Assessment of risk factors in medical data using improved Binary Artificial Fish Swarm Algorithm with Classification upon Evaluation from F-Test

2022 ◽  
Vol 13 (1) ◽  
pp. 0-0
Keyword(s):  
Risk Factors ◽  
Heart Disease ◽  
Research Work ◽  
Insulin Level ◽  
Central Position ◽  
F Test ◽  
Artificial Fish Swarm Algorithm ◽  
Artificial Fish Swarm ◽  
Bayes Algorithm ◽  
Improved Accuracy

The objective of this research work is to effectively deploy improved Binary Artificial Fish Swarm optimization Algorithm (BAFSA) with the data classification techniques. The improvement has been made with accordance to the condition of visual scope and the movement of fish to update towards the central position and chasing behavior towards best point of movement has been modified among the given population. The experimental results show that feature selection by BAFSA and classification by Decision trees and Gaussian Naïve bayes algorithm provides an improved accuracy of about 89.6% for Pima Indian diabetic dataset, 91.1% for lenses dataset and 94.4% for heart disease dataset. Statistical analysis has also been made using Fisher’s F-Test for two sample variance and the selected risk factors such as glucose, insulin level, blood pressure for diabetics datasets, spectacle prescription, tear production rate for lenses dataset and trestbps, cholesterol level, thalach, chest pain type for heart disease dataset are found to be significant with R2<0.001 respectively.

Assessment of risk factors in medical data using improved Binary Artificial Fish Swarm Algorithm with Classification upon Evaluation from F-Test

2022 ◽  
Vol 13 (1) ◽  
pp. 0-0
Keyword(s):  
Risk Factors ◽  
Heart Disease ◽  
Research Work ◽  
Insulin Level ◽  
Central Position ◽  
F Test ◽  
Artificial Fish Swarm Algorithm ◽  
Artificial Fish Swarm ◽  
Bayes Algorithm ◽  
Improved Accuracy

The objective of this research work is to effectively deploy improved Binary Artificial Fish Swarm optimization Algorithm (BAFSA) with the data classification techniques. The improvement has been made with accordance to the condition of visual scope and the movement of fish to update towards the central position and chasing behavior towards best point of movement has been modified among the given population. The experimental results show that feature selection by BAFSA and classification by Decision trees and Gaussian Naïve bayes algorithm provides an improved accuracy of about 89.6% for Pima Indian diabetic dataset, 91.1% for lenses dataset and 94.4% for heart disease dataset. Statistical analysis has also been made using Fisher’s F-Test for two sample variance and the selected risk factors such as glucose, insulin level, blood pressure for diabetics datasets, spectacle prescription, tear production rate for lenses dataset and trestbps, cholesterol level, thalach, chest pain type for heart disease dataset are found to be significant with R2<0.001 respectively.

scholarly journals Prediction of Heart Disease Using Machine Learning Algorithms

2018 ◽  
Vol 7 (2.32) ◽  
pp. 363 ◽  
Author(s):  
N Rajesh ◽  
Maneesha T ◽  
Shaik Hafeez ◽  
Hari Krishna
Keyword(s):  
Risk Factors ◽  
Heart Disease ◽  
Decision Trees ◽  
Naive Bayes ◽  
Heart Diseases ◽  
Naïve Bayes ◽  
Machine Learning Algorithms ◽  
Common Disease ◽  
Bayes Algorithm ◽  
The One

Heart disease is the one of the most common disease. This disease is quite common now a days we used different attributes which can relate to this heart diseases well to find the better method to predict and we also used algorithms for prediction. Naive Bayes, algorithm is analyzed on dataset based on risk factors. We also used decision trees and combination of algorithms for the prediction of heart disease based on the above attributes. The results shown that when the dataset is small naive Bayes algorithm gives the accurate results and when the dataset is large decision trees gives the accurate results.  

scholarly journals Factors of raised C-reactive protein as a potential determinant of Diabetic patients and Risk factor for Coronary Heart Disease

