{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:RK65BUHBWYRM2FV2UMFWTPJALK","short_pith_number":"pith:RK65BUHB","canonical_record":{"source":{"id":"2604.27277","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T00:21:36Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"1aab93bb7b167328414952a51b7440d16f473adc1be657eeb9899ee952a1b134","abstract_canon_sha256":"4130b87d6702e492ed571669aaa8ca08a2bd068dbecd7543f43005857d7f6755"},"schema_version":"1.0"},"canonical_sha256":"8abdd0d0e1b622cd16baa30b69bd205ab6333733f94396f437f0d90ddce06261","source":{"kind":"arxiv","id":"2604.27277","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.27277","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"arxiv_version","alias_value":"2604.27277v2","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.27277","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_12","alias_value":"RK65BUHBWYRM","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_16","alias_value":"RK65BUHBWYRM2FV2","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_8","alias_value":"RK65BUHB","created_at":"2026-05-27T01:04:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:RK65BUHBWYRM2FV2UMFWTPJALK","target":"record","payload":{"canonical_record":{"source":{"id":"2604.27277","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T00:21:36Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"1aab93bb7b167328414952a51b7440d16f473adc1be657eeb9899ee952a1b134","abstract_canon_sha256":"4130b87d6702e492ed571669aaa8ca08a2bd068dbecd7543f43005857d7f6755"},"schema_version":"1.0"},"canonical_sha256":"8abdd0d0e1b622cd16baa30b69bd205ab6333733f94396f437f0d90ddce06261","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T01:04:58.559262Z","signature_b64":"giLkK4YD9WD/5FFF0G0vhj5XoJM5fsZOeaIluFQmL9Kw1CheoRCDf9KyQN+1dnkFYsN88C7y4trAPTtKdA0YAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8abdd0d0e1b622cd16baa30b69bd205ab6333733f94396f437f0d90ddce06261","last_reissued_at":"2026-05-27T01:04:58.558600Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T01:04:58.558600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.27277","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-27T01:04:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lJbj6EU4kxi/XhETIFMxqH/EtPZnuLz71MECjYeOk9iq8eVhOdtTYXRh3ZVYhJeE5abHu5QBhZTSQLCsf/OeDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T10:42:54.655953Z"},"content_sha256":"dd315b20854524eafa4330eb898a6379f42726fe355d5720ec4b951cb11fe2b3","schema_version":"1.0","event_id":"sha256:dd315b20854524eafa4330eb898a6379f42726fe355d5720ec4b951cb11fe2b3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:RK65BUHBWYRM2FV2UMFWTPJALK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A self-supervised model trained on millions of brain MRI slices yields a unified representation that supports diverse clinical tasks with a frozen encoder and lightweight heads.","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Chih-Wei Chang, Harini Veeraraghavan, Mingzhe Hu, Mojtaba Safari, Shansong Wang, Xiaofeng Yang, Yizhou Wu, Yuheng Li","submitted_at":"2026-04-30T00:21:36Z","abstract_excerpt":"Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data. Here we show that a single self-supervised representation can generalize across heterogeneous brain MRI endpoints. We trained BrainDINO, a self-distilled foundation model, on approximately 6.6 million unlabeled axial slices from 20 datasets encompassing broad variation in population, disease, and acquisition setting. Using a frozen encoder with lightweight task heads, BrainDINO supported transfer across tumor segmentation, neu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"a single self-supervised representation can generalize across heterogeneous brain MRI endpoints... large-scale slice-wise self-supervised learning can yield a unified brain MRI representation that supports diverse neuroimaging tasks without volumetric pretraining or full-network fine-tuning","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the 20 datasets and the selected clinical endpoints (tumor segmentation, classification tasks, survival modeling, etc.) are sufficiently representative to support the claim of broad generalizability across all brain MRI applications and acquisition settings.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"BrainDINO delivers a single self-supervised brain MRI representation that generalizes to tumor segmentation, disease classification, brain age estimation, and other tasks without volumetric pretraining or full fine-tuning.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A self-supervised model trained on millions of brain MRI slices yields a unified representation that supports diverse clinical tasks with a frozen encoder and lightweight heads.