Estimate of the HOMA-IR Cut-off Value Identifying Subjects at Risk of Insulin Resistance Using a Machine Learning Approach

Abdessalem, Abdelhamid; Zidoum, Hamza; Zadjali, F.; Hedjam, Rachid; Al-Ansari, Aliya; Bayoumi, Riad; Al‐Yahyaee, Said; Hassan, Mohammed O.; Albarwani, Sulayma

doi:10.18295/squmj.4.2021.030

Cited by 19 publications

(17 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HOMA-IR value more than equal to 2.5 was considered as presence of insulin resistant and HOMA-B% ≤50% was considered as poor pancreatic beta-cell reserve. 12…”

Section: Methodsmentioning

confidence: 99%

Pancreatic Beta-cell function and degree of Insulin resistance among newly detected type 2 diabetics and their correlation with anthropometric, glucose, and lipid parameters: An Observational cross-sectional study

Saha

2022

Asian J Med Sci

View full text Add to dashboard Cite

Background: Insulin resistance is a major cause for developing type 2 diabetes. However, simultaneously pancreatic beta-cell dysfunction must coexist in for clinical occurrence of type 2 diabetes (T2D). Hence, knowledge regarding residual beta-cell function and degree of insulin resistance is required while treating type 2 diabetic patients. Aims and Objectives: The present study was done to estimate degree of insulin resistance (homeostatic model assessment insulin resistance - HOMA-IR) and pancreatic beta-cell functional capacity (HOMA-B%) among newly detected type 2 diabetics and correlation of these with anthropometric, glucose, and lipid parameters. Materials and Methods: This was an observational cross-sectional study conducted in 100 newly diagnosed type 2 diabetic patients. Detailed anthropometric and clinical examination were carried out. Venous blood samples were drawn for fasting plasma glucose, c-peptide, fasting insulin level, hemoglobin A1C (HbA1C), lipid profile, and postprandial glucose. HOMA-IR and HOMA-B% were calculated using HOMA 2 calculator and correlations were calculated between the study variables. Results: The mean age of the study population was 45.55±11.64 years and 58% of study participants were male. The mean HOMA-IR and HOMA-B% were 2.55±1.75 and 40.67±23.55%, respectively. HOMA-IR positively correlated with abdominal circumference, triglyceride to HDLc ratio, and negatively correlated with HDLc. There were statistically significant negative correlations between HOMA-B% and fasting glucose(r=−0.48, P<0.001), 2 hr post prandial glucose (r=−0.37, P<0.001 and HbA1C (r=−0.24, P=0.01). Conclusion: This study found more reduced beta-cell function compared to reduced insulin sensitivity in new T2D mellitus patients. Hence, this kind of functional assessment needs to be done while selecting appropriate anti-diabetic drugs for a particular patient.

show abstract

“…HOMA-IR value more than equal to 2.5 was considered as presence of insulin resistant and HOMA-B% ≤50% was considered as poor pancreatic beta-cell reserve. 12…”

Section: Methodsmentioning

confidence: 99%

Pancreatic Beta-cell function and degree of Insulin resistance among newly detected type 2 diabetics and their correlation with anthropometric, glucose, and lipid parameters: An Observational cross-sectional study

Saha

2022

Asian J Med Sci

View full text Add to dashboard Cite

show abstract

“…However, its cut-off values vary with age, gender and among different ethnic groups (26)(27)(28). Similar to this, its cut-off values have not been clearly defined among European populations (27,(29)(30)(31).…”

Section: Tolerability and Compliancementioning

confidence: 98%

“…HOMA-IR is a useful tool to distinguish healthy individuals from those with IR and those with T2DM (24,25). However, its cut-off values vary with age, gender and among different ethnic groups (26)(27)(28). Similar to this, its cut-off values have not been clearly defined among European populations (27,(29)(30)(31).…”

Section: Tolerability and Compliancementioning

confidence: 99%

The Effect of Commonly Used Fixed-Dose Single Pill Combinations of Renin-Angiotensin-System Blockers and Calcium Channel Blockers on HOMA-IR Index in Hypertensive Patients with Impaired Fasting Glucose: a 12-Week Randomized Open-Label Prospective Study

