Watch stream on YouTube
Abstract
Medical coding is a process of codifying clinical notes to appropriate diagnosis and procedure codes from the standard taxonomies such as ICD (International Classification of Diseases) and CPT (Current Procedure Terminology). Automated medical coding is a research direction of great interest to the healthcare industry especially in the Revenue Cycle Management (RCM) space, as a solution to the traditional human-powered coding limitations. The current manual processes cost the companies $264B+ annual wasteful spend in billing & coding. The most significant drawbacks in manual coding are its Turn-Around Time (TAT), typically 24-48 hours, and the inability to scale to large volumes of data. Automatic medical coding addresses both problems by applying AI and Natural Language Understanding (NLU) by mimicking and automating the manual coding processes. In this seminar, we will discuss a novel solution, GrabQC - Graph based Query Contextualization method that automatically extracts queries from the clinical text, contextualizes the queries using a Graph Neural Network (GNN) model and obtains the ICD Codes using an external IR system.
Link to paper: https://link.springer.com/chapter/10.1007/978-3-030-75762-5_19
Bio of the speakers
Sudarsun Santhiappan
Sudarsun Santhiappan is currently the Chief Scientific Officer and Co-founder at BUDDI.AI. He is an incisive leader with over two decades of invaluable experience in leading AI Research and Development Projects for an end-to-end Artificially Intelligent Product Development, System / Software Design, and Development in assignments and deputations. Adroit in managing the AI Research and Product Development Lifecycle from conceptualization, finalization of product specifications, technology implementation, client handling, change request management, and cross-functional coordination for effective positioning of the products and services. His research interests include Multi-modal Machine Learning, NLP, Text Mining, Scalable & Explainable AI.
Jeshuren Chelladurai
Jeshuren Chelladurai is a Research Scientist at BUDDI.AI and also a Ph.D. research scholar at IIT Madras. He is extremely passionate about Intelligent Application development and its potential to change the way companies do business. His research interests include Knowledge graphs and Natural Language Processing for Clinical text.