Associations between Diabetes Mellitus and Selected Cancers

International Journal of Molecular Sciences, Год журнала: 2024, Номер 25(13), С. 7476 - 7476

Опубликована: Июль 8, 2024

Cancer is one of the major causes mortality and second leading cause death. Diabetes mellitus a serious growing problem worldwide, its prevalence continues to grow; it 12th An association between diabetes cancer has been suggested for more than 100 years. common disease diagnosed among patients with cancer, evidence indicates that approximately 8-18% have diabetes, investigations suggesting an some particular cancers, increasing risk developing cancers such as pancreatic, liver, colon, breast, stomach, few others. Breast colorectal increased from 20% 30% there 97% intrahepatic cholangiocarcinoma or endometrial cancer. On other hand, number therapies increase mellitus. Complications due in may influence choice therapy. Unfortunately, mechanisms associations are still unknown. The aim this review summarize selected update on underlying association.

Язык: Английский

---

Comparative Study of XGBoost and Logistic Regression for Predicting Sarcopenia in Postsurgical Gastric Cancer Patients

Research Square (Research Square), Год журнала: 2025, Номер unknown

Опубликована: Апрель 2, 2025

Background: The use of machine learning (ML) techniques, particularly XGBoost and logistic regression, to predict sarcopenia among postsurgical gastric cancer patients has gained significant attention in recent research. Sarcopenia, characterized by the progressive loss skeletal muscle mass strength, is a serious concern these due its association with poor postoperative outcomes, including increased morbidity mortality. In this study, was used establish risk prediction model for undergoing gastrectomy facilitate early intervention reduce incidence complications. Methods: Gastric who underwent surgery at tertiary comprehensive hospital Nanjing (China) from January 2022 December 2023 were retrospectively included their clinical follow-up data collected. multivariate regression analysis screen factors related results two models compared. area under receiver operating characteristic (ROC) curve (AUC), sensitivity specificity calculated evaluate predictive value model. SHAP (SHapley Additive exPlanations) method explain determine impact features on Results: A total 231 whom 128 (55.4%) developed sarcopenia. univariate LASSO (Least Absolute Shrinkage Selection Operator) cross-validated, 5 key study variables ultimately determined: serum albumin, comorbid diabetes, operation style, nutritional score, ECOG (Eastern Cooperative Oncology Group) performance status score. slightly better AUC (0.987, 95% CI: 0.976-0.998) than (0.918, 0.873-0.963) training set. showed that model, albumin have greater after surgery, especially diabetes score most significant, followed style least impact. Conclusions: In summary, learning-based constructed provides valuable decision support tool screening

Язык: Английский

---

Triglyceride-glucose index: carotid intima-media thickness and cardiovascular risk in a European population

Cardiovascular Diabetology, Год журнала: 2025, Номер 24(1)

Опубликована: Янв. 13, 2025

Abstract Background The triglyceride-glucose (TyG) index is now widely recognized as a marker of insulin resistance and has been linked to the development prognosis atherosclerotic cardiovascular diseases (ASCVD) in numerous populations, particularly Eastern world. Although there are fewer reports from Western world, they sometimes contradictory, absence definitive data on relationship between raised TyG risk suggested opportunity testing this biochemical against well-established vascular such carotid intima media thickness (c-IMT). Methods Primary prevention patients were selected cohort individuals who underwent c-IMT measurement 1984 2018 at Dyslipidemia Center ASST Grande Ospedale Metropolitano Niguarda Milan, Italy. was calculated Ln [fasting TG (mg/dL)×fasting glucose (mg/dL)/2]. Carotid ultrasonography performed using echographic measurements far walls left right common, internal carotids, bifurcations. Patients followed for up 20 years with periodic evaluation parameters. ASCVD events monitored through hospital records, where all regularly examined. Results analysis included 3108 mean age 54.9 ± 13.1 years. Participants generally non-obese, an average BMI 24.6 3.5 Kg/m 2 . Among women, 83.1% postmenopausal. 8.65 0.59. There significant association measurements. Those highest quartiles had significantly higher IMT max compared those lower quartiles. These associations consistent across sites examined remained after adjusting potential confounders. Kaplan-Meier survival revealed increased incidence two Conclusions sensitive European population moderate risk, assessed by measurements, large Lipid Clinic patients.

