Machine learning as new promising technique for selection of significant features in obese women with type 2 diabetes

2020 ◽  
Vol 41 (1) ◽  
Author(s):  
Hajie Lotfi ◽  
Saeed Pirmoradi ◽  
Rasoul Mahmoudi ◽  
Mohammad Teshnehlab ◽  
Roghayeh Sheervalilou ◽  
...  
Keyword(s):  
Machine Learning ◽  
Experimental Data ◽  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Model Assessment ◽  
Obese Women ◽  
Data Set ◽  
Obesity And Diabetes ◽  
Significant Difference

AbstractBackgroundThe global trend of obesity and diabetes is considerable. Recently, the early diagnosis and accurate prediction of type 2 diabetes mellitus (T2DM) patients have been planned to be estimated according to precise and reliable methods, artificial networks and machine learning (ML).Materials and methodsIn this study, an experimental data set of relevant features (adipocytokines and anthropometric levels) obtained from obese women (diabetic and non-diabetic) was analyzed. Machine learning was used to select significant features [by the separability-correlation measure (SCM) algorithm] for classification of women with the best accuracy and the results were evaluated using an artificial neural network (ANN).ResultsAccording to the experimental data analysis, a significant difference (p < 0.05) was found between fasting blood sugar (FBS), hemoglobin A1c (HbA1c) and visfatin level in two groups. Moreover, significant correlations were determined between HbA1c and FBS, homeostatic model assessment (HOMA) and insulin, total cholesterol (TC) level and body mass index (BMI) in non-diabetic women and insulin and HOMA, FBS and HbA1c, insulin and HOMA, systolic blood pressure (SBP) and diastolic blood pressure (DBP), BMI and TC and HbA1c and TC in the diabetic group. Furthermore, there were significant positive correlations between adipocytokines except for the resistin and leptin levels for both groups. The excellent (FBS and HbA1c), good (HOMA) and fair (visfatin, adiponectin and insulin) discriminators of diabetic women were determined based on specificities and sensitivities level. The more selected features in the ML method were FBS, apelin, visfatin, TC, HbA1c and adiponectin.ConclusionsThus, the subset of features involving FBS, apelin, visfatin and HbA1c are significant features and make the best discrimination between groups. In this study, based on statistical and ML results, the useful biomarkers for discrimination of diabetic women were FBS, HbA1c, HOMA, insulin, visfatin, adiponectin and apelin. Eventually, we designed useful software for identification of T2DM and the healthy population to be utilized in clinical diagnosis.

scholarly journals Docking protein 1 and free fatty acids are associated with insulin resistance in patients with type 2 diabetes mellitus

2021 ◽  
Vol 49 (11) ◽  
pp. 030006052110482
Author(s):  
Xiaoqin Ha ◽  
Xiaoling Cai ◽  
Huizhe Cao ◽  
Jie Li ◽  
Bo Yang ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Insulin Resistance ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Effective Means ◽  
Model Assessment ◽  
High Concentration ◽  
Obesity And Diabetes ◽  
Docking Protein

Objective Insulin resistance (IR) is a key defect in type 2 diabetes mellitus (T2DM); therefore, effective means of ameliorating IR are sought. Methods We performed a retrospective cohort study of 154 patients with T2DM and 39 with pre-diabetes (pre-DM). The effects of IR and a high concentration of FFA on gene expression were determined using microarray analysis and quantitative reverse transcription polymerase chain reaction (RT-qPCR) in patients with T2DM or pre-DM. Results Serum FFA concentration and homeostasis model assessment of IR (HOMA-IR) were significantly higher in patients with T2DM but no obesity and in those with pre-DM than in controls. HOMA-IR was significantly associated with T2DM. RT-qPCR showed that the expression of FBJ murine osteosarcoma viral oncogene homolog ( FOS) and AE binding protein 1 ( AEBP1) was much lower in the circulation of participants with obesity and diabetes. RT-qPCR showed that the expression of docking protein 1 ( DOK1) was significantly lower in the blood of participants with diabetes but no obesity and in those with pre-DM than in controls. Conclusions FFA and DOK1 are associated with IR in patients with T2DM but no obesity or pre-DM. The downregulation of DOK1 might inhibit lipid synthesis and induce lipolysis, inducing or worsening IR.

