org\/10.2174\/bsp_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Nested Continual Learning with Elastic Weight Consolidation for Progressive Alzheimer\u2019s Disease Classification from MRI"],"prefix":"10.2174","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8113-0277","authenticated-orcid":true,"given":"Abdurrahim","family":"Akg\u00fcndo\u011fdu","sequence":"first","affiliation":[{"name":"Department of Electrical and Electronics Engineering, Istanbul University-Cerrahpa\u015fa, \u0130stanbul, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6118-9335","authenticated-orcid":true,"given":"\u015eerife","family":"\u00c7eli\u0307kba\u015f","sequence":"additional","affiliation":[{"name":"Biomedical Device Technology, Nam\u0131k Kemal University, Tekirda\u011f, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"965","container-title":["Current Medical Imaging Formerly Current Medical Imaging Reviews"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.eurekaselect.com\/article\/download?doi=10.2174\/0115734056512153260820074645","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.eurekaselect.com\/257957\/article","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.eurekaselect.com\/article\/download?doi=10.2174\/0115734056512153260820074645","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T09:16:57Z","timestamp":1788427017000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.eurekaselect.com\/257957\/article"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,31]]},"references-count":0,"alternative-id":["LiveAll1"],"URL":"https:\/\/doi.org\/10.2174\/0115734056512153260820074645","relation":{},"ISSN":["1573-4056"],"issn-type":[{"value":"1573-4056","type":"print"}],"subject":[],"published":{"date-parts":[[2026,8,31]]},"assertion":[{"value":"Peer Reviewed","order":0,"name":"review_status","label":"Review Status","group":{"name":"peer_review_details","label":"Peer Review Details"}},{"value":"Double blind","order":1,"name":"review_process","label":"Review Process","group":{"name":"peer_review_details","label":"Peer Review Details"}},{"value":"Checked with iThenticate","order":0,"name":"screening_status","label":"Screening Status","group":{"name":"plagiarism_screening","label":"Plagiarism Screening"}},{"value":"2026-05-03","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-25","order":1,"name":"revised","label":"Revised","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-02","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-31","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}