A Benchmark Data Set for Model-Based Glycemic Control in Critical Care

Introduction: Currently, it is almost impossible to diagnose a patient at the onset of sepsis due to the lack of real-time metrics with high sensitivity and specificity. The purpose of the present study is to determine the diagnostic value of model-based insulin sensitivity (SI) as a new sepsis biomarker in critically ill patients, and compare its performance to classical inflammatory parameters. Materials and method: We monitored hourly SI levels in septic (n=19) and non-septic (n=19) critically ill patients in a 24-hour follow-up study. Patients with type I or type II diabetes mellitus were excluded. SI levels were calculated by a validated glycemic control software, STAR TGC (Stochastic TARgeted Tight Glycemic Controller) (Christchurch, NZ). STAR TGC uses a physiological glucose-insulin system model coupled with stochastic models that capture SI variability in real time. Results: The median SI levels were lower in the sepsis group than in the non-sepsis group (1.9 x 10-4 L/mU/min vs 3.7 x 10-4 L/mU/min, P <0.0001). The areas under the receiver operating characteristic curve (AUROC) of the model-based SI for distinguishing non-sepsis from sepsis was 0.911, superior to white cells count (AUROC 0.611) and temperature (AUROC 0.618). The optimal cut-off value of the test was 2.9 x 10-4 L/mU/min. At this cut-off value, the sensitivity and specificity was 88.9% and 84.2%, respectively. The positive predictive value was 84.2%, while the negative predictive value was 88.9%. Conclusion: The early and relevant decrease of SI in sepsis suggests that it might be a promising novel biomarker of sepsis in critical care. Low SI is diagnostic of sepsis, while high SI rules out sepsis, and these may be determined non-invasively in real-time from glycemic control protocol data.

Download Full-text

Interface Design and Human Factors Considerations for Model-Based Tight Glycemic Control in Critical Care

Journal of Diabetes Science and Technology ◽

10.1177/193229681200600115 ◽

2012 ◽

Vol 6 (1) ◽

pp. 125-134 ◽

Cited By ~ 4

Author(s):

Logan Ward ◽

James Steel ◽

Aaron Le Compte ◽

Alicia Evans ◽

Chia-Siong Tan ◽

...

Keyword(s):

Critical Care ◽

Glycemic Control ◽

Human Factors ◽

Interface Design ◽

Tight Glycemic Control ◽

Model Based

Download Full-text

Insulin Sensitivity, Its Variability and Glycemic Outcome: A model-based analysis of the difficulty in achieving tight glycemic control in critical care

IFAC Proceedings Volumes ◽

10.3182/20110828-6-it-1002.01281 ◽

2011 ◽

Vol 44 (1) ◽

pp. 1745-1750

Author(s):

J. Geoffrey Chase ◽

Aaron J. Le Compte ◽

Jean-Charles Preiser ◽

Christopher G. Pretty ◽

Katherine T. Moorhead ◽

...

Keyword(s):

Critical Care ◽

Insulin Sensitivity ◽

Glycemic Control ◽

Tight Glycemic Control ◽

Model Based ◽

Model Based Analysis

Download Full-text

Tight glycemic control in critical care – The leading role of insulin sensitivity and patient variability: A review and model-based analysis

Computer Methods and Programs in Biomedicine ◽

10.1016/j.cmpb.2010.11.006 ◽

2011 ◽

Vol 102 (2) ◽

pp. 156-171 ◽

Cited By ~ 77

Author(s):

J. Geoffrey Chase ◽

Aaron J. Le Compte ◽

Fatanah Suhaimi ◽

Geoffrey M. Shaw ◽

Adrienne Lynn ◽

...

Keyword(s):

Critical Care ◽

Insulin Sensitivity ◽

Glycemic Control ◽

Tight Glycemic Control ◽

Model Based ◽

Leading Role ◽

Model Based Analysis

Download Full-text

A Novel Spectral Clustering based on Local Distribution

International Journal of Electrical and Computer Engineering (IJECE) ◽

10.11591/ijece.v5i2.pp361-370 ◽

2015 ◽

Vol 5 (2) ◽

pp. 361

Author(s):

Jyotsna Kumar Mandal ◽

Parthajit Roy

Keyword(s):

Outlier Detection ◽

Spectral Clustering ◽

Second Phase ◽

Local Distribution ◽

Data Set ◽

Benchmark Data ◽

Model Based ◽

Clustering Model ◽

Proposed Model ◽

Novel Variation

This paper proposed a novel variation of spectral clustering model based on a novel affinitymetric that considers the distribution of the neighboring points to learn the underlayingstructures in the data set. Proposed affinity metric is calculated using Mahalanobis distancethat exploits the concept of outlier detection for identifying the neighborhoods of the datapoints. RandomWalk Laplacian of the representative graph and its spectra has been consideredfor the clustering purpose and the first k number of eigenvectors have been consideredin the second phase of clustering. The model has been tested with benchmark data and thequality of the output of the proposed model has been tested in various clustering indicesscales.