scholarly journals Home Blood Pressure Measurement in the Morning Related to Cancer in Patients with Type 2 Diabetes Mellitus: 10 Years-Term Results of a Longitudinal Study and Cancer

2015 ◽  
Vol 7 (1) ◽  
pp. 1-6
Author(s):  
Kyuzi Kamoi
Keyword(s):  
Diabetes Mellitus ◽  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Risk Factor ◽  
Cause Of Death ◽  
Japanese Patients ◽  
Pressure Measurements ◽  
Significant Difference

Previous longitudinal studies have demonstrated that blood pressure measurements at home (HBP) in the wakening- up display stronger predictive power for death, and vascular complications in patients with type 2 diabetes mellitus (T2DM) than clinic blood pressure measurements (CBP). The leading cause of death was cancer. Patients with T2DM have associated with cancer, and high CBP is a risk factor for cancer. Therefore, this study investigated whether HBP or CBP is related to cancer event in patients with T2DM for 10 years. At baseline, 400 Japanese patients with T2DM were classified as hypertensive (HT) or normotensive (NT) based on HBP and CBP. Mean (± SD) duration was 95 ± 35 months. Primary and secondary endpoints were death and cancer, respectively. Differences in outcome between HT and NT were analyzed using survival curves from Kaplan-Meier analysis and log-rank testing. Associated risk factors were assessed using Cox proportional hazards. On basis of HBP, death and event of cancer were significantly higher in HT than in NT. The leading cause of death was cancer. On basis of CBP, there was no significant difference in the incidence of death and event of cancer between patients with HT and NT at baseline. Associated risk factor for cancer was T2DM. Home morning HT may be reflected more keenly state of cancer than clinic HT, which may be superior to clinic NT. When we meet with such patients, it is important that cancer may be one of many causes for morning HT in Japanese patients with T2DM.

scholarly journals Relationship between homocysteine and non-dipper pattern in patients with type 2 diabetes mellitus

2012 ◽  
Vol 56 (5) ◽  
pp. 285-290 ◽  
Author(s):  
Serdal Korkmaz ◽  
Abdulkerim Yilmaz ◽  
Gürsel Yildiz ◽  
Fatih Kiliçli ◽  
Serhat Içağasioğlu
Keyword(s):  
Diabetes Mellitus ◽  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Control Group ◽  
Diabetic Patients ◽  
Nocturnal Blood Pressure ◽  
Type 2 Dm ◽  
Significant Difference

OBJECTIVE: The rate of reduction of nocturnal blood pressure (NBP) is lesser than normal in patients with type 2 diabetes mellitus (type 2 DM). Hyperhomocysteinemia (HHC) disrupts vascular structure and function, no matter the underlying causes. The risk of development of vascular disease is greater in diabetic patients with hyperhomocysteinemia than in patients with normal homocystein levels. The aim of the study was to investigate whether there are differences of homocystein levels in dipper and non-dippers patients with type 2 DM. SUBJECTS AND METHODS: We compared 50 patients (33 females, 17 males) with type 2 DM and 35 healthy individuals (18 females, 17 males ) in a control group. Ambulatory blood pressure monitoring (ABPM) was performed and homocysteine levels were measured in all patients. RESULTS: We found that the percentage of non-dipper pattern was 72% in patients with type 2 DM and 57% in control group. In diabetic and control individuals, homocystein levels were higher in non-dipper (respectively 13.4 ± 8.1 µmol/L and 11.8 ± 5 µmol/L) than in dipper subjects (respectively, 11.8 ± 5.8 µmol/L and 10.1 ± 4.2 µmol/L), but there was no significant difference between the two groups (respectively, p = 0.545, p = 0.294). CONCLUSION: In both groups, homocystein levels were higher in non-dipper than in dipper participants, but there was no significant difference between the groups. High homocystein levels and the non-dipper pattern increases cardiovascular risk. Therefore, the relationship between nocturnal blood pressure changes and homocystein levels should be investigated in a larger study.