2018 ◽  
Vol 5 (3) ◽  
pp. 3656-3661
Author(s):  
Sharma Sushil Kumar ◽  
Rastogi Parag
Keyword(s):  
Risk Factors ◽  
Coronary Heart Disease ◽  
Type 1 Diabetes ◽  
Heart Disease ◽  
Research Work ◽  
Islet Autoimmunity ◽  
Diabetic Patients ◽  
C Reactive Protein ◽  
Reactive Protein

Elevated C-reactive protein (CRP) levels have previously been described before the onset of type 1 diabetes and gestational diabetes. We hypothesized that inflammation, as reflected by elevated CRP levels, can help predict development of islet autoimmunity or type 1 diabetes. The outcome of this research is to establish potential determinants of raised CRP concentrations in type 1 diabetic patients. Sensitive assay showed ‘low-level’ CRP concentrations in 147 type 1 patients (83M, 64F, median age 30 years, range 13–67). We have done step by step variant examination to relate these CRP levels to known cardiovascular risk factors and demographic data. Only four patients had established Coronary Heart Disease (median CRP 3.43 mg/l vs. 0.85 mg/l, p=0.035). In subjects without overt CHD, multivariate analysis revealed increase in subject age (p=0.0027), BMI (p=0.001) and HbA1 (p=0.013) to be associated with a higher CRP concentration, as was female sex (p=0.025) and a history of CHD in a first-degree relative (p=0.018, n=58). Elevated CRP levels were positively associated with cardiovascular and renal risk factors: age, body mass index, blood pressure, serum cholesterol level, smoking, plasma glucose level and elevated urinary albumin excretion and presence of hypertension were unrelated. This research work advises that certain of the risk factors connected with CHD in type 1 patients are also individually predictive of high CRP concentrations. The reasons for this, and whether intervention would prove valuable, require further analysis

Insulin resistance associated with metabolic syndrome as an indicator of cardiovascular risk

2016 ◽  
Vol 94 (3) ◽  
pp. 189-193
Author(s):  
Aleksandr V. Dontsov ◽  
L. V. Vasil’eva
Keyword(s):  
Risk Factors ◽  
Insulin Resistance ◽  
Metabolic Syndrome ◽  
Coronary Heart Disease ◽  
Heart Disease ◽  
Cardiovascular Risk ◽  
Healthy Subjects ◽  
Tumour Necrosis ◽  
Insulin Level ◽  
Necrosis Factor

Aim. To study blood insulin level in patients with coronary heart disease (CHD) with and without metabolic syndrome (MS) and its relation to cardiovascular risk factors. Materials and methods. We examined 127 patients with stable coronary heart disease (mean age 59.4±5.7 yr) including 63 with MS and 64 without it. The control group consisted of 80 practically healthy subjects. Bloods insulin was determined by immunochemoluminescence, glycated hemoglobin (HbA) by immunoturbidimetry, total cholesterol, HDLP cholesterol, and triglycerides by enzymatic colorimetric method, oxidized LDLP, IL-1β, IL-6, and tumour necrosis factor-a by enzyme immunoassay. The degree of depression was estimated using the Zung scale. Results. Blood insulin level in healthy subjects, CHD patients with and without MS was 6.3 (6.20;6.62), 15.5 (13.96, 16.3) and 9.5 (9.2, 10.1) mcIE/ml respectively (p<0.001). HOMA-IR directly correlated with MBI, waist circumference, HbA total cholesterol, triglyceride, oxidized LDLP, IL-1β, IL-6, and tumour necrosis factor-a levels and negatively with the HDLP cholesterol level. Conclusion. In patients with CHD, metabolic syndrome is associated with a set of additional cardiovascular risk factors, viz. hyperinsulinemia, insulin resistance, increased HbA level, dyslipidemia, oxidative modification of LDLP, activation of proinflammatory cytokines, and depressive disorders. Close correlation of HOMA-IR with certain pathogenetic factors of CHD allow to use it as an indicator of cardiovascular risk in patients with CHD and MS.

Sign in / Sign up

Export Citation Format

Share Document