Download Full-text

Data Entry Errors and Design for Model-Based Tight Glycemic Control in Critical Care

Journal of Diabetes Science and Technology ◽

10.1177/193229681200600116 ◽

2012 ◽

Vol 6 (1) ◽

pp. 135-143 ◽

Cited By ~ 3

Author(s):

Logan Ward ◽

James Steel ◽

Aaron Le Compte ◽

Alicia Evans ◽

Chia-Siong Tan ◽

...

Keyword(s):

Critical Care ◽

Glycemic Control ◽

Data Entry ◽

Tight Glycemic Control ◽

Model Based

Download Full-text

MODES: model-based optimization on distributed embedded systems

Machine Learning ◽

10.1007/s10994-021-06014-6 ◽

2021 ◽

Author(s):

Junjie Shi ◽

Jiang Bian ◽

Jakob Richter ◽

Kuan-Hsun Chen ◽

Jörg Rahnenführer ◽

...

Keyword(s):

Machine Learning ◽

Embedded Systems ◽

Learning Model ◽

Black Box ◽

Distributed Embedded Systems ◽

Data Set ◽

Individual Model ◽

Model Based ◽

Machine Learning Model ◽

Distributed Machine Learning

AbstractThe predictive performance of a machine learning model highly depends on the corresponding hyper-parameter setting. Hence, hyper-parameter tuning is often indispensable. Normally such tuning requires the dedicated machine learning model to be trained and evaluated on centralized data to obtain a performance estimate. However, in a distributed machine learning scenario, it is not always possible to collect all the data from all nodes due to privacy concerns or storage limitations. Moreover, if data has to be transferred through low bandwidth connections it reduces the time available for tuning. Model-Based Optimization (MBO) is one state-of-the-art method for tuning hyper-parameters but the application on distributed machine learning models or federated learning lacks research. This work proposes a framework $$\textit{MODES}$$ MODES that allows to deploy MBO on resource-constrained distributed embedded systems. Each node trains an individual model based on its local data. The goal is to optimize the combined prediction accuracy. The presented framework offers two optimization modes: (1) $$\textit{MODES}$$ MODES -B considers the whole ensemble as a single black box and optimizes the hyper-parameters of each individual model jointly, and (2) $$\textit{MODES}$$ MODES -I considers all models as clones of the same black box which allows it to efficiently parallelize the optimization in a distributed setting. We evaluate $$\textit{MODES}$$ MODES by conducting experiments on the optimization for the hyper-parameters of a random forest and a multi-layer perceptron. The experimental results demonstrate that, with an improvement in terms of mean accuracy ($$\textit{MODES}$$ MODES -B), run-time efficiency ($$\textit{MODES}$$ MODES -I), and statistical stability for both modes, $$\textit{MODES}$$ MODES outperforms the baseline, i.e., carry out tuning with MBO on each node individually with its local sub-data set.

Download Full-text

Improving glycemic control in critical care patients

Nursing Critical Care ◽

10.1097/01.ccn.0000325851.61317.83 ◽

2008 ◽

Vol 3 (4) ◽

pp. 30-35

Author(s):

Julie L. Stone ◽

Linda L. Hutchinson

Keyword(s):

Critical Care ◽

Glycemic Control ◽

Critical Care Patients

Download Full-text

Research of Personalized Recommendation Technology Based on Knowledge Graphs

Applied Sciences ◽

10.3390/app11157104 ◽

2021 ◽

Vol 11 (15) ◽

pp. 7104

Author(s):

Xu Yang ◽

Ziyi Huan ◽

Yisong Zhai ◽

Ting Lin

Keyword(s):

Neural Network ◽

Hot Spot ◽

Experimental Results ◽

Graph Representation ◽

Personalized Recommendation ◽

Knowledge Graph ◽

Learning Technology ◽

Data Set ◽

Model Based ◽

Knowledge Graphs

Nowadays, personalized recommendation based on knowledge graphs has become a hot spot for researchers due to its good recommendation effect. In this paper, we researched personalized recommendation based on knowledge graphs. First of all, we study the knowledge graphs’ construction method and complete the construction of the movie knowledge graphs. Furthermore, we use Neo4j graph database to store the movie data and vividly display it. Then, the classical translation model TransE algorithm in knowledge graph representation learning technology is studied in this paper, and we improved the algorithm through a cross-training method by using the information of the neighboring feature structures of the entities in the knowledge graph. Furthermore, the negative sampling process of TransE algorithm is improved. The experimental results show that the improved TransE model can more accurately vectorize entities and relations. Finally, this paper constructs a recommendation model by combining knowledge graphs with ranking learning and neural network. We propose the Bayesian personalized recommendation model based on knowledge graphs (KG-BPR) and the neural network recommendation model based on knowledge graphs(KG-NN). The semantic information of entities and relations in knowledge graphs is embedded into vector space by using improved TransE method, and we compare the results. The item entity vectors containing external knowledge information are integrated into the BPR model and neural network, respectively, which make up for the lack of knowledge information of the item itself. Finally, the experimental analysis is carried out on MovieLens-1M data set. The experimental results show that the two recommendation models proposed in this paper can effectively improve the accuracy, recall, F1 value and MAP value of recommendation.

Download Full-text