{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3MPJKXBSZD3B5R5JQMEANR4BX4","short_pith_number":"pith:3MPJKXBS","schema_version":"1.0","canonical_sha256":"db1e955c32c8f61ec7a9830806c781bf3a970b5825516d858eb452e863414bc3","source":{"kind":"arxiv","id":"2411.02372","version":2},"attestation_state":"computed","paper":{"title":"Learning General-Purpose Biomedical Volume Representations using Randomized Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Adrian V. Dalca, Benjamin Billot, Clinton J. Wang, Hallee E. Wong, Mengwei Ren, Neel Dey, P. Ellen Grant, Polina Golland","submitted_at":"2024-11-04T18:40:46Z","abstract_excerpt":"Current volumetric biomedical foundation models struggle to generalize as public 3D datasets are small and do not cover the broad diversity of medical procedures, conditions, anatomical regions, and imaging protocols. We address this by creating a representation learning method that instead anticipates strong domain shifts at training time itself. We first propose a data engine that synthesizes highly variable training samples that would enable generalization to new biomedical contexts. To then train a single 3D network for any voxel-level task, we develop a contrastive learning method that pr"},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2411.02372","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-11-04T18:40:46Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"cbcc7c56a26e72b395ce556b8e56d81e2354759de0d567ed90816556eb9f212d","abstract_canon_sha256":"d973aa386254b3d9558f18488da123f9ce6df2f1271a1f5be13752eaaaf8f82d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:22:09.376645Z","signature_b64":"KRzYDTevqS9hwmJzdmh6WSdB+sbD7S17g555fsNQ/Z5ANpSuGBC2U/Rp3Nr1Lujo6x+PlQcINOekcAArSs2zBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db1e955c32c8f61ec7a9830806c781bf3a970b5825516d858eb452e863414bc3","last_reissued_at":"2026-07-05T10:22:09.376051Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:22:09.376051Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning General-Purpose Biomedical Volume Representations using Randomized Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Adrian V. Dalca, Benjamin Billot, Clinton J. Wang, Hallee E. Wong, Mengwei Ren, Neel Dey, P. Ellen Grant, Polina Golland","submitted_at":"2024-11-04T18:40:46Z","abstract_excerpt":"Current volumetric biomedical foundation models struggle to generalize as public 3D datasets are small and do not cover the broad diversity of medical procedures, conditions, anatomical regions, and imaging protocols. We address this by creating a representation learning method that instead anticipates strong domain shifts at training time itself. We first propose a data engine that synthesizes highly variable training samples that would enable generalization to new biomedical contexts. To then train a single 3D network for any voxel-level task, we develop a contrastive learning method that pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.02372","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2411.02372/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2411.02372","created_at":"2026-07-05T10:22:09.376115+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.02372v2","created_at":"2026-07-05T10:22:09.376115+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.02372","created_at":"2026-07-05T10:22:09.376115+00:00"},{"alias_kind":"pith_short_12","alias_value":"3MPJKXBSZD3B","created_at":"2026-07-05T10:22:09.376115+00:00"},{"alias_kind":"pith_short_16","alias_value":"3MPJKXBSZD3B5R5J","created_at":"2026-07-05T10:22:09.376115+00:00"},{"alias_kind":"pith_short_8","alias_value":"3MPJKXBS","created_at":"2026-07-05T10:22:09.376115+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.22226","citing_title":"Cardiovascular disease classification using radiomics and geometric features from cardiac CT","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13397","citing_title":"A Multimodal Clinically Informed Coarse-to-Fine Framework for Longitudinal CT Registration in Proton Therapy","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4","json":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4.json","graph_json":"https://pith.science/api/pith-number/3MPJKXBSZD3B5R5JQMEANR4BX4/graph.json","events_json":"https://pith.science/api/pith-number/3MPJKXBSZD3B5R5JQMEANR4BX4/events.json","paper":"https://pith.science/paper/3MPJKXBS"},"agent_actions":{"view_html":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4","download_json":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4.json","view_paper":"https://pith.science/paper/3MPJKXBS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.02372&json=true","fetch_graph":"https://pith.science/api/pith-number/3MPJKXBSZD3B5R5JQMEANR4BX4/graph.json","fetch_events":"https://pith.science/api/pith-number/3MPJKXBSZD3B5R5JQMEANR4BX4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4/action/storage_attestation","attest_author":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4/action/author_attestation","sign_citation":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4/action/citation_signature","submit_replication":"https://pith.science/pith/3MPJKXBSZD3B5R5JQMEANR4BX4/action/replication_record"}},"created_at":"2026-07-05T10:22:09.376115+00:00","updated_at":"2026-07-05T10:22:09.376115+00:00"}