RuCCoD: Towards Automated ICD Coding in Russian
Aleksandr Nesterov, Andrey Sakhovskiy, Ivan Sviridov, Airat Valiev, Vladimir Makharev, Petr Anokhin, Galina Zubkova, Elena Tutubalina
2025-03-10
Summary
This paper talks about RuCCoD, a study that explores how to use AI to automate medical coding in Russian, which is a language with limited resources for healthcare technology
What's the problem?
Medical coding, especially in Russian, is difficult because there aren't enough tools or datasets to help doctors and researchers. Manual coding can be slow and prone to errors, which affects the accuracy of diagnoses and treatments
What's the solution?
The researchers created a new dataset for Russian medical coding with over 10,000 annotated entities and 1,500 unique codes. They tested advanced AI models like BERT and LLaMA to see how well they could predict medical codes. By using the best-performing model, they showed that AI-generated codes were more accurate than those manually assigned by doctors, improving diagnosis prediction significantly
Why it matters?
This matters because automating medical coding in Russian could make healthcare faster and more reliable. It could help doctors focus on treating patients instead of spending time on paperwork, and it opens up opportunities to use AI in other languages with limited healthcare resources
Abstract
This study investigates the feasibility of automating clinical coding in Russian, a language with limited biomedical resources. We present a new dataset for ICD coding, which includes diagnosis fields from electronic health records (EHRs) annotated with over 10,000 entities and more than 1,500 unique ICD codes. This dataset serves as a benchmark for several state-of-the-art models, including BERT, LLaMA with LoRA, and RAG, with additional experiments examining transfer learning across domains (from PubMed abstracts to medical diagnosis) and terminologies (from UMLS concepts to ICD codes). We then apply the best-performing model to label an in-house EHR dataset containing patient histories from 2017 to 2021. Our experiments, conducted on a carefully curated test set, demonstrate that training with the automated predicted codes leads to a significant improvement in accuracy compared to manually annotated data from physicians. We believe our findings offer valuable insights into the potential for automating clinical coding in resource-limited languages like Russian, which could enhance clinical efficiency and data accuracy in these contexts.