scholarly journals Decreased Plasma Maresin 1 Concentration Is Associated with Diabetic Foot Ulcer

2020 ◽  
Vol 2020 ◽  
pp. 1-7 ◽  
Author(s):  
Tian Miao ◽  
Bangliang Huang ◽  
Niexia He ◽  
Lihua Sun ◽  
Guangsheng Du ◽  
...  
Keyword(s):  
Insulin Resistance ◽  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Insulin Secretion ◽  
Diabetic Foot ◽  
Foot Ulcer ◽  
Diabetic Foot Ulcer ◽  
Model Assessment ◽  
Homeostasis Model Assessment

Aims. To assess the maresin 1 (MaR1) contents in type 2 diabetic patients with or without diabetic foot ulcer and to analyze the association of MaR1 concentrations with several metabolism-related parameters. Methods. Plasma MaR1 concentrations were analyzed in 96 subjects with normal glucose tolerant (NC, n=43), type 2 diabetes (T2DM, n=40), or diabetic foot ulcer (DFU, n=13). The intravenous glucose tolerance test (IVGTT) and biochemical parameters were measured in all participants. Results. Plasma MaR1 concentrations were significant decreased in type 2 diabetes patient with or without DFU compared with NC (both P<0.001) and were lowest in DFU patients among these 3 groups. (DFU vs. T2DM, P<0.05). Plasma MaR1 concentrations were negatively correlated with BMI, waist circumference (Wc), waist hip ratio (WHR), systolic blood pressure (SBP), diastolic blood pressure (DBP), LDL-c, FPG, 2hPG, HbA1c, and homeostasis model assessment for insulin resistance (HOMA-IR) (all P<0.05) and were positively correlated with HDL-c, acute insulin response (AIR), area under the curve of the first-phase (0-10 min) insulin secretion (AUC), and homeostasis model assessment for beta-cell function (HOMA-β) (all P<0.05). After adjusting for age and sex, Wc, WHR, TG, FPG, 2hPG, HbA1c, HOMA-IR, AIR, AUC, and HOMA-β remain statistically significant (all P<0.05). Conclusions. Plasma MaR1 concentration were decreased in T2DM with or without DFUs and were the lowest in DFU patients. The decreased plasma MaR1 strongly associated with obesity, impaired glucose and lipid metabolism, reduced first-phase of glucose-stimulated insulin secretion, and enhanced insulin resistance.

scholarly journals Electroacupuncture and Rosiglitazone Combined Therapy as a Means of Treating Insulin Resistance and Type 2 Diabetes Mellitus: A Randomized Controlled Trial

2013 ◽  
Vol 2013 ◽  
pp. 1-9 ◽  
Author(s):  
Rong-Tsung Lin ◽  
Huei-Chin Pai ◽  
Yu-Chen Lee ◽  
Chung-Yuh Tzeng ◽  
Chin-Hsien Chang ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Insulin Resistance ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Combined Therapy ◽  
Hypoglycemic Activity ◽  
Model Assessment ◽  
Significant Difference ◽  
Plasma Ffa

Aims.To evaluate the efficacy of rosiglitazone (TZD) and electroacupuncture (EA) combined therapy as a treatment for type 2 diabetes mellitus (T2DM) patients by randomized single-blind placebo controlled clinical trial.Methods.A total of 31 newly diagnostic T2DM patients, who fulfilled the study's eligibility criteria, were recruited. The individuals were randomly assigned into two groups, the control group (TZD,N=15) and the experimental group (TZD + EA,N=16). Changes in their plasma free fatty acid (FFA), glucose, and insulin levels, together with their homeostasis model assessment (HOMA) indices, were statistically compared before and after treatment. Hypoglycemic activity (%) was also compared between these two groups.Results.There was no significant difference in hypoglycemic activity between the TZD and TZD + EA group. The effectiveness of the combined therapy seems to derive from an improvement in insulin resistance and a significant lowering of the secreted insulin rather than the effect of TZD alone on T2DM. The combined treatment had no significant adverse effects. A lower plasma FFA concentration is likely to be the mechanism that causes this effect.Conclusion.This combined therapy seems to suppress endogenous insulin secretion by improving insulin resistance via a mechanism involving a reduction in plasma FFA. This trial is registered with ClinicalTrials.govNCT01577095.

