Type 2 diabetes and susceptibility to COVID-19: a machine learning analysis DOI Creative Commons

M Shabestari,

Reyhaneh Azizi, Akram Ghadiri‐Anari

et al.

BMC Endocrine Disorders, Journal Year: 2024, Volume and Issue: 24(1)

Published: Oct. 21, 2024

Type 2 diabetes mellitus (T2DM) was one of the most prevalent comorbidities among patients with coronavirus disease 2019 (COVID-19). Interactions between different metabolic parameters contribute to susceptibility virus; thereby, this study aimed rank importance clinical and laboratory variables as risk factors for COVID-19 or protective against it by applying machine learning methods. This is a retrospective cohort conducted at single center, focusing on population T2DM. The attended Yazd Diabetes Research Center in Yazd, Iran, from February 20, 2020, October 21, 2020. Clinical data were collected within three months before onset pandemic Iran. 59 infected COVID-19, while not. dataset split into 70% training 30% test sets. Principal Component Analysis (PCA) applied data. important components selected using 'sequential feature selector' scored Linear Discriminant model. PCA loadings then multiplied PCs' scores determine original contracting COVID-19. HDL-C, followed eGFR, showed strong negative correlation virus. Higher levels HDL-C eGFR offer protection T2DM population. But, ratio BUN creatinine did not show any correlation. Conversely, AIP, TyG index TG positive such way that higher these increase diastolic BP, TyG-BMI index, MAP, BMI, weight, TC, FPG, HbA1C, Cr, systolic BUN, LDL-C decreased, respectively. atherogenic plasma, triglyceride glucose are significant individuals Meanwhile, high-density lipoprotein cholesterol factor.

Language: Английский

Impact of COVID-19 on metabolic parameters in patients with type 2 diabetes mellitus DOI Creative Commons

M Shabestari,

Forouzan Salari,

Reyhaneh Azizi

et al.

BMC Pulmonary Medicine, Journal Year: 2025, Volume and Issue: 25(1)

Published: Feb. 3, 2025

The Coronavirus Disease 2019 (COVID-19) pandemic has disproportionately affected individuals with Type 2 Diabetes Mellitus (T2DM), making them more susceptible to viral infections. Additionally, COVID-19 and the associated lockdown restrictions have influenced metabolic regulatory mechanisms in this population. This study aims evaluate impact of infection measures on physiological parameters T2DM. retrospective cohort included 118 a prior diagnosis Medical records were reviewed for laboratory tests conducted within three months before onset Iran. Fifty-nine patients confirmed during first underwent follow-up six post-diagnosis. An age- gender-matched group 59 noninfected after pandemic's onset. Clinical analyzed compared each group. In positive group, significant reductions observed triglycerides (TG) (P = 0.001), total cholesterol (TC) 0.028), body mass index (BMI) 0.034), atherogenic plasma (AIP) 0.027), triglyceride-glucose (TyG) triglyceride-glucose-BMI (TyG-BMI) < 0.001) following pre-pandemic levels. Other variables remained unchanged. negative noted TC low-density lipoprotein (LDL-C) 0.01). T2DM mild moderate exhibited improvements TC, TG, BMI, insulin-related indices. Lockdown decreased LDL-C levels without history infection.

Language: Английский

Citations

0
Type 2 diabetes and susceptibility to COVID-19: a machine learning analysis DOI Creative Commons

M Shabestari,

Reyhaneh Azizi, Akram Ghadiri‐Anari

et al.

BMC Endocrine Disorders, Journal Year: 2024, Volume and Issue: 24(1)

Published: Oct. 21, 2024

Type 2 diabetes mellitus (T2DM) was one of the most prevalent comorbidities among patients with coronavirus disease 2019 (COVID-19). Interactions between different metabolic parameters contribute to susceptibility virus; thereby, this study aimed rank importance clinical and laboratory variables as risk factors for COVID-19 or protective against it by applying machine learning methods. This is a retrospective cohort conducted at single center, focusing on population T2DM. The attended Yazd Diabetes Research Center in Yazd, Iran, from February 20, 2020, October 21, 2020. Clinical data were collected within three months before onset pandemic Iran. 59 infected COVID-19, while not. dataset split into 70% training 30% test sets. Principal Component Analysis (PCA) applied data. important components selected using 'sequential feature selector' scored Linear Discriminant model. PCA loadings then multiplied PCs' scores determine original contracting COVID-19. HDL-C, followed eGFR, showed strong negative correlation virus. Higher levels HDL-C eGFR offer protection T2DM population. But, ratio BUN creatinine did not show any correlation. Conversely, AIP, TyG index TG positive such way that higher these increase diastolic BP, TyG-BMI index, MAP, BMI, weight, TC, FPG, HbA1C, Cr, systolic BUN, LDL-C decreased, respectively. atherogenic plasma, triglyceride glucose are significant individuals Meanwhile, high-density lipoprotein cholesterol factor.

Language: Английский

Citations

0