Tsui Laboratory Research Overview
Tsui Lab research projects have a direct effect on hospital staff, patients, and public health by decreasing costs, morbidity and mortality, as well as increasing efficiency and quality of care.
The Tsui Laboratory has developed IMPreSIv system, which predicts infant mortality risk in real-time based on a large data-driven approach. For infant mortality prediction, Dr. Tsui developed the IMPreSIv system using pregnancy and birth delivery data and government birth/death records. Implementing technologies and interventions for clinicians and service providers will reduce risks of infant mortality in prenatal and postnatal care.
Given increased challenges in intensive care, there is a critical need to develop real-time systems for remote monitoring and early warning of patient deterioration. The Tsui Lab built and evaluated the Intensive Care Warning Index (I-WIN) at a tertiary-care children’s hospital. I-WIN has key components including real-time data acquisition component, distributed AI platform, and a graphical user interface. I-WIN currently provides real-time streaming vital signs and early warning of deterioration. The Tsui Lab evaluated I-WIN’s accuracy in data collection and deterioration prediction. I-WIN had 100 percent waveform data accuracy measured by mean square errors by comparing with an FDA-approved waveform monitor system.
To address population needs, conditions, and information about available resources amid the COVID-19 crisis, the Tsui Lab developed real-time monitoring of COVID-19 Outbreak in Population with Information Needs and Guides (COPING) through this interactive survey.
The Tsui Lab developed a suicide risk prediction system, which demonstrated accurate risk prediction based on a large data-driven approach using electronic health records and social media. Learning Suicide Attempt Prediction (LSAP) uses machine learning and natural language processing (NLP) of unstructured narrative clinical notes along with structured data.
The System for Hospital Adaptive Readmission Prediction and Management (SHARP) is a real-time decision support system that estimates individual risk of hospital readmission within 30 days.
The Tsui Laboratory and Children’s Hospital of Pittsburgh of UPMC announced a research partnership to develop a software analytics tool, the Cardiac ICU Warning Index (C-WIN), which will enable early prediction of catastrophic events such as cardiac arrest or emergency endotracheal tube intubation in patients with congenital heart disease.
Expertise Knowledge Platform (EKP) provides functions/services to address the unmet needs an organization faces on how to connect people with different expertise needs or overview of individuals/departments’ strength/weakness.