Past Research
Home Care Nursing Notes: Natural Language Processing Analysis of Prediabetes Symptoms
Project Title
Early detection of pre-diabetes high-risk patients using homecare nursing notes
Project Duration
2021.03.01~2024.02.29
Principal Institution
Ministry of Science and ICT
Projective Objective
To develop and evaluate a natural language processing (NLP) algorithm for analyzing prediabetes symptoms to identify high-risk individuals within the community. This algorithm will be applied to home care nursing notes to classify high-risk patients and facilitate early detection through the assessment of patient characteristics.
[Phase 1 Objective]
– Develop and validate an NLP algorithm for analyzing prediabetes symptoms to identify individuals at high risk for prediabetes.
[Phase 2 Objective]
– Apply the developed NLP algorithm to home care nursing notes to classify high-risk patients based on the number and frequency of prediabetes-related symptoms identified.
[Phase 3 Objective]
– Link the classified high-risk patient data with electronic health records (EHRs) to identify personal, social, and clinical characteristics associated with prediabetes risk.
Publication
Published:
– Jeon E, Kim A, Lee J, Heo H, Lee H, Woo K. Developing a Classification algorithm for Prediabetes Risk Detection from Home care Nursing Notes: Using Natural Language Processing. CIN: Computers, Informatics, Nursing. 2023;41(7):539-547. Available from: https://dx.doi.org/10.1097/CIN.0000000000001000 (Jean E. and Kim A. contributed equally to this work.) ▶ Link
– Kim A, Jean E, Lee H, Heo H, Woo K. Risk factors for prediabetes in community-dwelling adults: A generalized estimating equation logistic regression approach with natural language processing insights. Research in Nursing & Health, 2024;47(6):620-634. ▶ Link
Under Review:
– One additional manuscript currently under peer review
Early Detection of Diabetes High-risk Patient Using Homecare Nursing Notes
- GRANT SUPPORT -
2021 – 2023
National Research Foundation of Korea (NRF) Early Career Grant
Title of Project: Early Detection of Diabetes High-risk Patients Using Homecare Nursing Notes
Goal: This study aims to develop and validate a natural language processing algorithm for detecting prediabetes symptoms and identifying high-risk patient characteristics through EHR data linkage.
Funding Agency: South Korea Ministry of Science and ICT
Role: Principal Investigator
Exploring the Association Between Caregiver Education on Diabetes
Self-management in the Community/at Home and Patient Hospitalization
- GRANT SUPPORT -
2020 – 2022
Korea Research Resettlement Fund
Title of Project: Exploring the Association Between Caregiver Education on Diabetes Self-management in the Community/at Home and Patient Hospitalization
Goal: This study aims to examine the effect of caregiver involvement in type 2 diabetes mellitus education within a community and patient diabetes care outcomes through a systematic literature search and meta-analysis.
Funding Agency: Seoul National University
Role: Principal Investigator
Exploring Prevalence of Wound Infections and Related Patient Characteristics in Homecare Using Natural Language Processing
- GRANT SUPPORT -
2019 – 2020
Eugenie and Joseph Doyle Research Partnership Fund
Title of Project: Exploring Prevalence of Wound Infections and Related Patient Characteristics in Homecare Using Natural Language Processing
Goal: This study aims to link Natural Language Processing (NLP) findings to a routinely collected data in homecare (Outcome and Assessment Information Set, OASIS) to capture important, under-reported critical patient information. The investigators will use NLP findings for wound infection-related information from clinical notes and link to a structured data (OASIS) to estimate the prevalence of wound infections and describe related patient characteristics in the homecare population.
Funding Agency: Center for Home Care Policy & Research, Visiting Nurse Service of New York
Role: Principal Investigator
Co-Investigators: Dr. Jingjing Shang, Dr. Maxim Topaz
Exploring Prevalence of Wound Infections and Related Patient Characteristics in Homecare Using Natural Language Processing
- GRANT SUPPORT -
2019 – 2020
Columbia University School of Nursing Intramural Pilot Grant
Title of Project: Exploring Prevalence of Wound Infections and Related Patient Characteristics in Homecare Using Natural Language Processing
Goal: This study aims to use advanced Natural Language Processing (NLP) methods to create and validate an NLP algorithm to extract wound infection-related information from clinical notes (~2.6 million) in the homecare population.
Funding Agency: Columbia University School of Nursing
Role: Principal Investigator
Co-Investigators: Dr. Maxim Topaz, Dr. Jingjing Shang