AI/ snomed-ct · clinical-nlp · healthcare-ai · benchmarks

AI Medical Coding Benchmark Finds Rare Codes Stump Models

A new arXiv benchmark built from MIMIC-IV discharge summaries shows even the best method finds only 15% of masked SNOMED CT codes on the first guess.

A new benchmark finds that the best available method for guessing a missing medical code gets it right on the first try only 15 percent of the time.

The paper, arXiv:2609.17855 ('SNOMED CT Concept Recommendation from Masked Clinical Context'), builds its test from the SNOMED CT Entity Linking Challenge dataset, itself derived from MIMIC-IV-Note discharge summaries. Researchers masked a clinical term from its surrounding context in 75,491 note-concept pairs across 272 discharge summaries, then asked several systems to guess the missing SNOMED CT concept from a candidate list built during training. A simple TF-IDF model beat every fancier alternative, including a retrieval-augmented hybrid, hitting Recall@1 of 14.81%, Recall@10 of 33.43%, and an MRR of 0.2114. Retrieval augmentation actually made results worse, dropping Recall@10 to 31.99%.

That gap matters because SNOMED CT is the backbone of clinical interoperability - it's how hospitals, insurers, and researchers are supposed to talk about the same diagnosis without agreeing on the words. The benchmark shows performance collapses for anything but common terms: Recall@10 falls to 7.74% for concepts seen only once or twice in training, versus 43.90% for well-represented ones, and nearly 1 in 10 test concepts never showed up in training at all.

In other words, coding tools are only as good as the paperwork that trained them, and no amount of retrieval-augmented machinery fixed that here - a useful reality check before anyone pitches an 'AI coding assistant' as ready for rare diagnoses.

TR

The Revision

Written by an AI system from the public sources credited above. How we write →