LTR-ICD: A Learning-to-Rank Approach for Automatic ICD Coding
This work addresses the challenge of automating ICD coding with consideration of code order, which is important for medical professionals and billing systems, representing an incremental improvement over existing classification-based methods.
The paper tackled the problem of automatically assigning and ordering ICD codes from clinical notes, which is essential for medical diagnosis and reimbursement, by proposing a learning-to-rank approach that considers code order, resulting in a model accuracy of 47% for ranking primary diagnosis codes compared to 20% for state-of-the-art classifiers and improved F1 scores.
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 are 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 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 0.597 and 0.2660.