I-Corps: Trustworthy Medical Code Recommendations

Saint John'S University, Jamaica NY

Investigators

Abstract

The broader impact/commercial potential of this I-Corps project is to revolutionize the medical coding process by increasing the efficiency, accuracy, and explainability of artificially intelligence-driven computer-assisted coding systems. The innovation has the potential to reduce human effort and errors, streamline medical claims and billing processes, and integrate information systems across healthcare providers, payers, and insurance companies. The increased transparency and trustworthiness of artificially intelligent generated codes, in turn, would lead to faster processing and resolution of medical bills and claims with fewer rejections. As a result, healthcare providers will save costs on providing quality healthcare services with better patient outcomes. The project’s broader significance also lies in its applicability to information systems in other industries that rely on coding and structured information, such as pharmaceuticals, bio-medical, legal, and accounting. This I-Corps project is based on the development of an innovative technology that combines deep learning attention mechanisms with symbolic artificially intelligent, specifically knowledge graphs, to improve the explainability of computer-assisted coding models. The approach focuses on providing a deeper understanding of the relationships between highlighted words and predicted standard medical codes, thereby enhancing the accuracy and efficiency of the medical coding process. By addressing the need for increased transparency in artificially intelligent-driven medical coding systems, the project aims to advance the field of artificially intelligent explainability and contribute to the scientific understanding of the subject. The technology has the potential to pave the way for future advancements in artificially intelligent-driven coding systems in various domains and industries reliant on highly structured knowledge bases. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.