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"928814841c8a8fac75677054addbb3972973b8284ddaf5129cdcb2267cdf3584"},"source":{"id":"2604.27277","kind":"arxiv","version":2},"verdict":{"id":"8b44807d-a0d0-474a-9e66-5f413dcb57a7","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T08:42:33.520839Z","strongest_claim":"a single self-supervised representation can generalize across heterogeneous brain MRI endpoints... large-scale slice-wise self-supervised learning can yield a unified brain MRI representation that supports diverse neuroimaging tasks without volumetric pretraining or full-network fine-tuning","one_line_summary":"BrainDINO delivers a single self-supervised brain MRI representation that generalizes to tumor segmentation, disease classification, brain age estimation, and other tasks without volumetric pretraining or full fine-tuning.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the 20 datasets and the selected clinical endpoints (tumor segmentation, classification tasks, survival modeling, etc.) are sufficiently representative to support the claim of broad generalizability across all brain MRI applications and acquisition settings.","pith_extraction_headline":"A self-supervised model trained on millions of brain MRI slices yields a unified representation that supports diverse clinical tasks with a frozen encoder and lightweight heads."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.27277/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T22:40:39.335488Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T19:25:04.198889Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"902de55835a706d711b7c2551c58a30a0fcea3f271298a36e0029fbf8763526b"},"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"},"verdict_id":"8b44807d-a0d0-474a-9e66-5f413dcb57a7"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-27T01:04:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0xp8mjY1Xq02yyjLtOS5g3wTT1vqbpJ3DBvMA2LA6FVJfA+Er7asvcm4yQW8yrtZ1JDoJW6JqH4jiyegwWKyCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T10:42:54.656449Z"},"content_sha256":"8226b5218bb891182e0bf95fad36ee2e4e2b68b33f64e51b25b4fb7159023af2","schema_version":"1.0","event_id":"sha256:8226b5218bb891182e0bf95fad36ee2e4e2b68b33f64e51b25b4fb7159023af2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RK65BUHBWYRM2FV2UMFWTPJALK/bundle.json","state_url":"https://pith.science/pith/RK65BUHBWYRM2FV2UMFWTPJALK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RK65BUHBWYRM2FV2UMFWTPJALK/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-29T10:42:54Z","links":{"resolver":"https://pith.science/pith/RK65BUHBWYRM2FV2UMFWTPJALK","bundle":"https://pith.science/pith/RK65BUHBWYRM2FV2UMFWTPJALK/bundle.json","state":"https://pith.science/pith/RK65BUHBWYRM2FV2UMFWTPJALK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RK65BUHBWYRM2FV2UMFWTPJALK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RK65BUHBWYRM2FV2UMFWTPJALK","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"4130b87d6702e492ed571669aaa8ca08a2bd068dbecd7543f43005857d7f6755","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T00:21:36Z","title_canon_sha256":"1aab93bb7b167328414952a51b7440d16f473adc1be657eeb9899ee952a1b134"},"schema_version":"1.0","source":{"id":"2604.27277","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.27277","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"arxiv_version","alias_value":"2604.27277v2","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.27277","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_12","alias_value":"RK65BUHBWYRM","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_16","alias_value":"RK65BUHBWYRM2FV2","created_at":"2026-05-27T01:04:58Z"},{"alias_kind":"pith_short_8","alias_value":"RK65BUHB","created_at":"2026-05-27T01:04:58Z"}],"graph_snapshots":[{"event_id":"sha256:8226b5218bb891182e0bf95fad36ee2e4e2b68b33f64e51b25b4fb7159023af2","target":"graph","created_at":"2026-05-27T01:04:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"a single self-supervised representation can generalize across heterogeneous brain MRI endpoints... large-scale slice-wise self-supervised learning can yield a unified brain MRI representation that supports diverse neuroimaging tasks without volumetric pretraining or full-network fine-tuning"},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the 20 datasets and the selected clinical endpoints (tumor segmentation, classification tasks, survival modeling, etc.) are sufficiently representative to support the claim of broad generalizability across all brain MRI applications and acquisition settings."