back to list

Project: Visual Analysis of early identification of diseases and associated complications in premature infants

Description

The Neonatal Intensive Care Unit (NICU) is designed to treat high-risk infants, such as those who were born prematurely, and to monitor vital signs, prevent sickness, and give life support. The NICU is where infants that are critically ill are often treated. Premature infants are more vulnerable to serious diseases because they are born with undeveloped physiological systems, necessitating prompt observation, diagnosis, and treatment from caregivers. The infant's overall health, monitored vital signs, respiratory support, use of antibiotics, the start and management of illness, duration of stay, and prognosis were all essential data that we used in this study.

Because the information gathered by the monitors within the NICU is time-stamped, the monitored vital signs are utilized as a multivariate time series (MTS). In order to explore the high-dimensional time series data, we propose Visual Analytics solutions applying dimensionality reduction techniques to the MTS.

According to prior research, it can be challenging to convert decision support models for early illness detection into practical treatment tools. Therefore, more research, including clinical trials, is required to enhance the model's clinical relevance for the early diagnosis of infant illness and its implications. These data also contain a high proportion of missing values and are irregularly sampled in terms of time and patients.

The objective of this research is to developed a visual analytics solutions for NICU data addressing the issues of MTS data with missing data. As a result, the project will give medical professionals an interactive solution to help them swiftly detect disorders that affect preterm infants and associated complications.

Details

Student

YS

Youming Shan

Supervisor

Anna Vilanova

Secondary supervisor

CC

Carola van Pul - Maxima Medical Center