{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Y6EOQ2DHUGASWEFVPE3PVQDTFL","short_pith_number":"pith:Y6EOQ2DH","schema_version":"1.0","canonical_sha256":"c788e86867a1812b10b57936fac0732ac272cfb41e142b6f991f4664f82cbcbc","source":{"kind":"arxiv","id":"2506.16494","version":3},"attestation_state":"computed","paper":{"title":"Manifold Learning for Personalized and Label-Free Detection of Cardiac Arrhythmias","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Nonlinear dimensionality reduction clusters ECG heartbeats from one person into normal and arrhythmic groups without labels or training.","cross_cats":["eess.SP"],"primary_cat":"cs.LG","authors_text":"Amir Reza Vazifeh, Jason W. Fleischer","submitted_at":"2025-06-19T17:39:57Z","abstract_excerpt":"Electrocardiograms (ECGs) provide non-invasive measurements of heart activity and are established tools for detecting cardiac arrhythmias. Although supervised machine learning has emerged as a promising approach for automated heartbeat classification, substantial variations in ECG signals across individuals and leads, combined with inconsistent labeling standards and dataset biases, make it difficult to develop generalizable models. Dimensionality reduction maps high-dimensional data into a lower-dimensional space while preserving the underlying structure, enabling visualization and pattern di"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2506.16494","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-19T17:39:57Z","cross_cats_sorted":["eess.SP"],"title_canon_sha256":"95464f62e213c808d4ab527434a07a5be53a986816aa090265bb2557a161e1df","abstract_canon_sha256":"93a98d1f7fb03b59271f29357f89147ebe6670985c5229d0b9d97308d4ca44d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:46.048134Z","signature_b64":"W2YIsr64NTIuIqMDMtEqfSAognGiCJb6IWay10cazXjdUjqYtSZwSSmE+Fd9k3MBoB2bFPTqTrV5lglHd6/MAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c788e86867a1812b10b57936fac0732ac272cfb41e142b6f991f4664f82cbcbc","last_reissued_at":"2026-06-23T01:12:46.047523Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:46.047523Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Manifold Learning for Personalized and Label-Free Detection of Cardiac Arrhythmias","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Nonlinear dimensionality reduction clusters ECG heartbeats from one person into normal and arrhythmic groups without labels or training.","cross_cats":["eess.SP"],"primary_cat":"cs.LG","authors_text":"Amir Reza Vazifeh, Jason W. Fleischer","submitted_at":"2025-06-19T17:39:57Z","abstract_excerpt":"Electrocardiograms (ECGs) provide non-invasive measurements of heart activity and are established tools for detecting cardiac arrhythmias. Although supervised machine learning has emerged as a promising approach for automated heartbeat classification, substantial variations in ECG signals across individuals and leads, combined with inconsistent labeling standards and dataset biases, make it difficult to develop generalizable models. Dimensionality reduction maps high-dimensional data into a lower-dimensional space while preserving the underlying structure, enabling visualization and pattern di"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"nonlinear dimensionality reduction (NLDR) algorithms, e.g., t-SNE and UMAP, can identify medically relevant features in ECG signals without pretraining or prior information. [...] applying NLDR to heartbeats of a single individual separates normal beats from arrhythmias into distinct clusters, identifiable in an unsupervised manner. Classification on 2D embeddings outperforms the original high-dimensional space, with a k-NN classifier discriminating individual recordings with >=80% accuracy and identifying arrhythmias with median accuracy >=98% and median F1-score >=85%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the distinct clusters formed in the 2D latent space correspond specifically to medically relevant arrhythmia categories rather than other sources of morphological variation or noise in the ECG signals.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Nonlinear dimensionality reduction on ECG signals enables unsupervised personalized arrhythmia detection with high accuracy on 2D embeddings using standard algorithms on the MIT-BIH database.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Nonlinear dimensionality reduction clusters ECG heartbeats from one person into normal and arrhythmic groups without labels or training.