Mortality prediction based on imbalanced high-dimensional ICU big data

Liu, Jiankang; Chen, Xian Xiang; Fang, Lipeng; Li, Jun Xia; Yang, Ting; Zhan, Qingyuan; Tong, Kai; Fang, Zhen

doi:10.1016/j.compind.2018.01.017

Cited by 34 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The studies are based on the regular ML techniques for mortality prediction [14][44] [45]. Cheng et al [46] illustrated the use of SVM results to improve the prediction of patient LOSLOS using association rules.…”

Section: B Ml-based Systemsmentioning

confidence: 99%

“…Gentimis et al [11] utilized lab test results to predict the patient's length of stay (LOS), and sepsis has been used for mortality prediction by [12] [13]. Liu et al [14] provided a mortality prediction model using the support vector machine (SVM). [15], [16] and [17] endorsed the use of LR to develop a clinical prediction model for mortality while [18] and [19] supported the use of random forest (RF) and decision tree (DT), respectively.…”

Section: Introductionmentioning

confidence: 99%

“…Although these studies proposed various ML models for mortality prediction, the resulting models have suffered unstable performance, especially in different ICU categories [15]. This is because mortality prediction in the ICU encounters many challenges relating to data diversity, high dimensionality, irregular sampling, and data imbalance [14] [20]. The high dimensionality increases the computational complexity and decreases the model accuracy.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Intensive Care Unit Mortality Prediction: An Improved Patient-Specific Stacking Ensemble Model

et al. 2020

View full text Add to dashboard Cite

The intensive care unit (ICU) admits the most seriously ill patients requiring extensive monitoring. Early ICU mortality prediction is crucial for identifying patients who are at great risk of dying and for providing suitable interventions to save their lives. Accordingly, early prediction of patients at high mortality risk will enable their provision of appropriate and timely medical services. Although various severity scores and machine-learning models have recently been developed for early mortality prediction, such prediction remains challenging. This paper proposes a novel stacking ensemble approach to predict the mortality of ICU patients. Our approach is more accurate and medically intuitive compared to the literature work. Data were prepared and feature selection was processed under the supervision of the ICU domain expert. The data were split into six modalities based on the expert's decisions. For the prediction process, a separate classifier was selected for each modality based on the performance of the classifiers. We utilized the most popular and diverse classifiers in the literature, including linear discriminant analysis, decision tree (DT), multilayer perceptron, k-nearest neighbor, and logistic regression (LR). Then, a stacking ensemble classifier was constructed and optimized based on the fusion of these five classifier decisions. The framework was evaluated using 10,664 patients from the medical information mart for intensive care (MIMIC III) benchmark dataset. To predict patient mortality, extensive experiments were conducted using the patients' time series data of different lengths. For each patient, the first 6, 12, and 24 hours of the first stay were tested. The results indicate that our model outperformed the state-of-the-art approaches in terms of accuracy (94.4%), F1 score (93.7%), precision (96.4%), recall (91.1%), and area under the receiver operator characteristic (ROC) curve (93.3%). These results demonstrate the ability and efficiency of our approach to predict ICU mortality.

show abstract

Section: B Ml-based Systemsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intensive Care Unit Mortality Prediction: An Improved Patient-Specific Stacking Ensemble Model

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Therefore, due to the importance of these factors, Roumani, May [27] compared the performance of several general data mining methods handling imbalanced data problem. Later, García, Robledo [28] and Liu, Chen [29] concentrated on dimensionality reduction as well as handling the imbalanced class problem, and they achieved excellent results in mortality prediction. A summarized list of research works on the mortality prediction of intensive care unit patients is presented in detail in Table 1.…”

Section: Introductionmentioning

confidence: 99%

A New Hybrid Predictive Model to Predict the Early Mortality Risk in Intensive Care Units on a Highly Imbalanced Dataset

et al. 2020

View full text Add to dashboard Cite

Due to the development of biomedical equipment and healthcare level, especially in the Intensive Care Unit (ICU), a considerable amount of data has been collected for analysis. Mortality prediction in the ICUs is considered as one of the most important topics in the healthcare data analysis section. A precise prediction of the mortality risk for patients in ICU could provide us with valuable information about patients' lives and reduce costs at the earliest possible stage. This paper aims to introduce a new hybrid predictive model using the Genetic Algorithm as a feature selection method and a new ensemble classifier based on the combination of Stacking and Boosting ensemble methods to create an early mortality prediction model on a highly imbalanced dataset. The SVM-SMOTE method is used to solve the imbalanced data problem. This paper compares the new model with various machine learning models to validate the efficiency of the introduced model. The achieved results using the shuffle 5-fold cross-validation and random hold-out methods indicate that the new hybrid model has the best performance among other classifiers. Additionally, the Friedman test is applied as a statistical significance test to examine the differences between classifiers. The results of the statistical analysis prove that the proposed model is more effective than other classifiers. Furthermore, the proposed model is compared to APACHE and SAPS scoring systems and is benchmarked against state-of-the-art predictive models applied to the MIMIC dataset for experimental validation and achieved promising results as it outperformed the state-of-the-art models.

show abstract

“…Автори [4] розрізняють два типи проблем якості даних: неповні дані ( відсутні та зсунені) і некоректні дані. В дослідженні [5] проблемами, що ускладнюють отримання якісних прогнозів станів пацієнтів реанімаційного відділення, визначені висока розмірність даних, їх незбалансованість та часова асинхронізація. В статті [6] також досліджуються дані пацієнтів реанімаційного відділення та основною проблемою якісного прогнозування майбутніх станів визначені відсутні дані.…”

unclassified

A Comprehensive Approach for Processing Heterogeneous Medical Data with Missing Values

Biloborodova¹,

Skarga-Bandurova²

2018

Bulletin of the National Technical University "KhPI" a Series of "Information and Modeling"

View full text Add to dashboard Cite

Показана необхідність врахування змішаних наборів даних з відсутніми значеннями. Запропоновано узагальнений підхід до обробки різнорідних медичних даних з урахуванням типу даних, механізму їх відсутності. Проведене опрацювання трьох різних наборів даних з використанням запропонованого підходу та контрольної моделі. Надано порівняння ефективності опрацювання даних. Іл.: 2. Табл.: 2. Бібліогр.: 12 назв. Ключові слова: різнорідні медичні дані, відсутні значення, обробка, контрольна модель.

show abstract

Mortality prediction based on imbalanced high-dimensional ICU big data

Cited by 34 publications

References 8 publications

Intensive Care Unit Mortality Prediction: An Improved Patient-Specific Stacking Ensemble Model

Intensive Care Unit Mortality Prediction: An Improved Patient-Specific Stacking Ensemble Model

A New Hybrid Predictive Model to Predict the Early Mortality Risk in Intensive Care Units on a Highly Imbalanced Dataset

A Comprehensive Approach for Processing Heterogeneous Medical Data with Missing Values

Contact Info

Product

Resources

About