Neonatal mortality prediction using real-time medical measurements

Gilchrist, Jeff; Ennett, Colleen M.; Frize, Monique; Bariciak, Erika

doi:10.1109/memea.2011.5966653

Cited by 11 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“….0 decision tree software was used by Gilchrist et al [22] in order to develop a real-time mortality prediction model for 12, 24 and 48 hours. To validate the model, a 5-by-2 cross validation technique was used.…”

Section: C5mentioning

confidence: 99%

New Approach based on Machine Learning for Short-Term Mortality Prediction in Neonatal Intensive Care Unit

Kefi¹,

Aloui²,

Naceur³

2019

IJACSA

View full text Add to dashboard Cite

Mortality remains one of the most important outcomes to predict in Intensive Care Units (ICUs). In fact, the sooner mortality is predicted, the better critical decisions are made by doctors based on patient's illness severity. In this paper, a new approach based on Machine Learning (ML) techniques for short-term mortality prediction in Neonatal Intensive Care Unit (NICU) is proposed. This approach relies on many steps. At first, relevant features are selected from available data upon neonates' admission and from the time-series variables collected within the two first hours of stay in the NICU from the Medical Information Mart for Intensive Care III (MIMIC-III). After that, to predict mortality, many classifiers were tested which are Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Classification and Regression Trees (CART), Logistic Regression (LR), Support Vector Machine (SVM), Naïve Bayes (NB) and Random Forest (RF). The experimental results showed that LDA was the best performing classifier with an accuracy equal to 0.947 and AUROC equal to 0.97 with 31 features. The third step of this approach is mortality time prediction using the Galaxy-Random Forest method achieving an f-score equal to 0.871. The proposed approach compared favorably in terms of time, accuracy and AUROC with existing scoring systems and ML techniques. It is the first work predicting neonates mortality based on ML techniques and time-series data after only two hours of admission to the NICU.

show abstract

Section: C5mentioning

confidence: 99%

New Approach based on Machine Learning for Short-Term Mortality Prediction in Neonatal Intensive Care Unit

Kefi¹,

Aloui²,

Naceur³

2019

IJACSA

View full text Add to dashboard Cite

show abstract

“…This data set contains 43 million data points from 357 patients with a 7.6% mortality rate, and 37.8% of patients were female. It was collected prospectively and automatically (Gilchrist et al, 2011a bed location, and will encode the patient ID accordingly with the correct data.…”

Section: Risk Estimation Data Setsmentioning

confidence: 99%

“…It may be possible to generate custom reports to extract some of the data from eClinDoc, but this cannot be done in an automated way, and must be manually extracted and imported into the CDR, which makes using the data in real-time for (Gilchrist et al, , 2010(Gilchrist et al, , 2011b. It also outlines the approach for new neonatal mortality risk models that use summary data (data from admission until up to the first 72 hours after NICU admission) and real-time data segments (data broken down into time segments extracted from the entire duration of the patient's NICU stay) stored in the CDR, with the ability to provide results in real-time (Gilchrist et al, 2011a).…”

Section: Risk Estimation Data Setsmentioning

confidence: 99%

“…This chapter presents the results of the hybrid EAV storage format design and per formance evaluation (Gilchrist et al, 2011b(Gilchrist et al, , 2010, and describes the design and implementation ofthe real-time CDR (Gilchrist et al, 2011b. It also provides the results of the new neonatal mortality risk models that use summary data (data from admission until up to the first 72 hours after NICU admission) and real-time data segments (data broken down into time segments extracted from the entire duration of the patient's NICU stay) stored in the CDR (Gilchrist et al, 2011a). Finally it describes the data access performance achieved to deliver the necessary information to the mortality model to estimate a patient's risk in real-time.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Clinical Decision Support System using Real-Time Data Analysis for a Neonatal Intensive Care Unit

Gilchrist¹

View full text Add to dashboard Cite

There are two main contributions to knowledge presented in this thesis: (1) an efficient Clinical Data Repository (CDR) with a novel data storage format, and (2) ne neonatal mortality risk estimation models that use data analysed in real-time fro the CDR.The CDR collects, stores, and retrieves clinical data in real-time. It uses a nov hybrid Entity-Attribute-Value (EAV) storage format that is faster, uses less comp queries, and allows the data to remain in its original data type while using the Thank you to my supervisor Dr. Monique Prize who inspired me to pursue a PhD after learning of her research work. I knew that I could make a difference if we wor together, and we did! Thank you to my co-supervisor Dr. Colleen M. Ennett who provided an invaluable industry perspective on our research work and who had many interesting ideas and improvements for the research we conducted together.A heartfelt thank you to Dr. Erika Bariciak at the Children's Hospital of Eastern Ontario (CHEO) who always made herself available to answer any questions that I had. Your help and support of my research project was instrumental in my success.I would like to thank and acknowledge the extra work that the NICU team at CHEO provided to help collect data for the patients in the unit including nursing manager Patricia O'Flaherty for all of her support, the patient service clerks, the nurses, the biomedical engineering team, especially Kim Greenwood, and the IS dept.Thank you to my defence committee for all of the useful feedback on my research

show abstract

“…This score functions as a weighted average of the precision and recall. The closer the classifier's F1-score is to 1, the higher the precision and recall values will be [60].…”

Section: F1-scorementioning

confidence: 99%

Applying Data Preparation Methods to Optimize Preterm Birth Prediction

Esty¹

View full text Add to dashboard Cite

The purpose of this work was to develop an accurate prediction model which can process information contained in antenatal databases to determine whether a baby will be born prematurely. The focus was on improved data preprocessing to add to methods developed by previous students in the Carleton MIRG (Medical Information technology Research Group) lab.The machine learning classifiers used included Decision Tree (DT) classifiers (for feature reduction) and the Artificial Neural Network (ANN) classifier (for model evaluation).Missing values and class imbalance was dealt with by applying software packages in the R statistical programming language.This research has shown a marked improvement in the accuracy of predicting preterm births.The final sensitivity and specificity results for the BORN (Better Outcomes Registry and Network) database were: Parous 89.2%, and 67.8%, Nulliparous 89.0% and 71.5%, and for PRAMS (Pregnancy Risk Assessment Monitoring System) database: Parous 84.1% and 71.4%, Nulliparous 83.8% and 76.0%. These improved results are promising. An accurate predictive tool will allow caregivers to implement preventative treatment strategies or to ensure delivery occurs in a tertiary health care Centre.iii

show abstract

Neonatal mortality prediction using real-time medical measurements

Cited by 11 publications

References 11 publications

New Approach based on Machine Learning for Short-Term Mortality Prediction in Neonatal Intensive Care Unit

New Approach based on Machine Learning for Short-Term Mortality Prediction in Neonatal Intensive Care Unit

Clinical Decision Support System using Real-Time Data Analysis for a Neonatal Intensive Care Unit

Applying Data Preparation Methods to Optimize Preterm Birth Prediction

Contact Info

Product

Resources

About