scholarly journals Predictive models for diabetes mellitus using machine learning techniques

2019 ◽  
Vol 19 (1) ◽  
Author(s):  
Hang Lai ◽  
Huaxiong Huang ◽  
Karim Keshavjee ◽  
Aziz Guergachi ◽  
Xin Gao
Keyword(s):  
Diabetes Mellitus ◽  
Body Mass Index ◽  
Logistic Regression ◽  
Blood Glucose ◽  
Random Forest ◽  
Decision Tree ◽  
High Density Lipoprotein ◽  
Predictive Models ◽  
Fasting Blood Glucose ◽  
Density Lipoprotein

Abstract Background Diabetes Mellitus is an increasingly prevalent chronic disease characterized by the body’s inability to metabolize glucose. The objective of this study was to build an effective predictive model with high sensitivity and selectivity to better identify Canadian patients at risk of having Diabetes Mellitus based on patient demographic data and the laboratory results during their visits to medical facilities. Methods Using the most recent records of 13,309 Canadian patients aged between 18 and 90 years, along with their laboratory information (age, sex, fasting blood glucose, body mass index, high-density lipoprotein, triglycerides, blood pressure, and low-density lipoprotein), we built predictive models using Logistic Regression and Gradient Boosting Machine (GBM) techniques. The area under the receiver operating characteristic curve (AROC) was used to evaluate the discriminatory capability of these models. We used the adjusted threshold method and the class weight method to improve sensitivity – the proportion of Diabetes Mellitus patients correctly predicted by the model. We also compared these models to other learning machine techniques such as Decision Tree and Random Forest. Results The AROC for the proposed GBM model is 84.7% with a sensitivity of 71.6% and the AROC for the proposed Logistic Regression model is 84.0% with a sensitivity of 73.4%. The GBM and Logistic Regression models perform better than the Random Forest and Decision Tree models. Conclusions The ability of our model to predict patients with Diabetes using some commonly used lab results is high with satisfactory sensitivity. These models can be built into an online computer program to help physicians in predicting patients with future occurrence of diabetes and providing necessary preventive interventions. The model is developed and validated on the Canadian population which is more specific and powerful to apply on Canadian patients than existing models developed from US or other populations. Fasting blood glucose, body mass index, high-density lipoprotein, and triglycerides were the most important predictors in these models.

scholarly journals Profil Lipid, Peroksidasi Lipid, dan Status Inflamatif Wanita Penderita Sindrom Metabolik

2011 ◽  
Vol 5 (5) ◽  
pp. 212
Author(s):  
Hery Winarsi ◽  
Siwi P.M. Wijayanti ◽  
Agus Purwanto
Keyword(s):  
Metabolic Syndrome ◽  
Body Mass Index ◽  
Blood Glucose ◽  
Blood Glucose Level ◽  
High Density Lipoprotein ◽  
Low Density Lipoprotein ◽  
Density Lipoprotein ◽  
Low Density ◽  
C Reactive Protein ◽  
Reactive Protein

Prevalensi sindrom metabolik (SM) di Indonesia (13,13%) tergolong tinggi dengan kecenderungan terus meningkat. Salah satu akibat SM adalah disfungsi endotel, sebagai awal penyakit kardiovaskuler yang diinduksi oleh stres oksidatif dan inflamatif. Penelitian ini bertujuan untuk mengeksplorasiprofil lipid, peroksidasi lipid, dan marker inflamasi pada wanita penderita SM di Purwokerto. Sebanyak 30 wanita dengan kadar gula darah diatas normal, obesitas body mass index (BMI) > 25 kg/m2, dan berusia 40-65 tahun dilibatkan sebagai responden yang dipilih melalui survei di PoliklinikPenyakit Dalam Rumah Sakit Margono Soekarjo. Kadar kolesterol total, trigliserida, high density lipoprotein, low density lipoprotein, malondialdehid, dan plasma C-reactive protein ditentukan dalam darah responden yang mempunyai kadar gula sewaktu > 200 mg/dL. Ditemukan bahwa wanitadengan SM rata-rata berumur 50,4 tahun; BMI 31,89 kg/m2; kadar gula darah 219,4 mg/dL; kolesterol total 216,73 mg/dL; trigliserida 218,13 mg/dL; HDL 46,59 mg/dL; LDL 146,27 mg/dL; MDA 2943,4 pmol/mL; C-RP 7,62 mg/L; dan tekanan darah 153/103 mmHg. Hasil ini menunjukkan bahwapenderita SM mengalami dislipidemia disertai dengan status antioksidan rendah dan inflamasi.Kata kunci: Wanita sindrom metabolik, profil lipid, lipid peroksida, malondialdehid, C-reactive proteinAbstractPrevalence of metabolic syndrome (MS) in Indonesia (13,13%) is high and tends to increase. One of the consequences of MS is endothelial dysfunction leading to cardiovascular disease which is inducted by oxidative stressand inflammation. The aim of the present research is to explore lipid profile, lipid peroxidation, and inflammatory marker level on metabolic syndrome women in Purwokerto. Thirty women with blood glucose level greater than normal, body mass index (BMI) > 25 kg/m2, 40-65 years of age were recruited as respondent through selection by a survey in Internal Medicine Polyclinic of Margono Soekarjo Hospital in Purwokerto. In respondents with blood glucose level > 200 mg/dL, total blood cholesterol level, high density lipoprotein, low density lipoprotein, malondialdehid, and plasma C-reactive protein were determined. It was found that the MS women were 50,4 years of age; BMI 31,89 kg/m2; blood glucose 219,4 mg/dL; total cholesterol 216,73 mg/dL; triglyceride 218,13 mg/dL; HDL 46,59 mg/dL; LDL 146,27mg/dL; MDA 2943,4 pmol/mL; C-RP 7,62 mg/L; and blood pressure 153/103 mmHg. It indicates that SM women experience dyslipidemia with low antioxidant and inflammation.Key words: Metabolic syndrome women, lipid profile, peroxide lipid, malondialdehid, C-reactive protein

