Understanding and Using Natural Language Processing
The cNAE: Natural Language Processing from the Electronic Health Record
Overview:
Estimates suggest that 80% of important information in electronic health records (EHRs) is untapped due to difficulty with using free-text notes. These unstructured data often hold key information necessary for clinical decision-making, quality improvement, and research. We will describe development of a clinical natural language processing (NLP) analytics engine (cNAE) which can easily be used by researchers to expand available data for use in research.
The cNIE: Using Natural Language Processing for Inference or Clincal Decisions
Overview:
The cNIE is designed to evaluate user-defined clinical inference rules in a real-time manner. Clinical Inference Rules (CIRs) can be computable phenotypes, classifiers, predictive models, etc. CIRs are versatile and can be used in a wide range of situations. The cNIE is unique in that it is a general-purpose engine with native cNLP capabilities with extremely low transaction latency. This capability enables what we now term as “transactional NLP” or “transaction clinical NLP”.
Speakers:
Kathleen Bobay, PhD, RN, NEA-BC, FAAN, is a professor, associate dean for faculty affairs, interim department chair, Health Informatics & Data Science, Parkinson School of Health Sciences & Public Health, professor, Marcella Niehoff School of Nursing. Bobay’s research focuses on measuring the value of nurses in patient outcomes. Her research team has demonstrated how staff nurses can help reduce readmissions and Emergency Department visits in medical-surgical patients after discharge by using standardized nursing and patient readiness for discharge assessments. Bobay’s current research involves the use of natural language processing to better identify social determinants of health and phenotyping for clinical decision support. Bobay has a PhD in Nursing Administration and Health Systems from the University of Michigan, and a MSN as a Family Nurse Practitioner from Michigan State University.
Ron Price is the associate vice president of Informatics and Clinical Research at Loyola University Chicago in Information Technology Services (ITS). Price's responsibilities include direction of technology teams that identify, implement and support computing initiatives that advance the institutions' strategic goals. Under his leadership, Informatics and Clinical Research has developed a range of educational applications, clinical research data repositories, award-winning web sites and advanced computing infrastructure. A significant focus of his current activities is the creation of high performance computing clusters and technologies supporting large-scale clinical analytics including clinical natural language processing and image analysis. Price joined the SSOM in 1987 and has held a variety of IT positions. Price is a Red Hat Certified Engineer (RHCE). Mr. Price holds a Bachelor’s degree from the University of North Carolina at Chapel Hill.
Jason Boyda is a programmer analyst responsible for developing and maintaining clinical research applications and software. His work primarily focuses on supporting large-scale, high performance clinical natural language processing systems through his expertise in the Golang programming language, and mobile application development that aims to improve clinical research. Boyda obtained his bachelor’s degree in Computer Science from the University of Iowa and joined Loyola in 2018, where he has co-authored a handful of publications in health sciences, and is currently working towards his master's degree in Health Informatics.
Originally recorded on September 9, 2021, as part of CHOIR's
Overview:
Estimates suggest that 80% of important information in electronic health records (EHRs) is untapped due to difficulty with using free-text notes. These unstructured data often hold key information necessary for clinical decision-making, quality improvement, and research. We will describe development of a clinical natural language processing (NLP) analytics engine (cNAE) which can easily be used by researchers to expand available data for use in research.
Originally recorded on Wednesday, October 20, 2021, as part of CHOIR's
Overview:
The cNIE is designed to evaluate user-defined clinical inference rules in a real-time manner. Clinical Inference Rules (CIRs) can be computable phenotypes, classifiers, predictive models, etc. CIRs are versatile and can be used in a wide range of situations. The cNIE is unique in that it is a general-purpose engine with native cNLP capabilities with extremely low transaction latency. This capability enables what we now term as “transactional NLP” or “transaction clinical NLP”.
Speakers:
Kathleen Bobay, PhD, RN, NEA-BC, FAAN, is a professor, associate dean for faculty affairs, interim department chair, Health Informatics & Data Science, Parkinson School of Health Sciences & Public Health, professor, Marcella Niehoff School of Nursing. Bobay’s research focuses on measuring the value of nurses in patient outcomes. Her research team has demonstrated how staff nurses can help reduce readmissions and Emergency Department visits in medical-surgical patients after discharge by using standardized nursing and patient readiness for discharge assessments. Bobay’s current research involves the use of natural language processing to better identify social determinants of health and phenotyping for clinical decision support. Bobay has a PhD in Nursing Administration and Health Systems from the University of Michigan, and a MSN as a Family Nurse Practitioner from Michigan State University.
Ron Price is the associate vice president of Informatics and Clinical Research at Loyola University Chicago in Information Technology Services (ITS). Price's responsibilities include direction of technology teams that identify, implement and support computing initiatives that advance the institutions' strategic goals. Under his leadership, Informatics and Clinical Research has developed a range of educational applications, clinical research data repositories, award-winning web sites and advanced computing infrastructure. A significant focus of his current activities is the creation of high performance computing clusters and technologies supporting large-scale clinical analytics including clinical natural language processing and image analysis. Price joined the SSOM in 1987 and has held a variety of IT positions. Price is a Red Hat Certified Engineer (RHCE). Mr. Price holds a Bachelor’s degree from the University of North Carolina at Chapel Hill.
Jason Boyda is a programmer analyst responsible for developing and maintaining clinical research applications and software. His work primarily focuses on supporting large-scale, high performance clinical natural language processing systems through his expertise in the Golang programming language, and mobile application development that aims to improve clinical research. Boyda obtained his bachelor’s degree in Computer Science from the University of Iowa and joined Loyola in 2018, where he has co-authored a handful of publications in health sciences, and is currently working towards his master's degree in Health Informatics.