{"prediction": {"id": "f4033292-4595-5178-ad0f-44642a1476cd", "work_id": "ae40c841-dc30-51fb-bee2-0aaca79b74c9", "study_id": "9549c009-9355-536b-ac72-3c5d719cb5ce", "kind": "Model evaluation", "pages": [6], "version": "Version of record", "source_work_id": "ae40c841-dc30-51fb-bee2-0aaca79b74c9", "title": "Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests.", "doi": "10.1038/s41746-025-01890-x", "pdf_sha256": "8e1c5015d8e6b9751bd7eab1ee1ebb34eaf38252dd0363b33241e6095a06e49c", "physical_page_count": 10}, "fields": {"comparator": {"value": "observed_imaging", "source_basis": "verified_profile", "pages": [6]}, "endpoint": {"value": "tumor_volume__relative_error", "source_basis": "original_pdf_review", "pages": [6]}, "family": {"value": "prediction_error", "source_basis": "verified_profile", "pages": [6]}, "measured_variable": {"value": "tumor_volume", "source_basis": "original_pdf_review", "pages": [1, 6]}, "measurement_role": {"value": "prediction_error", "source_basis": "original_pdf_review", "pages": [6]}, "metric": {"value": "relative_error", "source_basis": "original_pdf_review", "pages": [6]}, "model": {"value": "machine_learning", "source_basis": "verified_profile", "pages": [1]}, "population": {"value": "human_observations", "source_basis": "verified_profile", "pages": [1, 6]}, "reference_type": {"value": "observed_imaging_reference", "source_basis": "verified_profile", "pages": [6]}, "reported_time_window": {"value": "Patient A follow-ups 2 and 3, days 582 and 786 from diagnosis in Figure 5A", "source_basis": "original_pdf_review", "pages": [6]}, "role": {"value": "prediction_error", "source_basis": "verified_profile", "pages": [6]}, "unit": {"value": "dimensionless", "source_basis": "verified_profile", "pages": [6]}}, "endpoints": [{"term": "prediction_error", "label": "Prediction or fitting error"}], "validations": [{"id": "63484269-242c-564d-8457-662611052952", "claim_id": "f4033292-4595-5178-ad0f-44642a1476cd", "coverage": "direct_comparison", "support": "supports", "source": "same_source", "overlap": "overlap", "comparability": "matched", "statistics": "not_reported", "anchors": [{"work_id": "ae40c841-dc30-51fb-bee2-0aaca79b74c9", "version": "Version of record", "pages": [6]}]}], "model_resources": {"total": 2, "page": 1, "limit": 50, "rows": [{"id": 50, "work_id": "ae40c841-dc30-51fb-bee2-0aaca79b74c9", "claim_id": "f4033292-4595-5178-ad0f-44642a1476cd", "kind": "model", "label": "prostateModel.py", "repository": "GitHub", "identifier": "prostateModel.py", "source_page": 8, "availability": "author_reported_public", "url": "https://github.com/daniel-camacho-gomez/ProstateNet/blob/ProstateNet/prostateModel.py", "scope": "claim_specific", "resource_version": null, "version_basis": null, "repository_commit": null, "repository_member": null, "source_file_sha256": "be36844d9185798aab3371ee13cc97e8432fb8b195f202b7c9d428638ea0b7a7", "source_basis": "ai_reviewed_primary_file", "priority": 0, "doi": "10.1038/s41746-025-01890-x", "title": "Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests.", "version": "Version of record", "pdf_sha256": "8e1c5015d8e6b9751bd7eab1ee1ebb34eaf38252dd0363b33241e6095a06e49c", "association_scope": "prediction", "resource_relation": "indexed_resource", "source_review": {"indexed": true, "source_basis": "ai_reviewed_primary_file"}}, {"id": 558, "work_id": "ae40c841-dc30-51fb-bee2-0aaca79b74c9", "claim_id": "f4033292-4595-5178-ad0f-44642a1476cd", "kind": "model", "label": "ProstateNet physics-informed tumor growth twin", "repository": "Author GitHub", "identifier": "daniel-camacho-gomez/ProstateNet", "source_page": 8, "availability": "author_reported_public", "url": "https://github.com/daniel-camacho-gomez/ProstateNet", "scope": "claim_specific", "resource_version": null, "version_basis": null, "repository_commit": null, "repository_member": null, "source_file_sha256": null, "source_basis": "legacy_curated_resource", "priority": 2, "doi": "10.1038/s41746-025-01890-x", "title": "Physics-informed machine learning digital twin for reconstructing prostate cancer tumor growth via PSA tests.", "version": "Version of record", "pdf_sha256": "8e1c5015d8e6b9751bd7eab1ee1ebb34eaf38252dd0363b33241e6095a06e49c", "association_scope": "prediction", "resource_relation": "indexed_resource", "source_review": {"indexed": true, "source_basis": "legacy_curated_resource"}}], "review_status": "indexed_source"}}