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"BrainDINO delivers a single self-supervised brain MRI representation that generalizes to tumor segmentation, disease classification, brain age estimation, and other tasks without volumetric pretraining or full fine-tuning."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A self-supervised model trained on millions of brain MRI slices yields a unified representation that supports diverse clinical tasks with a frozen encoder and lightweight heads."}],"snapshot_sha256":"928814841c8a8fac75677054addbb3972973b8284ddaf5129cdcb2267cdf3584"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T22:40:39.335488Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T19:25:04.198889Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.27277/integrity.json","findings":[],"snapshot_sha256":"902de55835a706d711b7c2551c58a30a0fcea3f271298a36e0029fbf8763526b","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data. Here we show that a single self-supervised representation can generalize across heterogeneous brain MRI endpoints. We trained BrainDINO, a self-distilled foundation model, on approximately 6.6 million unlabeled axial slices from 20 datasets encompassing broad variation in population, disease, and acquisition setting. Using a frozen encoder with lightweight task heads, BrainDINO supported transfer across tumor segmentation, neu","authors_text":"Chih-Wei Chang, Harini Veeraraghavan, Mingzhe Hu, Mojtaba Safari, Shansong Wang, Xiaofeng Yang, Yizhou Wu, Yuheng Li","cross_cats":["cs.AI","cs.CV"],"headline":"A self-supervised model trained on millions of brain MRI slices yields a unified representation that supports diverse clinical tasks with a frozen encoder and lightweight heads.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T00:21:36Z","title":"BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.27277","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-07T08:42:33.520839Z","id":"8b44807d-a0d0-474a-9e66-5f413dcb57a7","model_set":{"reader":"grok-4.3"},"one_line_summary":"BrainDINO delivers a single self-supervised brain MRI representation that generalizes to tumor segmentation, disease classification, brain age estimation, and other tasks without volumetric pretraining or full fine-tuning.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A self-supervised model trained on millions of brain MRI slices yields a unified representation that supports diverse clinical tasks with a frozen encoder and lightweight heads.","strongest_claim":"a single self-supervised representation can generalize across heterogeneous brain MRI endpoints... large-scale slice-wise self-supervised learning can yield a unified brain MRI representation that supports diverse neuroimaging tasks without volumetric pretraining or full-network fine-tuning","weakest_assumption":"That the 20 datasets and the selected clinical endpoints (tumor segmentation, classification tasks, survival modeling, etc.) are sufficiently representative to support the claim of broad generalizability across all brain MRI applications and acquisition settings."}},"verdict_id":"8b44807d-a0d0-474a-9e66-5f413dcb57a7"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:dd315b20854524eafa4330eb898a6379f42726fe355d5720ec4b951cb11fe2b3","target":"record","created_at":"2026-05-27T01:04:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"4130b87d6702e492ed571669aaa8ca08a2bd068dbecd7543f43005857d7f6755","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T00:21:36Z","title_canon_sha256":"1aab93bb7b167328414952a51b7440d16f473adc1be657eeb9899ee952a1b134"},"schema_version":"1.0","source":{"id":"2604.27277","kind":"arxiv","version":2}},"canonical_sha256":"8abdd0d0e1b622cd16baa30b69bd205ab6333733f94396f437f0d90ddce06261","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8abdd0d0e1b622cd16baa30b69bd205ab6333733f94396f437f0d90ddce06261","first_computed_at":"2026-05-27T01:04:58.558600Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:04:58.558600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"giLkK4YD9WD/5FFF0G0vhj5XoJM5fsZOeaIluFQmL9Kw1CheoRCDf9KyQN+1dnkFYsN88C7y4trAPTtKdA0YAA==","signature_status":"signed_v1","signed_at":"2026-05-27T01:04:58.559262Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.27277","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dd315b20854524eafa4330eb898a6379f42726fe355d5720ec4b951cb11fe2b3","sha256:8226b5218bb891182e0bf95fad36ee2e4e2b68b33f64e51b25b4fb7159023af2"],"state_sha256":"d98db01634347053ca33be5d4ca26710f425f23e101e2c5956502c41755eb751"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hTyiq7+aRprDJq5GmJxTh6kGNSgB6Cn+qga9LettDNpoG5TRK3+kiE35WPOYMImLfTHgzpmE6yWUOH3VooEsCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-29T10:42:54.658630Z","bundle_sha256":"c94612c11fa4ffce3eac42611289b17dc692815b4ab80b0d0abe37735f2068fc"}}