GrabQC: Graph Based Query Contextualization for Automated ICD Coding

Automated medical coding is a process of codifying clinical notes to appropriate diagnosis and procedure codes automatically from the standard taxonomies such as ICD (International Classification Diseases) CPT (Current Procedure Terminology). The manual involves identification entities followed by querying commercial or non-commercial Information Retrieval (IR) system that follows Centre for Medicare Medicaid Services (CMS) guidelines. We propose automate this constructing query IR using auto-extracted notes. \textbf{GrabQC}, \textbf{Gra}ph \textbf{b}ased \textbf{Q}uery \textbf{C}ontextualization method extracts queries text, contextualizes Graph Neural Network (GNN) model obtains Codes an external system. also labelling dataset training model. perform experiments on two datasets text in three different setups assert effectiveness our approach. experimental results show proposed better than compared baselines all settings.