Automatic ICD-10 codes association to diagnosis: Bulgarian case

Abstract

This paper presents an approach for the automatic association of diagnoses in Bulgarian language to ICD-10 codes. Since this task is currently performed manually by medical professionals, the ability to automate it would save time and allow doctors to focus more on patient care. The presented approach employs a fine-tuned language model (i.e. BERT) as a multi-class classification model. As there are several different types of BERT models, we conduct experiments to assess the applicability of domain and language specific model adaptation. To train our models we use a big corpora of about 350,000 textual descriptions of diagnosis in Bulgarian language annotated with ICD-10 codes. We conduct experiments comparing the accuracy of ICD-10 code prediction using different types of BERT language models. The results show that the MultilingualBERT model (Accuracy Top 1-81%; Macro F1-86%, MRR Top 5-88%) outperforms other models. However, all models seem to suffer from the class imbalance in the training dataset. The achieved accuracy of prediction in the experiments can be evaluated as very high, given the huge amount of classes and noisiness of the data. The result also provides evidence that the collected dataset and the proposed approach can be useful in building an application to help medical practitioners with this task and encourages further research to improve the prediction

Automatic Transformation of Clinical Narratives into Structured Format

“…For the purposes of ICD-10 diagnosis classification task, a dataset created by Velichkov et al (2020a) was used with additional cleaning and preprocessing to reduce the noise. It contains 345,591 sample diagnoses with their corresponding ICD-10 codes.…”

“…We compile a dataset of Bulgarian medical articles starting with a dataset from Velichkov et al (2020a) and enhancing it by scraping Bulgarian medical websites, containing general medical articles. The final dataset consists of about 10,000 articles.…”

“…Zhang et al (2018) propose an entropy-based model for temporal information extraction for medical facts and medical events from Chinese clinical text. Some preliminary work was reported for structuring clinical information in Bulgarian -for extraction of various numerical values related to lab test by Boytcheva et al (2015), medication extraction by Boytcheva (2011), and annotation with ICD-10 codes by Velichkov et al (2020a). The automatic generation of diabetes register from unstructured clinical texts using rule-based NLP techniques is presented in Tcharaktchiev et al (2018).…”