{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JS6TSMOLH6HISDKUVKAFZKENN7","short_pith_number":"pith:JS6TSMOL","schema_version":"1.0","canonical_sha256":"4cbd3931cb3f8e890d54aa805ca88d6fe7a7b10d3d05160be82bd3edd2e71e48","source":{"kind":"arxiv","id":"2605.14251","version":1},"attestation_state":"computed","paper":{"title":"Generative Deep Learning for Computational Destaining and Restaining of Unregistered Digital Pathology Images","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Conditional GANs can destain and restain unregistered pathology slides from external institutions after simple preprocessing","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aarushi Kulkarni, Alarice Lowe, Pratik Shah","submitted_at":"2026-05-14T01:37:58Z","abstract_excerpt":"Conditional generative adversarial networks (cGANs) have enabled high-fidelity computational staining and destaining of hematoxylin and eosin (H&E) in digital pathology whole-slide images (WSI). However, their ability to generalize to out-of-distribution WSI across institutions without retraining remains insufficiently characterized. Previously developed cGAN models trained on 102 registered prostate core biopsy WSIs from Brigham and Women's Hospital were evaluated on 82 spatially unregistered WSIs acquired at Stanford University. To mitigate domain shift without retraining, a preprocessing pi"},"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":false},"canonical_record":{"source":{"id":"2605.14251","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T01:37:58Z","cross_cats_sorted":[],"title_canon_sha256":"c2b8f2d2ddda082c63f9fdaa35a937fd3382bc8a2fd35a7596d59ad8e2ef6b21","abstract_canon_sha256":"c2447c0d590f1a879af67baa813ccbadf6f583e63a4d29379f134e36124ba25c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:10.564713Z","signature_b64":"NtU6Vo5nW/jueQNxSbeIcMOLsGNMKoWjHhQw+ll85p3hQddl2Ui5iJa3DN2wCOzlkNR/2WtsBPVK0LjRc2gjCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cbd3931cb3f8e890d54aa805ca88d6fe7a7b10d3d05160be82bd3edd2e71e48","last_reissued_at":"2026-05-17T23:39:10.564240Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:10.564240Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative Deep Learning for Computational Destaining and Restaining of Unregistered Digital Pathology Images","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Conditional GANs can destain and restain unregistered pathology slides from external institutions after simple preprocessing","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aarushi Kulkarni, Alarice Lowe, Pratik Shah","submitted_at":"2026-05-14T01:37:58Z","abstract_excerpt":"Conditional generative adversarial networks (cGANs) have enabled high-fidelity computational staining and destaining of hematoxylin and eosin (H&E) in digital pathology whole-slide images (WSI). However, their ability to generalize to out-of-distribution WSI across institutions without retraining remains insufficiently characterized. Previously developed cGAN models trained on 102 registered prostate core biopsy WSIs from Brigham and Women's Hospital were evaluated on 82 spatially unregistered WSIs acquired at Stanford University. To mitigate domain shift without retraining, a preprocessing pi"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"H&E restaining from computationally destained outputs outperformed direct staining from ground-truth unstained inputs across all metrics (PCC: 0.798 vs. 0.715; SSIM: 0.756 vs. 0.718; PSNR: 20.08 vs. 18.51 dB), suggesting that preprocessing quality may be more limiting than model capacity.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That histogram-based stain normalization and channel-wise intensity calibration are sufficient to mitigate domain shift for unregistered WSIs without introducing artifacts that affect pathological interpretation, particularly for malignant structures.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"cGAN-based virtual H&E destaining and restaining generalizes to unregistered external pathology WSIs via preprocessing, with destaining PCC of 0.854 and restaining outperforming direct staining.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Conditional GANs can destain and restain unregistered pathology slides from external institutions after simple preprocessing","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9091ac6ce46dac297e4cfd61d431e1e514b3808cef82e0d418f7c9acc46aa50e"},"source":{"id":"2605.14251","kind":"arxiv","version":1},"verdict":{"id":"39788bf9-d1d4-45ea-9c97-f65b67d9344c","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:57:54.254778Z","strongest_claim":"H&E restaining from computationally destained outputs outperformed direct staining from ground-truth unstained inputs across all metrics (PCC: 0.798 vs. 0.715; SSIM: 0.756 vs. 0.718; PSNR: 20.08 vs. 18.51 dB), suggesting that preprocessing quality may be more limiting than model capacity.","one_line_summary":"cGAN-based virtual H&E destaining and restaining generalizes to unregistered external pathology WSIs via preprocessing, with destaining PCC of 0.854 and restaining outperforming direct staining.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That histogram-based stain normalization and channel-wise intensity calibration are sufficient to mitigate domain shift for unregistered WSIs without introducing artifacts that affect pathological interpretation, particularly for malignant structures.","pith_extraction_headline":"Conditional GANs can destain and restain unregistered pathology slides from external institutions after simple preprocessing"},"references":{"count":18,"sample":[{"doi":"","year":2025,"title":"U.s. cancer statistics prostate cancer stat bite,","work_id":"ab9b8e05-d687-4259-ad41-c7117fe960cb","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2008,"title":"The 20-core prostate biopsy protocol–a new gold standard?","work_id":"83997d09-c6bf-4a3f-8b93-f614675affe1","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2014,"title":"Tissue processing and hematoxylin and eosin staining,","work_id":"569a7e07-a850-41aa-bd0c-236de2a5eddf","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2015,"title":"Automated histology analysis: Opportunities for signal processing,","work_id":"5e30a135-d81d-4a0b-a368-cc61f91b30c2","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2010,"title":"Identification of a cell of origin for human prostate cancer,","work_id":"e03f5d88-f6cf-4aa5-8a83-d43ab7cf6e67","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":18,"snapshot_sha256":"b06127e6ec9cf14c9b7af62222216c48c762c00ba18a494906fa6540d7cc06e7","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":"2605.14251","created_at":"2026-05-17T23:39:10.564331+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.14251v1","created_at":"2026-05-17T23:39:10.564331+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14251","created_at":"2026-05-17T23:39:10.564331+00:00"},{"alias_kind":"pith_short_12","alias_value":"JS6TSMOLH6HI","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"JS6TSMOLH6HISDKU","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"JS6TSMOL","created_at":"2026-05-18T12:33:37.589309+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/JS6TSMOLH6HISDKUVKAFZKENN7","json":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7.json","graph_json":"https://pith.science/api/pith-number/JS6TSMOLH6HISDKUVKAFZKENN7/graph.json","events_json":"https://pith.science/api/pith-number/JS6TSMOLH6HISDKUVKAFZKENN7/events.json","paper":"https://pith.science/paper/JS6TSMOL"},"agent_actions":{"view_html":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7","download_json":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7.json","view_paper":"https://pith.science/paper/JS6TSMOL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.14251&json=true","fetch_graph":"https://pith.science/api/pith-number/JS6TSMOLH6HISDKUVKAFZKENN7/graph.json","fetch_events":"https://pith.science/api/pith-number/JS6TSMOLH6HISDKUVKAFZKENN7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7/action/storage_attestation","attest_author":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7/action/author_attestation","sign_citation":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7/action/citation_signature","submit_replication":"https://pith.science/pith/JS6TSMOLH6HISDKUVKAFZKENN7/action/replication_record"}},"created_at":"2026-05-17T23:39:10.564331+00:00","updated_at":"2026-05-17T23:39:10.564331+00:00"}