Liontos

Biros

Papagiannopoulos

et al. 2022

Maedica

View full text Add to dashboard Cite

Backround: The effect of antihypertensive drugs on glucose homeostasis and insulin resistance remains an issue under investigation. There is evidence that renin-angiotensin system (RAS) blockers may favorably affect glucose metabolism, while treatment with calcium channel blockers (CCBs) is considered to have an overall neutral metabolic effect. However, the effects on glycemic indices may differ among agents within the same class of antihypertensive drugs. Objective: To evaluate the effects of different fixed-dose single pill combinations of RAS blockers with CCBs on homeostatic model assessment for insulin resistance (HOMA-IR). Methods: Drug-naive patients with arterial hypertension (AH) and impaired fasting glucose (IFG) were randomly allocated to open-label fixed, single pill combinations of valsartan 160 mg/day plus amlodipine 5 mg/day (VAL/AMLO group, n = 54), delapril 30 mg/day and manidipine 10 mg/day (DEL/MANI group, n = 53) or telmisartan 80 mg/day and amlodipine 5 mg/day (TEL/AMLO group, n = 51) for 12 weeks. Glycemic indices and HOMA-IR were determined at baseline and post-treatment. Results: A total of 158 patients were included. All treatment combinations effectively reduced blood pressure (systolic and diastolic) to similar levels (all p < 0.001). A decrease in the HOMA-IR index by 22.55% (p < 0.01) was noted following treatment with TEL/AMLO, while an increase by 1.4% (p = 0.57) and 12.65% (p = 0.072) was observed in the VAL/AMLO group and the DEL/MANI group, respectively. These changes were significantly different between TEL/AMLO and DEL/MANI (p < 0.05) as well as between TEL/AMLO and VAL/AMLO (p < 0.001). Conclusions: Despite similar antihypertensive action, the effect of fixed, single pill combinations with TEL/AMLO, VAL/AMLO and DEL/MANI on insulin resistance is in favor of TEL/AMLO. Trial registration: The study protocol was published online in https://diavgeia.gov.gr/ (No: ΒΘ6Σ46906Η-ΑΕΣ) via the Ministry of Digital Governance, after receiving approval from the Scientific Council and Administrative Council of University Hospital of Ioannina (No. of approval: 1/12-06-2014 (issue 150). https://diavgeia.gov.gr/decision/view/%CE%92%CE%986%CE%A346906%CE%97- %CE%91%CE%95%CE%A3 https://diavgeia.gov.gr/doc/%CE%92%CE%986%CE%A346906%CE%97- %CE%91%CE%95%CE%A3?inline=true

show abstract

“…An unsupervised machine learning method was utilized to assess the homeostatic model assessment-insulin resistance (HOMA-IR) cut-off to find individuals at risk of IR based on clinical data [ 160 ]. First, HOMA-IR-correlated features were determined by using a clustering algorithm, and two clusters with the lowest overlap in their HOMA-IR amounts were retrieved.…”

Section: The Application Of ML and Dl Models For The Management Predi...mentioning

confidence: 99%

“…Nonlinear and linear predictive algorithms [152] 2787 consecutive participants Combination of elastic network with RF, SVM, and back-propagation artificial neural network (BP-ANN) algorithms as well as LR [155] 1772 paired data varying from 65 ~ 492 mg/dl and GBT [133] Insulin resistance predicting models 8842 Koreans participants LR, XGBoost, random forest, and ANN [159] 1344 samples HOMA-IR model [160] 2433 T2DM patients MIL-Boost [161] 968 patients not affected by T2DM (FIMMG_obs dataset)…”

Section: Healthy Subjectsmentioning

confidence: 99%

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

Afsaneh¹,

Sharifdini²,

Ghazzaghi³

et al. 2022

Diabetol Metab Syndr

View full text Add to dashboard Cite

Diabetes as a metabolic illness can be characterized by increased amounts of blood glucose. This abnormal increase can lead to critical detriment to the other organs such as the kidneys, eyes, heart, nerves, and blood vessels. Therefore, its prediction, prognosis, and management are essential to prevent harmful effects and also recommend more useful treatments. For these goals, machine learning algorithms have found considerable attention and have been developed successfully. This review surveys the recently proposed machine learning (ML) and deep learning (DL) models for the objectives mentioned earlier. The reported results disclose that the ML and DL algorithms are promising approaches for controlling blood glucose and diabetes. However, they should be improved and employed in large datasets to affirm their applicability.

show abstract

Estimate of the HOMA-IR Cut-off Value Identifying Subjects at Risk of Insulin Resistance Using a Machine Learning Approach

Cited by 19 publications

References 27 publications

Pancreatic Beta-cell function and degree of Insulin resistance among newly detected type 2 diabetics and their correlation with anthropometric, glucose, and lipid parameters: An Observational cross-sectional study

Pancreatic Beta-cell function and degree of Insulin resistance among newly detected type 2 diabetics and their correlation with anthropometric, glucose, and lipid parameters: An Observational cross-sectional study

The Effect of Commonly Used Fixed-Dose Single Pill Combinations of Renin-Angiotensin-System Blockers and Calcium Channel Blockers on HOMA-IR Index in Hypertensive Patients with Impaired Fasting Glucose: a 12-Week Randomized Open-Label Prospective Study

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

Contact Info

Product

Resources

About