scholarly journals Osteoprotegerin Expression and Serum Values in Obese Women with Type 2 Diabetes Mellitus

2021 ◽  
Author(s):  
Nearmeen M. Rashad ◽  
Amal S. El-Shal ◽  
Sally M Shalaby ◽  
Walaa M Sarhan ◽  
Hanim M. Abdel-Nour
Keyword(s):  
Diabetes Mellitus ◽  
Risk Factors ◽  
Type 2 Diabetes ◽  
Type 2 Diabetes Mellitus ◽  
Serum Levels ◽  
Obese Patients ◽  
Obese Women ◽  
Expression Levels ◽  
Obesity And Diabetes

Abstract Obesity and diabetes prevalence are increasing worldwide. We aimed by this work to detect the possible association of osteoprotegerin (OPG) gene expression with visceral adiposity indices and cardiometabolic risk factors among obese patients. This is research enrolled 150 controls and 150 obese cases subdivided into two subgroups non-diabetic (n = 70) and 80 patients with type 2 diabetes mellitus (T2DM). Circulating OPG expression levels were determined by RT-PCR. Serum OPG concentrations were assessed by ELISA.Our results explored that OPG serum levels were higher in the control group compared to obese women and obese diabetics had higher serum levels of OPG in comparison to obese non-diabetic patients. In obese group, regarding expression levels of OPG were higher than controls in diabetic obese patients, the blood expression levels of OPG were higher than non-diabetics patients. We found positive correlations between parameters of MetS and obesity indices, among them, the highest positive correlation found between VAI. After adjustment of the traditional risk factors, stepwise linear regression analysis test revealed that, OPG expression levels were independently correlated with HbA1c, HDL-cholesterol, and WHR. ROC analysis demonstrated that cutoff values of OPG expression levels and serum levels were 2.226 and 7.5; the AUC were 0.833 (95% CI = 0.821–0.946) and 0.981, respectively. Additionally, the sensitivities and the specificities of OPG expression were (90% and 86.2%), and OPG serum levels (98% and %94), respectively.OPG mRNA and serum levels are useful diagnostic biomarkers discriminate metabolic risk among obesity with or without T2DM.

Evaluation of the effects of the national diabetes control and prevention program on a sample of iranian people with type 2 diabetes

2021 ◽  
pp. 1-13
Author(s):  
Simin Haghravan ◽  
Fatemeh Mohammadi-Nasrabadi ◽  
Maryam Rafraf ◽  
Mohammad Asghari Jafarabadi
Keyword(s):  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Prevention Program ◽  
Time Effect ◽  
Anthropometric Indices ◽  
Hba1 C ◽  
Significant Difference ◽  
Before And After ◽  
Control And Prevention

