Transforming Care in Psychiatry: Leveraging NLP to optimize Inpatient Violence and Delirium Screening in Acute Care Settings

Clinical notes and reports have become increasingly accessible through the extensive use of EHR (Electronic Health Records) platforms across various clinical settings. Conversely, certain clinical conditions, such as inpatient violence and delirium, pose significant challenges. These conditions can cause psychological suffering and disruptions, affecting both staff and patients.

Screening for these conditions faces numerous challenges, including the subjectivity involved in defining violence or a confusional state, the use of inconsistent tools, clinical complexity, short staffing, time constraints, cultural sensitivity, and potential gaps in risk assessment during transitions from inpatient to outpatient care. To address these challenges, we have developed and implemented applications using John Snow Labs’ Healthcare NLP, capable of ingesting various clinical documents, processing them, and then integrating them with both structured and semi-structured data to generate risk stratification scores.

In this talk, we will share insights into the computational design and workflows, as well as the methods being used to operationalize and integrate NLP tools within clinical workflows. Key Takeaways: The rationale behind of the need for a real-time NLP tool in personalized medicine The architecture of the NLP applications including components and the way they should talk to each other Highlight the challenges and the lessons learned from developing and deploying process and our approach in building capacity for knowledge sharing and collective learning in an acute care setting

About the speaker

Arash Kia

Director, Clinical Data Science at the Mount Sinai Health System

Arash Kia, M.D., M.Sc., is a Clinical Data Science Lead at the Mount Sinai Health System. He is a clinical practitioner with 7+ years of experience as a researcher and data scientist in the field of life sciences and health care technology.

He works to identify clinical opportunities for optimization, translate the opportunities into technology development language, design and develop machine learning applications, and incorporate them into clinical practice workflow as AI-based clinical decision support tools.

Arash has designed, developed, and deployed predictive & recommendation engines for readmission, sepsis, clinical deterioration, severe malnutrition, and others.



Sessions: April 2nd – 3rd 2024
Trainings: April 15th – 19th 2024


Presented by