aiqbench analysis
This research reframes automatic ICD coding as a retrieval and ranking problem rather than simple classification, significantly improving the accuracy of primary diagnosis identification. By explicitly modeling the order of diagnostic codes, the proposed framework addresses a critical gap in current state-of-the-art methods that often overlooks this essential sequence for medical billing and diagnosis. The approach demonstrates measurable gains in both ranking metrics and standard classification scores, offering a more robust solution for clinical note processing.
LTR-ICD: A Ranking-Aware Framework for Automatic ICD Coding
arXiv:2510.13922v2 Announce Type: replace Abstract: Clinical notes contain unstructured text provided by clinicians during patient encounters. These notes are usually accompanied by a sequence of diagnostic codes following the International Classification of Diseases (ICD). Correctly assigning and ordering ICD codes is essential for medical diagnosis and reimbursement. However, automating this task remains challenging. State-of-the-art methods treated this problem as a classification task, leading to ignoring the order of ICD codes that is essential for different purposes. In this work, as a first attempt, we approach this task from a retrieval system perspective to consider the order of codes, thus formulating this problem as a classification and ranking task. Our results and analysis show that the proposed framework has a superior ability to identify high-priority codes compared to other methods. For instance, our model's accuracy in correctly ranking primary diagnosis codes is 47%, compared to 20% for the state-of-the-art classifier. Additionally, in terms of classification metrics, the proposed model achieves a micro- and macro-F1 scores of 0.6065 and 0.2904, respectively, surpassing the previous best model with scores of 0.6035 and 0.2741.
Abstract:Clinical notes contain unstructured text provided by clinicians during patient encounters. These notes are usually accompanied by a sequence of diagnostic codes following the International Classification of Diseases (ICD). Correctly assigning and ordering ICD codes is essential for medical diagnosis and reimbursement. However, automating this task remains challenging. State-of-the-art methods treated this problem as a classification task, leading to ignoring the order of ICD codes that is essential for different purposes. In this work, as a first attempt, we approach this task from a retrieval system perspective to consider the order of codes, thus formulating this problem as a classification and ranking task. Our results and analysis show that the proposed framework has a superior ability to identify high-priority codes compared to other methods. For instance, our model's accuracy in correctly ranking primary diagnosis codes is 47%, compared to 20% for the state-of-the-art classifier. Additionally, in terms of classification metrics, the proposed model achieves a micro- and macro-F1 scores of 0.6065 and 0.2904, respectively, surpassing the previous best model with scores of 0.6035 and 0.2741.
Machine Learning (cs.LG); Computation and Language (cs.CL); Information Retrieval (cs.IR)
arXiv:2510.13922 [cs.LG]
(or arXiv:2510.13922v2 [cs.LG] for this version)
M. Mansoori, A. Soliman, and F. Etminani. 2026. In Proceedings of the 17th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics (BCB '26)
https://doi.org/10.1145/3807503.3819375 DOI(s) linking to related resources
Full text under the publisher's open licence · 224 words
Related coverage
Handling Imbalanced Classification: What Works Better Than SMOTE
Analytics Vidhya · Jul 12, 2026
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · Sep 17, 2026
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
A shared agentic platform for Wood Mackenzie, on Amazon Bedrock AgentCore
AWS · Sep 17, 2026