{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/arxiv-2606-17062","title":"RadSEM: A Finding-by-Finding Metric for Clinical Consistency in Radiology Reports","arxiv_id":"2606.17062","date":"2026-06-03","proceeding":null,"authors":["Zhenhong Yang","Zhuoyun Liu","Jintao Fei","Wen Tang","Shichao Quan","Jun Zhao","Jun Xu"],"abstract":"Radiology report evaluation must distinguish clinical compatibility from surface similarity, because negation, laterality, or normal-abnormal polarity can reverse a finding. We propose RadSEM (Radiology Sentence-Level Evaluation Metric), a constrained LLM-assisted metric for reference-based evaluation of radiology Findings. RadSEM rewrites reference and generated reports into ordered atomic finding sentences, each expressing one site-finding proposition. It then performs contradiction-constrained many-to-many matching: incompatible pairs such as \"effusion\" and \"no effusion\" receive no credit, while compatible granularity differences can receive partial credit. A deterministic stage weights pairs by part-whole and abnormal-detail relationships, counts unmatched findings, and produces an abnormal-focused weighted F1 score. Thus, the LLM supports structured rewriting and local alignment rather than acting as an opaque judge. We evaluate RadSEM with SSREE, a controlled monotonicity stress test built from 2,448 de-identified reports expanded into five graded corruption levels. RadSEM achieves Kendall tau_b of 0.957, all-pairs concordance of 97.8%, adjacent concordance of 95.0%, and strict five-level ordering for 81.9% of reports, outperforming radiology-specific and general text metrics while avoiding the failure in which polarity-inverted reports regain lexical overlap. On the same SSREE set, RadSEM outperforms the Ref-anchored RadSEM-Alt policy, improving adjacent concordance from 90.7% to 95.0% and strict ordering from 67.2% to 81.9%. On a 599-triplet synonym/antonym subset, RadSEM prefers synonyms in 597 cases (99.67%). These results suggest that explicit finding units, contradiction-aware matching, and abnormal-focused deterministic scoring make report scoring more interpretable and sensitive to clinically meaningful errors. Code is available at https://github.com/jdh-algo/RadSEM.","url_abs":"https://arxiv.org/abs/2606.17062","url_pdf":"https://arxiv.org/pdf/2606.17062","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2606.17062","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2606.17062"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/jdh-algo/RadSEM","reach":null}],"summary":{"ran":1,"ran_honours":3,"ran_fixture":1,"ran_draft_wrong":4,"unverified":1},"by_repo_kind":{"found_in_text":{"samples":10,"ran":9,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":10,"samples":[{"code_sha256_prefix":"94a0f279d6a53355","entry":"_Dinic","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"94a0f279d6a53355"}},{"code_sha256_prefix":"8a5448269b496388","entry":"_allocated_match_count","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8a5448269b496388"}},{"code_sha256_prefix":"c320337f1d082ac5","entry":"_class_score","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c320337f1d082ac5"}},{"code_sha256_prefix":"b007cffe0fac66d5","entry":"_clean_sentence","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b007cffe0fac66d5"}},{"code_sha256_prefix":"c14a672a6c0c2a7f","entry":"_collect_weighted_pairs","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c14a672a6c0c2a7f"}},{"code_sha256_prefix":"3edd17b8dafac67b","entry":"_deduplicate_pair_edges","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3edd17b8dafac67b"}},{"code_sha256_prefix":"017e9ec74afc4b41","entry":"_pair_weight","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"017e9ec74afc4b41"}},{"code_sha256_prefix":"41f6fad5e30dcf89","entry":"_unique_pair_weights","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"41f6fad5e30dcf89"}},{"code_sha256_prefix":"1fef4c265bddcc6a","entry":"_unmatched_counts","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1fef4c265bddcc6a"}},{"code_sha256_prefix":"29b27aedb8af8403","entry":"calculate_score","repo":"jdh-algo/RadSEM","repo_kind":"found_in_text","path":"step/step3.py","file_url":"https://github.com/jdh-algo/RadSEM/blob/HEAD/step/step3.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"29b27aedb8af8403"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}