scholarly journals Metformin reduces circulating malondialdehyde-modified low-density lipoprotein in type 2 diabetes mellitus

2014 ◽  
Vol 37 (4) ◽  
pp. 243 ◽  
Author(s):  
Masahiro Ohira ◽  
Takashi Yamaguchi ◽  
Atsuhito Saiki ◽  
Noriko Ban ◽  
Hidetoshi Kawana ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Blood Glucose ◽  
Low Density Lipoprotein ◽  
Fasting Blood Glucose ◽  
Density Lipoprotein ◽  
Hypoglycemic Agents ◽  
Low Density

Purpose: Type 2 diabetes is known to be associated with increasing cardiovascular mortality. Malondialdehyde-modified LDL (MDA-LDL) is an oxidized LDL and is increased in patients with diabetes or hypertriglyceridemia. Elevated MDA-LDL has been reported to be a risk factor of atherosclerosis or cardiovascular disease. Sitagliptin is a dipeptidyl peptidase-4 inhibitor and a new class of hypoglycemic agents. In this study, the effects of increasing the dose of metformin and add-on sitagliptin on MDA-LDL were examined in type 2 diabetes patients. Methods: Seventy patients with type 2 diabetes, inadequately controlled despite on-going treatment with metformin 500 mg/day, were enrolled in this randomized controlled trial. The patients received additional metformin (500 mg/day) or sitagliptin (50 mg/day) for 6 months, and changes in metabolic parameters including MDA-LDL were evaluated. Results: After 6 months of treatment, add-on sitagliptin (n=35) improved fasting blood glucose (FBG) and hemoglobin A1c (HbA1c) to significantly greater extent than increasing the dose of metformin (n=35). There were no differences in total cholesterol and low-density lipoprotein cholesterol levels between two groups. MDA-LDL levels (mean±S.E.) decreased significantly with increasing the dose of metformin (from 94.40±6.35 to 77.83±4.74 U/L, P < 0.005), but remained unchanged with add-on sitagliptin treatment (from 89.94±5.59 to 98.46±6.78 U/L, p > 0.05). Multiple linear regression analysis identified increasing the dose of metformin treatment as the only independent factor associated with decreased MDA-LDL (β coefficient 0.367, P < 0.0119), and no significant correlation between change in MDA-LDL and fasting blood glucose or HbA1c. Conclusion: These results suggest that increasing the dose of metformin improves serum MDA-LDL levels in type 2 diabetes mellitus.

scholarly journals A Scientific Validation of Antihyperglycemic and Antihyperlipidemic Attributes of Trichosanthes dioica

2013 ◽  
Vol 2013 ◽  
pp. 1-7 ◽  
Author(s):  
Prashant Kumar Rai ◽  
Sharad Kumar Gupta ◽  
Amrita Kumari Srivastava ◽  
Rajesh Kumar Gupta ◽  
Geeta Watal
Keyword(s):  
Diabetes Mellitus ◽  
Blood Glucose ◽  
Diabetes Management ◽  
Fasting Blood Glucose ◽  
Density Lipoprotein ◽  
Postprandial Glucose ◽  
Tolerance Test ◽  
Antidiabetic Activity ◽  
Traditional Use ◽  
Trichosanthes Dioica

