Biomarker Discovery for Risk Stratification of Cardiovascular Events using an Improved Genetic Algorithm

2006 ◽  
Author(s):  
Xiaobo Zhou ◽  
Honghui Wang ◽  
Jun Wang ◽  
Gerard Hoehn ◽  
Joseph Azok ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Risk Stratification ◽  
Cardiovascular Events ◽  
Biomarker Discovery ◽  
Improved Genetic Algorithm

Identification of biomarkers for risk stratification of cardiovascular events using genetic algorithm with recursive local floating search

PROTEOMICS ◽  
2009 ◽  
Vol 9 (8) ◽  
pp. 2286-2294 ◽  
Author(s):  
Xiaobo Zhou ◽  
Honghui Wang ◽  
Jun Wang ◽  
Yuan Wang ◽  
Gerard Hoehn ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Risk Stratification ◽  
Cardiovascular Events

Task scheduling algorithm based on improved genetic algorithm in cloud computing environment

2020 ◽  
Vol 13 ◽  
Author(s):  
Ge Weiqing ◽  
Cui Yanru
Keyword(s):  
Genetic Algorithm ◽  
Cloud Computing ◽  
Task Scheduling ◽  
Scheduling Algorithm ◽  
Task Completion ◽  
Improved Genetic Algorithm ◽  
Fitness Functions ◽  
Task Scheduling Algorithm ◽  
Simulation Results

Background: In order to make up for the shortcomings of the traditional algorithm, Min-Min and Max-Min algorithm are combined on the basis of the traditional genetic algorithm. Methods: In this paper, a new cloud computing task scheduling algorithm is proposed, which introduces Min-Min and Max-Min algorithm to generate initialization population, and selects task completion time and load balancing as double fitness functions, which improves the quality of initialization population, algorithm search ability and convergence speed. Results: The simulation results show that the algorithm is superior to the traditional genetic algorithm and is an effective cloud computing task scheduling algorithm. Conclusion: Finally, this paper proposes the possibility of the fusion of the two quadratively improved algorithms and completes the preliminary fusion of the algorithm, but the simulation results of the new algorithm are not ideal and need to be further studied.

Combining improved genetic algorithm and matrix semi-tensor product (STP) in color image encryption

2021 ◽  
Vol 183 ◽  
pp. 108041
Author(s):  
Xiuli Chai ◽  
Xiangcheng Zhi ◽  
Zhihua Gan ◽  
Yushu Zhang ◽  
Yiran Chen ◽  
...  
Keyword(s):  
Genetic Algorithm ◽  
Tensor Product ◽  
Image Encryption ◽  
Color Image ◽  
Improved Genetic Algorithm ◽  
Color Image Encryption

scholarly journals Simple cardiovascular risk stratification using anthropometric measures instead of serum cholesterol. The MORGAM Prospective Cohort Project

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
A Kristensen ◽  
V Rosberg ◽  
J Vishram-Nielsen ◽  
M Pareek ◽  
A Linneberg ◽  
...  
Keyword(s):  
Cardiovascular Risk ◽  
Risk Stratification ◽  
Total Cholesterol ◽  
Serum Cholesterol ◽  
Cardiovascular Events ◽  
Smoking Status ◽  
Predictive Ability ◽  
Serum Total Cholesterol ◽  
Anthropometric Measures ◽  
Blood Pressures

Abstract Background Body composition predicts cardiovascular outcomes, but it is uncertain whether anthropometric measures can replace the more expensive serum total cholesterol for cardiovascular risk stratification in low resource settings. Purpose The purpose of the study was to compare the additive prognostic ability of serum total cholesterol with that of body mass index (BMI), waist/hip ratio (WHR), and estimated fat mass (EFM, calculated using a validated prediction equation), individually and combined. Methods We used data from the MORGAM (MONICA, Risk, Genetics, Archiving, and Monograph) Prospective Cohort Project, an international pooling of cardiovascular cohorts, to determine the relationship between anthropometric measures, serum cholesterol, and cardiovascular events, using multivariable Cox proportional-hazards regression analysis. We further investigated the ability of these measures to enhance prognostication beyond a simpler prediction model, consisting of age, sex, smoking status, systolic blood pressures, and country, using comparison of area under the receiver operating characteristics curve (AUCROC) derived from binary logistic regression models. The primary endpoint was major adverse cardiovascular events (MACE), defined as a composite of death from coronary heart disease, myocardial infarction, or stroke. Results The study population consisted of 52,188 apparently healthy subjects (56.3% men) aged 47±12 years ranging from 20 to 84, derived from 37 European cohorts, with baseline between 1982–2002 all followed for 10 years during which MACE occurred in 2465 (4.7%) subjects. All anthropometric measures (BMI: hazard ratio (HR) 1.04 [95% confidence interval (CI): 1.03–1.05] per kg/m2; WHR: HR 7.5 [4.0–14.0] per unit; EFM: HR 1.02 [1.01–1.02] per kg) as well as serum total cholesterol (HR 1.20 [1.16–1.24] per mmol/l) were significantly associated with MACE (P<0.001 for all), independently of age, sex, smoking status, systolic blood pressures, and country. The addition of serum cholesterol significantly improved the predictive ability of the simple model (AUCROC 0.818 vs. 0.814, P<0.001), as did the combination of WHR, BMI, and EFM (AUCROC 0.817 vs. 0.814, P=0.004). When assessed individually, BMI (AUCROC 0.816 vs. 0.814, P=0.004) and WHR (AUCROC 0.815 vs. 0.814, P=0.02) improved model performance, while EFM narrowly missed significance (AUCROC 0.815 vs. 0.814, P=0.06). There was no significant difference in the predictive ability of a model including serum cholesterol versus that including all three anthropometric measures (AUCROC 0.818 vs. 0.817, P=0.13). The figure shows the pertinent areas under the ROC curve in predicting MACE. Conclusion In this large population-based cohort study, the addition of a combination of anthropometric measures, i.e. BMI, WHR, and EFM, raised the predictive ability of a simple prognostic model comparable to that obtained by the addition of serum total cholesterol. Figure 1 Funding Acknowledgement Type of funding source: None

Sign in / Sign up

Export Citation Format

Share Document