Time : - 11:00 AM to 12:00 Noon
Venue : -A007, R & D Block
Title: Information Extraction Problems in Healthcare Applications: demands and challenges
Abstract: The task of Information Extraction (IE) is to automatically extract structured information from unstructured and/or semi-structured machine-readable documents and other electronically represented sources. Well-known techniques of Natural Language Processing (NLP) are used to process human language texts for IE. Information extraction in the medical domain involves handling a number of vital tasks around an unstructured document like EHR/EMR, clinical records etc, which include identification of medical terms, attributes such as negation, uncertainty, severity, relationships between entities and mapping terms in the document to concepts in domain specific ontologies.
Oracle Language Services (also called, OCI Language Services) have adopted different healthcare models (mainly deep learning-based with recent addition of LLM models) to accomplish these tasks. The business units in Oracle aim to leverage AI/ML building blocks offered by OCI Language Services to build applications and ML models for use cases such as Readmission Predictive Risk Models, disease specific Risk Models, Clinical decision support systems etc., for which OCI Language Services need to develop foundational healthcare NLP models such as Health entity extraction (HNER), Health entity linking to medical standards (HMEL), Health Assertion Status Detection (HAD) and Health Relation Extraction (HRE). In this talk, we introduce these services from the standpoint of Healthcare services and also discuss functional aspects and deployment challenges of different Healthcare NLP models from the standpoint of rigorous performance thresholds and strict latency criteria for healthcare services.