The present study was undertaken to scientifically validate the antidiabetic activity of aqueous fruit extract of Trichosanthes dioica Roxb. (Family: Cucurbitaceae) which has been traditionally used for managing diabetes mellitus. This plant commonly known as “Sespadula” in English has not been explored scientifically so far for its glycemic potential except by our research group. The study was conducted with variable doses on normal, mild, and severe diabetics models, and several biochemical parameters including blood glucose level (BGL) were assessed. Maximum fall in BGL of 23.8% in normal rats and of 31.3% in mild diabetic rats was observed during their fasting blood glucose (FBG) and glucose tolerance test (GTT) with the dose of 1000 mg kg−1. In severely diabetic animals after 4 weeks treatment with FBG, postprandial glucose, total cholesterol, and triglyceride levels were reduced by 28.7, 30.7, 57.2, and 18.5%, whereas high density lipoprotein, total protein, hemoglobin, and body weight were increased by 33.0, 36.7, 15.7 and 16.7%, respectively. Moreover, urine sugar was reduced from +4 to +1. Thus, the study scientifically validates the traditional use of T. diocia in diabetes management and could be developed as an effective oral agent for treating diabetes mellitus and complications associated with it.

Pioglitazone Used for Reduction in Clozapine-Associated Hypertriglyceridemia

2002 ◽  
Vol 18 (5) ◽  
pp. 254-256
Author(s):  
Jessica Wehner Lea ◽  
Steven C Stoner
Keyword(s):  
Blood Glucose ◽  
Insulin Sensitivity ◽  
High Density Lipoprotein ◽  
Clinical Utility ◽  
Fasting Blood Glucose ◽  
Density Lipoprotein ◽  
Clozapine Therapy ◽  
Case Summary ◽  
Uncontrolled Diabetes ◽  
Lipid Abnormalities

Objective: To report a case of reduction of triglycerides and hemoglobin (Hb)A1C concentrations as a result of the use of the thiazolinedione pioglitazone in a psychiatric patient stabilized on clozapine therapy. Case Summary: A 49-year-old obese white man, who had uncontrolled diabetes and elevated triglycerides, was stabilized on an antipsychotic regimen using clozapine. After failing sulfonylurea therapy, metformin therapy, and fibric acid therapies, pioglitazone was used to reduce fasting blood glucose and HbA1c concentrations and reduce mean triglyceride concentrations. Discussion: Because of the necessity to keep this patient on his current antipsychotic medication and the possible association of exacerbation of diabetes and lipid abnormalities with clozapine, it was necessary to attempt to control the patient's blood glucose and triglycerides through therapeutic means. Numerous attempts to use other diabetic therapies failed. After the addition of pioglitazone to the patient's regimen, significant improvement was observed in the laboratory values, indicating an improvement in diabetes control. Conclusions: It appears that pioglitazone has played a role in reducing fasting plasma glucose concentrations and improving insulin sensitivity, while reducing triglycerides and increasing high-density lipoprotein concentrations in this patient. Expanded trials and further investigation are recommended to evaluate the clinical utility of pioglitazone in antipsychotic-associated diabetes and lipid abnormalities.

scholarly journals METS-IR, a Novel Score to Evaluate Insulin Sensitivity, is Associated with the Urinary Albumin–Creatinine Ratio in Chinese Adults: A Cross-sectional REACTION Study

2021 ◽  
Author(s):  
Wanlu Su ◽  
Jie Wang ◽  
Songyan Yu ◽  
Kang Chen ◽  
Wenhua Yan ◽  
...  
Keyword(s):  
Insulin Resistance ◽  
Blood Pressure ◽  
Body Mass Index ◽  
Blood Glucose ◽  
Fasting Blood Glucose ◽  
Density Lipoprotein ◽  
Urinary Albumin ◽  
Creatinine Ratio ◽  
Chinese Adults ◽  
Albumin Creatinine Ratio

Abstract Background: The metabolic score for insulin resistance (METS-IR) is a novel noninsulin-based metabolic index used as a substitution marker of insulin resistance and for cardiovascular disease evaluation. However, whether METS-IR is associated with the urinary albumin–creatinine ratio (UACR) is not well known. Therefore, we explored the age- and sex-related associations between METS-IR and UACR and compared the discriminative ability of the METS-IR index and its components for elevated UACR. Methods: This study included 37,290 subjects from the REACTION study (Risk Evaluation of cAncers in Chinese diabeTic Individuals: a IONgitudinal study). METS-IR was calculated as follows:(Ln[2×fasting blood glucose{mg/dL}+fasting triglyceride level{mg/dL}]×body mass index)/(Ln [high-density lipoprotein cholesterol{mg/dL}]). Participants were divided into four groups on the basis of METS-IR: <25%, 25%–49%, 50%–74%, and ≥75%. Logistic regression analyses were conducted to determine the associations between METS-IR vs. its components{(fasting blood glucose(FBG), fasting triglyceride level, body mass index, and high-density lipoprotein cholesterol} with UACR. Results: Participants with the highest quartile METS-IR, particularly men,presented a more significant trend towards elevated UACR than towards its components (odds ratio [OR]: 1.260, 95% confidence interval [CI]:1.152–1.378, P<0.001 in all subjects; OR: 1.321, 95% CI: 1.104–1.579, P=0.002 in men; OR: 1.201, 95% CI: 1.083–1.330, P < 0.001 in women). There were significant associations between METS-IR and UACR in younger participants (<65 years for women and 55–64 years for men) and those with an estimated glomerular filtration rate ≥ 90 mL/min/1.73 m2. Increased METS-IR was significantly associated with UACR in men with FBG ≥ 5.6 mmol/L or postprandial blood glucose ≥ 7.8 mmol/L and systolic blood pressure ≥ 120mmHg or diastolic blood pressure ≥ 80 mmHg. The relationships were significant in women with diabetes and hypertension. Conclusions: Increased METS-IR was significantly associated with elevated UACR, and its discriminative power for elevated UACR was superior to that of its components. The findings support the clinical significance of METS-IR for evaluating the cardiometabolic risk and renal function damage among Chinese adults.