Diabetes prevention programs have been developed in the different countries and it is therefore necessary to gain a better understanding of factors affecting the effectiveness of these programs for each society. Accordingly, the purpose of this study was to evaluate the effect of the National Diabetes Control and Prevention Program (NDCPP) on the biochemical and anthropometric indices among a sample of Iranian patients with type 2 diabetes mellitus (T2DM). A total of 384 patients with T2DM from the primary health care system of Tehran and Tabriz cities during January to June 2020, were included in the study. A two-part questionnaire consisting of items related to socio-economic and demographic characteristics was used and biochemical and anthropometric indices were measured at the beginning and the end of the 3-month intervention period. There was a significant difference in fasting blood sugar (FBS), hemoglobin glycosides (HbA1 C), systolic blood pressure (SBP) and weight before and after the NDCPP (P <  0.001). Also, there was significant improvement in carbohydrate (P = 0.015) and protein intake (P = 0.027) after the NDCPP compared to before the study. No significant difference was observed in waist circumference (WC) (P = 0.689), body mass index (P = 0.784), diastolic blood pressure (P = 0.647), and other dietary nutrient intake before and after the NDCPP. Repeated measure ANOVA revealed a significant decrease in FBS (Time effect P = 0.019) and weight (Time effect P = 0.006) after the NDCPP. Also, residing in different cities had a significant effect on FBS (time×city effect P <  0.001), HbA1 C (time×city effect P <  0.001), and SBP (time×city effect P = 0.037) after the NDCPP. The NDCPP resulted in weight loss, improved HbA1 C, and FBS. These findings suggested that NDCPP had a positive effect on controlling T2DM among Iranian population and can be considered as a way to prevent complications of T2DM.

scholarly journals Clusters of people with type 2 diabetes in the general population: unsupervised machine learning approach using national surveys in Latin America and the Caribbean

2021 ◽  
Vol 9 (1) ◽  
pp. e001889
Author(s):  
Rodrigo M Carrillo-Larco ◽  
Manuel Castillo-Cara ◽  
Cecilia Anza-Ramirez ◽  
Antonio Bernabé-Ortiz
Keyword(s):  
Machine Learning ◽  
Type 2 Diabetes ◽  
Optimal Number ◽  
Training Data ◽  
Data Set ◽  
Number Of Clusters ◽  
National Surveys ◽  
The Caribbean ◽  
Optimal Number Of Clusters

IntroductionWe aimed to identify clusters of people with type 2 diabetes mellitus (T2DM) and to assess whether the frequency of these clusters was consistent across selected countries in Latin America and the Caribbean (LAC).Research design and methodsWe analyzed 13 population-based national surveys in nine countries (n=8361). We used k-means to develop a clustering model; predictors were age, sex, body mass index (BMI), waist circumference (WC), systolic/diastolic blood pressure (SBP/DBP), and T2DM family history. The training data set included all surveys, and the clusters were then predicted in each country-year data set. We used Euclidean distance, elbow and silhouette plots to select the optimal number of clusters and described each cluster according to the underlying predictors (mean and proportions).ResultsThe optimal number of clusters was 4. Cluster 0 grouped more men and those with the highest mean SBP/DBP. Cluster 1 had the highest mean BMI and WC, as well as the largest proportion of T2DM family history. We observed the smallest values of all predictors in cluster 2. Cluster 3 had the highest mean age. When we reflected the four clusters in each country-year data set, a different distribution was observed. For example, cluster 3 was the most frequent in the training data set, and so it was in 7 out of 13 other country-year data sets.ConclusionsUsing unsupervised machine learning algorithms, it was possible to cluster people with T2DM from the general population in LAC; clusters showed unique profiles that could be used to identify the underlying characteristics of the T2DM population in LAC.

scholarly journals Renal Effects of Sulodexide in Type 2 Diabetic Patients without Nephrotic Range Proteinuria

2020 ◽  
Vol 2020 ◽  
pp. 1-6
Author(s):  
Kachonsak Yongwatana ◽  
Ouppatham Supasyndh ◽  
Bancha Satirapoj
Keyword(s):  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Renal Function ◽  
Diabetic Nephropathy ◽  
Control Group ◽  
Diabetic Patients ◽  
Long Term Effects ◽  
Renal Disease Progression ◽  
Significant Difference

