{"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/explainable-prediction-of-medical-codes-from","title":"Explainable Prediction of Medical Codes from Clinical Text","arxiv_id":"1802.05695","date":"2018-02-15","proceeding":"NAACL 2018 6","authors":["James Mullenbach","Sarah Wiegreffe","Jon Duke","Jimeng Sun","Jacob Eisenstein"],"abstract":"Clinical notes are text documents that are created by clinicians for each\npatient encounter. They are typically accompanied by medical codes, which\ndescribe the diagnosis and treatment. Annotating these codes is labor intensive\nand error prone; furthermore, the connection between the codes and the text is\nnot annotated, obscuring the reasons and details behind specific diagnoses and\ntreatments. We present an attentional convolutional network that predicts\nmedical codes from clinical text. Our method aggregates information across the\ndocument using a convolutional neural network, and uses an attention mechanism\nto select the most relevant segments for each of the thousands of possible\ncodes. The method is accurate, achieving precision@8 of 0.71 and a Micro-F1 of\n0.54, which are both better than the prior state of the art. Furthermore,\nthrough an interpretability evaluation by a physician, we show that the\nattention mechanism identifies meaningful explanations for each code assignment","url_abs":"http://arxiv.org/abs/1802.05695v2","url_pdf":"http://arxiv.org/pdf/1802.05695v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"explainable-prediction-of-medical-codes-from","repo_url":"https://github.com/jamesmullenbach/caml-mimic","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"explainable-prediction-of-medical-codes-from","repo_url":"https://github.com/dalgu90/icd-coding-benchmark","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"explainable-prediction-of-medical-codes-from","repo_url":"https://github.com/HeoTaksung/MIMIC-III_CNN_Attention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"medical-code-prediction","task_name":"Medical Code Prediction"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/medical-code-prediction-on-mimic-iii","task":"Medical Code Prediction","dataset":"MIMIC-III","model":"CAML","rank_in_archive_order":12,"of":18,"metrics":{"Macro-AUC":"89.5","Macro-F1":"8.8","Micro-AUC":"98.6","Micro-F1":"53.9","Precision@15":"56.1","Precision@8":"70.9"},"uses_additional_data":false},{"leaderboard":"/sota/medical-code-prediction-on-mimic-iii","task":"Medical Code Prediction","dataset":"MIMIC-III","model":"DR-CAML","rank_in_archive_order":13,"of":18,"metrics":{"Macro-AUC":"89.7","Macro-F1":"8.6","Micro-AUC":"98.5","Micro-F1":"52.9","Precision@15":"54.8","Precision@8":"69.0"},"uses_additional_data":false},{"leaderboard":"/sota/medical-code-prediction-on-mimic-iii","task":"Medical Code Prediction","dataset":"MIMIC-III","model":"SVM","rank_in_archive_order":14,"of":18,"metrics":{"Micro-F1":"44.1"},"uses_additional_data":false},{"leaderboard":"/sota/medical-code-prediction-on-mimic-iii","task":"Medical Code Prediction","dataset":"MIMIC-III","model":"CNN","rank_in_archive_order":15,"of":18,"metrics":{"Macro-AUC":"80.6","Macro-F1":"4.2","Micro-AUC":"96.9","Micro-F1":"41.9","Precision@15":"44.3","Precision@8":"58.1"},"uses_additional_data":false},{"leaderboard":"/sota/medical-code-prediction-on-mimic-iii","task":"Medical Code Prediction","dataset":"MIMIC-III","model":"Bi-GRU","rank_in_archive_order":16,"of":18,"metrics":{"Macro-AUC":"82.2","Macro-F1":"3.8","Micro-AUC":"97.1","Micro-F1":"41.7","Precision@15":"44.5","Precision@8":"58.5"},"uses_additional_data":false},{"leaderboard":"/sota/medical-code-prediction-on-mimic-iii","task":"Medical Code Prediction","dataset":"MIMIC-III","model":"Logistic Regression","rank_in_archive_order":18,"of":18,"metrics":{"Macro-AUC":"56.1","Macro-F1":"1.1","Micro-AUC":"93.7","Micro-F1":"27.2","Precision@15":"41.1","Precision@8":"54.2"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1802.05695","atlas_url":"https://app.syntology.ai/?focus=1802.05695","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1802.05695"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dalgu90/icd-coding-benchmark","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HeoTaksung/MIMIC-III_CNN_Attention","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jamesmullenbach/caml-mimic","reach":null}],"summary":{"ran_violates":1},"by_repo_kind":{"official":{"samples":1,"ran":1,"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":0,"samples":[{"code_sha256_prefix":"a0d6592556186142","entry":"early_stop","repo":"jamesmullenbach/caml-mimic","repo_kind":"official","path":"learn/training.py","file_url":"https://github.com/jamesmullenbach/caml-mimic/blob/HEAD/learn/training.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a0d6592556186142"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}