scholarly journals PPAR-αAgonist Fenofibrate Decreased Serum Irisin Levels in Type 2 Diabetes Patients with Hypertriglyceridemia

PPAR Research ◽  
2015 ◽  
Vol 2015 ◽  
pp. 1-8 ◽  
Author(s):  
Xiaomeng Feng ◽  
Xia Gao ◽  
Yumei Jia ◽  
Heng Zhang ◽  
Qingrong Pan ◽  
...  
Keyword(s):  
Type 2 Diabetes ◽  
Blood Glucose ◽  
High Density Lipoprotein ◽  
Metabolic Disorders ◽  
Fasting Blood Glucose ◽  
Density Lipoprotein ◽  
Group A ◽  
Group B ◽  
Diabetes Patients

Irisin is related to insulin resistance and metabolic disorders. The physiologic effects of irisin are partially mediated through peroxisome proliferator-activated receptor-α(PPAR-α). We investigated the effect of fenofibrate, a PPAR-αagonist, on serum irisin in type 2 diabetes patients with hypertriglyceridemia. This study evaluated cross-sectional and interventional studies of 25 type 2 diabetes patients with hypertriglyceridemia (group A) and 40 controls (group B). Group A was treated with fenofibrate (200 mg/day) for 8 weeks. Serum irisin and clinical characteristics were examined. Serum irisin was significantly higher in group A compared with group B (45.15±10.48versus35.38±9.97 ng/ml,P<0.001) and correlated with body mass index (r=0.314,P=0.011), fasting blood glucose (r=0.399,P=0.001), total cholesterol (r=0.256,P=0.040), and high-density lipoprotein cholesterol (r=0.247,P=0.047). In multiple regression analysis after controlling for confounders, only fasting blood glucose (β=5.615,P<0.001) and high-density lipoprotein cholesterol (β=19.483,P<0.001) were independently related to serum irisin. After 8 weeks of fenofibrate treatment, serum irisin significantly decreased in group A compared with baseline (45.15±10.48versus38.74±12.54 ng/ml,P=0.011). Conclusively, fenofibrate decreased serum irisin in type 2 diabetes patients with hypertriglyceridemia, indicating that PPAR-αagonists may protect against metabolic disorders by improving irisin resistance.

scholarly journals Association Between Body Mass Index and Cardiometabolic Risks Among Malay Obese Adults

2017 ◽  
Vol 28 (2) ◽  
pp. 202-216 ◽  
Author(s):  
Azlina Daud ◽  
Siti Zuhaidah Shahadan
Keyword(s):  
Blood Pressure ◽  
Body Mass Index ◽  
Blood Glucose ◽  
Waist Circumference ◽  
High Density Lipoprotein ◽  
Density Lipoprotein ◽  
Low Density ◽  
Weak Correlation ◽  
Obese Adults ◽  
Hs Crp

Cardiometabolic risk (CMR) is a cluster of risk factors that may predict the occurrence of cardiovascular diseases and diabetes mellitus. This study investigated the association between CMR with body mass index (BMI) among obese adults. The CMRs, including waist circumference, systolic and diastolic blood pressure, blood glucose, total cholesterol, low-density and high-density lipoprotein, triglyceride, and high-sensitivity C-reactive protein (hs-CRP) level, were quantified in 82 obese adults. The findings indicate the highest correlation coefficient value ( r) was .618 between BMI and waist circumference; a weak correlation between mass index and systolic ( r = .276) and diastolic ( r = .311) blood pressure and hs-CRP ( r = .384); and a very weak correlation between BMI level and fasting blood glucose ( r = .069), total cholesterol ( r = .014), low-density lipoprotein (.013), high-density lipoprotein (−.154), and triglyceride (.173). Future strategies for CMR reduction among obese adults should focus on waist circumference and hs-CRP level.

Sign in / Sign up

Export Citation Format

Share Document