{"paper":{"title":"Shift- and stretch-invariant non-negative matrix factorization with an application to brain tissue delineation in emission tomography data","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"A frequency-domain non-negative matrix factorization accounts for temporal shifts and stretching to better delineate brain tissue in emission tomography.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anders S. Olsen, Claus Svarer, Gitte M. Knudsen, Jesper L. Hinrich, Miriam L. Navarro, Morten M{\\o}rup","submitted_at":"2026-04-09T12:22:04Z","abstract_excerpt":"Dynamic neuroimaging data, such as emission tomography measurements of radiotracer transport in blood or cerebrospinal fluid, often exhibit diffusion-like properties. These introduce distance-dependent temporal delays, scale-differences, and stretching effects that limit the effectiveness of conventional linear modeling and decomposition methods. To address this, we present the shift- and stretch-invariant non-negative matrix factorization framework. Our approach estimates both integer and non-integer temporal shifts as well as temporal stretching, all implemented in the frequency domain, wher"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We demonstrate on synthetic data and brain emission tomography data that the model is able to account for stretching to provide more detailed characterization of brain tissue structure.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That frequency-domain phase modifications and zero-padding/truncation can accurately recover non-integer shifts and stretches without introducing significant artifacts or violating non-negativity in real emission tomography data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A new NMF variant estimates integer and non-integer temporal shifts plus stretching in the frequency domain to improve brain tissue delineation in emission tomography data.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A frequency-domain non-negative matrix factorization accounts for temporal shifts and stretching to better delineate brain tissue in emission tomography.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5aa382414ce40b6d87d043f6944ad4ec0cb2158e79cdbd4e964117294d870c2e"},"source":{"id":"2604.08161","kind":"arxiv","version":1},"verdict":{"id":"c1e10141-5d6f-4548-8866-91be70b68d57","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T16:51:46.580657Z","strongest_claim":"We demonstrate on synthetic data and brain emission tomography data that the model is able to account for stretching to provide more detailed characterization of brain tissue structure.","one_line_summary":"A new NMF variant estimates integer and non-integer temporal shifts plus stretching in the frequency domain to improve brain tissue delineation in emission tomography data.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That frequency-domain phase modifications and zero-padding/truncation can accurately recover non-integer shifts and stretches without introducing significant artifacts or violating non-negativity in real emission tomography data.","pith_extraction_headline":"A frequency-domain non-negative matrix factorization accounts for temporal shifts and stretching to better delineate brain tissue in emission tomography."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.08161/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":20,"sample":[{"doi":"10.22038/aojnmb.2022.63827.1448","year":2023,"title":"Partial volume effect in SPECT & PET imaging and impact on radionuclide dosimetry es- timates","work_id":"2caa512e-18ce-42ec-933f-06c968b4b9e6","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1007/s002590050425","year":1999,"title":"Automatic seg- mentation of dynamic neuroreceptor single-photon emission tomog- raphy images using fuzzy clustering","work_id":"9a9eac5c-0ac0-439c-918b-00ced1fb36d6","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/b978-012389760-2/50061-","year":1996,"title":"CHAPTER 59 - A Cluster Analysis Approach for the Characteriza- tion of Dynamic PET Data","work_id":"cf726da8-d90d-4f78-b9ba-d9aa7c266036","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1996,"title":"Delineation and quantitation of brain lesions by fuzzy clustering in Positron Emission Tomography,","work_id":"e61c92b6-09c8-48ea-89d5-79dcf5e9a383","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Segmentation of Dynamic Total-Body [18F]-FDG PET Images Using Unsupervised Cluster- ing,","work_id":"c1783015-304b-437b-923a-c57b6f0e23ae","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":20,"snapshot_sha256":"cdc0a348405c620f58a84aae02a5426689f1e01a2869324d4314fe0488a79f7a","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}