{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CUT6QMA6AKBJLUXZWR73F6ATPH","short_pith_number":"pith:CUT6QMA6","schema_version":"1.0","canonical_sha256":"1527e8301e028295d2f9b47fb2f81379fd71db32bd314dcc345429c41680abf2","source":{"kind":"arxiv","id":"2607.20582","version":1},"attestation_state":"computed","paper":{"title":"Bayesian uncertainty estimation improves clinical decision making in medical AI agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Christiane Kuhl, Daniel Truhn, Dyke Ferber, Frederik Hauke, Jakob Nikolas Kather, Patrick Wienholt, Sven Nebelung","submitted_at":"2026-07-22T11:54:23Z","abstract_excerpt":"Machine learning models for medical image analysis typically lack a reliable measure of confidence, limiting their use in ambiguous or atypical cases. Here we show that Monte Carlo dropout, applied to a multi-task chest-radiograph classifier (eight thoracic findings, 137,593 training images), provides an epistemic uncertainty signal that tracks generalisation across training-set scales and flags confident yet error-prone predictions. Adding this signal to the point prediction raised error-detection AUROC from 0.74 to 0.77 ($\\Delta$AUROC +0.023, 95% CI [+0.014, +0.033]). In a controlled 2x2 fac"},"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":"2607.20582","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-22T11:54:23Z","cross_cats_sorted":["cs.AI","cs.MA"],"title_canon_sha256":"14f42e0ba806ee8f86b8b4eb3c8a5b5ff9a3d7e9ce8e1c76d434a6823dd2fc73","abstract_canon_sha256":"810db1c9a7fc22d9a01f39ae2053e1b1daf496ea8a65275b4f1c3aadae132019"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-24T00:23:23.776471Z","signature_b64":"FDYcWa2C+7mPBmq+Kc6vd7jwq9Q5yqN9annszlqPuRtDkb4AiWKOD1AkpuG+TJkRNU3St3jbowrANFA24LjYBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1527e8301e028295d2f9b47fb2f81379fd71db32bd314dcc345429c41680abf2","last_reissued_at":"2026-07-24T00:23:23.775597Z","signature_status":"signed_v1","first_computed_at":"2026-07-24T00:23:23.775597Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian uncertainty estimation improves clinical decision making in medical AI agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.LG","authors_text":"Christiane Kuhl, Daniel Truhn, Dyke Ferber, Frederik Hauke, Jakob Nikolas Kather, Patrick Wienholt, Sven Nebelung","submitted_at":"2026-07-22T11:54:23Z","abstract_excerpt":"Machine learning models for medical image analysis typically lack a reliable measure of confidence, limiting their use in ambiguous or atypical cases. Here we show that Monte Carlo dropout, applied to a multi-task chest-radiograph classifier (eight thoracic findings, 137,593 training images), provides an epistemic uncertainty signal that tracks generalisation across training-set scales and flags confident yet error-prone predictions. Adding this signal to the point prediction raised error-detection AUROC from 0.74 to 0.77 ($\\Delta$AUROC +0.023, 95% CI [+0.014, +0.033]). In a controlled 2x2 fac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20582","kind":"arxiv","version":1},"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/2607.20582/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":"2607.20582","created_at":"2026-07-24T00:23:23.776047+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.20582v1","created_at":"2026-07-24T00:23:23.776047+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20582","created_at":"2026-07-24T00:23:23.776047+00:00"},{"alias_kind":"pith_short_12","alias_value":"CUT6QMA6AKBJ","created_at":"2026-07-24T00:23:23.776047+00:00"},{"alias_kind":"pith_short_16","alias_value":"CUT6QMA6AKBJLUXZ","created_at":"2026-07-24T00:23:23.776047+00:00"},{"alias_kind":"pith_short_8","alias_value":"CUT6QMA6","created_at":"2026-07-24T00:23:23.776047+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH","json":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH.json","graph_json":"https://pith.science/api/pith-number/CUT6QMA6AKBJLUXZWR73F6ATPH/graph.json","events_json":"https://pith.science/api/pith-number/CUT6QMA6AKBJLUXZWR73F6ATPH/events.json","paper":"https://pith.science/paper/CUT6QMA6"},"agent_actions":{"view_html":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH","download_json":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH.json","view_paper":"https://pith.science/paper/CUT6QMA6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.20582&json=true","fetch_graph":"https://pith.science/api/pith-number/CUT6QMA6AKBJLUXZWR73F6ATPH/graph.json","fetch_events":"https://pith.science/api/pith-number/CUT6QMA6AKBJLUXZWR73F6ATPH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH/action/storage_attestation","attest_author":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH/action/author_attestation","sign_citation":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH/action/citation_signature","submit_replication":"https://pith.science/pith/CUT6QMA6AKBJLUXZWR73F6ATPH/action/replication_record"}},"created_at":"2026-07-24T00:23:23.776047+00:00","updated_at":"2026-07-24T00:23:23.776047+00:00"}