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2f66d4320a027521befd5244096df00cb7736de7ea265122e5dd25ed55adedaa"},"source":{"id":"2506.16494","kind":"arxiv","version":3},"verdict":{"id":"b4ed038f-fc91-4fcb-a675-0b9995b0a025","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T08:28:45.957452Z","strongest_claim":"nonlinear dimensionality reduction (NLDR) algorithms, e.g., t-SNE and UMAP, can identify medically relevant features in ECG signals without pretraining or prior information. [...] applying NLDR to heartbeats of a single individual separates normal beats from arrhythmias into distinct clusters, identifiable in an unsupervised manner. Classification on 2D embeddings outperforms the original high-dimensional space, with a k-NN classifier discriminating individual recordings with >=80% accuracy and identifying arrhythmias with median accuracy >=98% and median F1-score >=85%.","one_line_summary":"Nonlinear dimensionality reduction on ECG signals enables unsupervised personalized arrhythmia detection with high accuracy on 2D embeddings using standard algorithms on the MIT-BIH database.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the distinct clusters formed in the 2D latent space correspond specifically to medically relevant arrhythmia categories rather than other sources of morphological variation or noise in the ECG signals.","pith_extraction_headline":"Nonlinear dimensionality reduction clusters ECG heartbeats from one person into normal and arrhythmic groups without labels or training."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.16494/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":57,"sample":[{"doi":"10.5334/gh.1288","year":2024,"title":"The Heart of the World","work_id":"58714330-965c-4d96-ae31-06294da69e96","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/j.ijcard.2024.132584","year":2025,"title":"R. G. Carbone and A.-M. Russell, ”Smoking cessation in heart and chronic respiratory disease: A healthy global strategy,” International Journal of Cardiology, vol. 418, Art no. 132584, 2025. doi: 10.1","work_id":"779e570d-469e-4d59-9460-d5eab84147a5","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"D. Desai and S. Hajouli, ”Arrhythmias,” StatPearls [Internet], Treasure Island (FL): StatPearls Publishing, updated June 5, 2023. Available: https://www.ncbi.nlm.nih.gov/books/NBK558923/","work_id":"59dcbeea-4c4f-46b0-8050-8f5b83d5c7cd","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1161/circulationaha.106.180200","year":2007,"title":"Kligfield et al., ”Recommendations for the Standardization and Interpretation of the Electrocardiogram,”Circulation, vol","work_id":"26f2e612-6541-42a1-84ba-816daea172ce","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/b978-0-12-818946-7.00004-4","year":2020,"title":"Rajoub, ”Machine learning in biomedical signal processing with ECG applications,” inBiomedical Signal Processing and Artificial Intelli- gence in Healthcare, W","work_id":"f35c66bb-3461-4e3b-a4d1-bf9ca34bf469","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":57,"snapshot_sha256":"7c71e75deb5f68462d3aa309bf52f9295487b1ccf6dceed81443e4b01dd1b6c0","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"07f8e509d4b7bde14b725b2cd06866c3755b972420760beecc1d254e7559672a"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2506.16494","created_at":"2026-06-23T01:12:46.047603+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.16494v3","created_at":"2026-06-23T01:12:46.047603+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.16494","created_at":"2026-06-23T01:12:46.047603+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y6EOQ2DHUGAS","created_at":"2026-06-23T01:12:46.047603+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y6EOQ2DHUGASWEFV","created_at":"2026-06-23T01:12:46.047603+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y6EOQ2DH","created_at":"2026-06-23T01:12:46.047603+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.04451","citing_title":"On Out-of-sample Embedding in UMAP","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL","json":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL.json","graph_json":"https://pith.science/api/pith-number/Y6EOQ2DHUGASWEFVPE3PVQDTFL/graph.json","events_json":"https://pith.science/api/pith-number/Y6EOQ2DHUGASWEFVPE3PVQDTFL/events.json","paper":"https://pith.science/paper/Y6EOQ2DH"},"agent_actions":{"view_html":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL","download_json":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL.json","view_paper":"https://pith.science/paper/Y6EOQ2DH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.16494&json=true","fetch_graph":"https://pith.science/api/pith-number/Y6EOQ2DHUGASWEFVPE3PVQDTFL/graph.json","fetch_events":"https://pith.science/api/pith-number/Y6EOQ2DHUGASWEFVPE3PVQDTFL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL/action/storage_attestation","attest_author":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL/action/author_attestation","sign_citation":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL/action/citation_signature","submit_replication":"https://pith.science/pith/Y6EOQ2DHUGASWEFVPE3PVQDTFL/action/replication_record"}},"created_at":"2026-06-23T01:12:46.047603+00:00","updated_at":"2026-06-23T01:12:46.047603+00:00"}