Background. Glycosaminoglycan plays an important role in the maintenance of glomerular charge selectivity of diabetic nephropathy. Sulodexide, a mixture of naturally occurring glycosaminoglycan polysaccharide components, has shown a nephroprotective effect in an experimental model of diabetic nephropathy. Although sulodexide reduced albuminuria in patients with type 1 and type 2 diabetes, long-term effects in patients with type 2 diabetes with significant proteinuria have not been established. Objectives. The study was aimed at investigating the effects of sulodexide on proteinuria and renal function in patients with type 2 diabetes and nephropathy. Methods. Fifty-two patients with proteinuria between 500 and 3000 mg/day received sulodexide 200 mg/day for 12 months, while 56 matched patients with type 2 diabetes constituted the control group. All patients received standard metabolic and blood pressure controls. Primary outcome was evaluated as percentage of reduced proteinuria compared with the control group. Renal function was assessed using estimated glomerular filtration rate (GFR). Results. Proteinuria significantly increased in the control group [0.9 (IQR 0.3 to 1.78) to 1.16 (IQR 0.44 to 2.23) g/gCr, P=0.001], whereas it remained stable in the sulodexide group [0.66 (IQR 0.23 to 0.67) to 0.67 (IQR 0.17 to 1.51) g/gCr, P=0.108]. At 12 months, proteinuria was higher by 19.4% (IQR 10.3 to 37.6) in the control group while proteinuria was lower by -17.7% (IQR -53.1 to 3.2) in the sulodexide group with a significant difference between groups (P=0.001). Renal function was noted as a change of estimated GFR, and serum creatinine decreased significantly during the study in both groups but did not significantly differ between groups. No significant changes in the blood pressure, fasting plasma glucose, and hemoglobin A1C were reported. Conclusion. In addition to standard treatment, sulodexide is efficient in maintaining proteinuria in patients with type 2 diabetes with nonnephrotic range proteinuria, but it did not provide an additional benefit concerning renal disease progression.

scholarly journals The Impact of Frailty on the Effectiveness and Safety of Intensive Glucose Control and Blood Pressure-Lowering Therapy for People With Type 2 Diabetes: Results From the ADVANCE Trial

2021 ◽  
Author(s):  
Tu N Nguyen ◽  
Katie Harris ◽  
Mark Woodward ◽  
John Chalmers ◽  
Mark Cooper ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Type 2 Diabetes ◽  
Glucose Control ◽  
Severe Hypoglycaemia ◽  
Intensive Glucose Control ◽  
Vascular Events ◽  
Significant Difference ◽  
Secondary Outcomes ◽  
The Impact

<b>Objective. </b>To develop a frailty index (FI) and explore the relationship of frailty to subsequent adverse outcomes on the effectiveness and safety of more intensive control of both blood glucose and blood pressure (BP), amongst participants with type-2 diabetes in the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial. <p><b>Research Design and Methods. </b>Cox proportional-hazard models were used to estimate the effectiveness and safety of intensive glucose control and BP intervention according to frailty (defined as FI>0.21) status. The primary outcomes were macrovascular events and microvascular events. The secondary outcomes were all-cause mortality, cardiovascular mortality, severe hypoglycaemia, and discontinuation of BP treatment due to hypotension/dizziness.</p> <p><b>Results.</b> There were 11140 participants (mean age 65.8, 42.5% female, 25.7% frail). Frailty was an independent predictor of all primary outcomes and secondary outcomes. The effect of intensive glucose treatment on primary outcomes showed some evidence of attenuation in the frail: HRs for combined major macro- and micro-vascular events 1.03, 95%CI 0.90-1.19 in the frail vs 0.84, 95%CI 0.74-0.94 non-frail (p=0.02). A similar trend was observed with BP intervention. <a>Severe hypoglycaemia rates (per 1000 person-years) were higher in the frail: 8.39 (6.15–10.63) vs. 4.80 (3.84–5.76) in non-frail</a> (p<0.001). There was no significant difference in discontinuation of BP treatment between frailty groups.</p> <p><b>Conclusions. </b>It was possible to retrospectively estimate frailty in a trial population, and this FI identified those at higher risk of poor outcomes. Participants with frailty had some attenuation of benefit from intensive glucose lowering and BP lowering treatments. </p>

Sign in / Sign up

Export Citation Format

Share Document