Язык: Английский

---

Obesity-driven hunger: From pathophysiology to intervention

Obesity Medicine, Год журнала: 2025, Номер unknown, С. 100588 - 100588

Опубликована: Фев. 1, 2025

Язык: Английский

---

Emerging biomarkers in type 2 diabetes mellitus

Advances in clinical chemistry, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Early prediction of postpartum dyslipidemia in gestational diabetes using machine learning models

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 7, 2025

This study addresses a gap in research on predictive models for postpartum dyslipidemia women with gestational diabetes mellitus (GDM). The goal was to develop machine learning-based model predict using early pregnancy clinical data, and the model's robustness evaluated through both internal temporal validation. Clinical data from 15,946 pregnant were utilized. After cleaning, divided into two sets: Dataset A (n = 1,116), used training evaluating model, B 707), Several learning algorithms applied, performance of assessed A, while validate across different time period. Feature significance Information Value (IV), importance analysis, SHAP (SHapley Additive exPlanations) analysis. results showed that among five tested, tree-based ensemble models, such as XGBoost, LightGBM, Random Forest, outperformed others predicting dyslipidemia. In these achieved accuracies 70.54%, 69.64%, respectively, AUC-ROC values 73.10%, 71.94%, 76.14%. Temporal validation indicated XGBoost performed best, achieving an accuracy 81.05% 87.92%. power strengthened by key variables total cholesterol, fasting glucose, triglycerides, BMI, cholesterol being identified most important feature. Further IV analyses confirmed pivotal role concluded XGBoost-based GDM strong consistent validations. By introducing new variables, can identify high-risk groups during pregnancy, supporting intervention potentially improving outcomes reducing complications.

Язык: Английский

---

Smart stimuli-responsive hydrogels for safe oral administration of Insulin: A Review

International Journal of Pharmaceutics, Год журнала: 2025, Номер unknown, С. 125487 - 125487

Опубликована: Март 1, 2025

Язык: Английский

---

Comparative study of XGBoost and logistic regression for predicting sarcopenia in postsurgical gastric cancer patients

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Апрель 14, 2025

The use of machine learning (ML) techniques, particularly XGBoost and logistic regression, to predict sarcopenia among postsurgical gastric cancer patients has gained significant attention in recent research. Sarcopenia, characterized by the progressive loss skeletal muscle mass strength, is a serious concern these due its association with poor postoperative outcomes, including increased morbidity mortality. In this study, was used establish risk prediction model for undergoing gastrectomy facilitate early intervention reduce incidence complications. Gastric who underwent surgery at tertiary comprehensive hospital Nanjing (China) from January 2022 December 2023 were retrospectively included their clinical follow-up data collected. multivariate regression analysis screen factors related results two models compared. area under receiver operating characteristic (ROC) curve (AUC), sensitivity specificity calculated evaluate predictive value model. SHAP (SHapley Additive exPlanations) method explain determine impact features on A total 231 whom 128 (55.4%) developed sarcopenia. univariate LASSO (Least Absolute Shrinkage Selection Operator) cross-validated, 5 key study variables ultimately determined: serum albumin, comorbid diabetes, operation style, nutritional score, ECOG (Eastern Cooperative Oncology Group) performance status score. slightly better AUC (0.987, 95% CI: 0.976-0.998) than (0.918, 0.873-0.963) training set. showed that model, albumin have greater after surgery, especially diabetes score most significant, followed style least impact. summary, learning-based constructed provides valuable decision support tool screening

Язык: Английский

---

MATHEMATICAL MODEL AS A METHOD OF EVALUATING THE ACTIVATION OF ONCOGENESIS IN PATIENTS WITH TYPE 2 DIABETES

Clinical and Preventive Medicine, Год журнала: 2025, Номер 2, С. 68 - 73

Опубликована: Апрель 17, 2025

Introduction. Numerous studies have proven the link between diabetes and cancer. Chronic metabolic disorders in type 2 (T2D) cause dysregulation of intracellular systems involved control cell survival, apoptosis, proliferation. The search for markers activation carcinogenesis continues. Aim. To develop a method assessing oncogenesis processes patients with T2D by creating mathematical model that takes into account complex impact activity components insulin signaling PI3K/Akt/mTOR. Materials methods. study 28 T2D, group consisted 16 practically healthy individuals. examination included determining indicators reflect carbohydrate metabolism compensation (glycemia, glycated hemoglobin (HbA1c)), levels growth factors (insulin, IGF-1), activity, such as phospho-PRAS40 phospho-p70S6K. Statistical analysis results was performed using STATISTIKA-12 software (StatSoft Inc., USA) statistical functions package Microsoft Excel. Using obtained data, developed discriminant analysis. Results. In significantly elevated fasting blood glucose, HbA1c, insulin, IGF-1, HOMA-IR were observed. Significantly increased phospho-p70S6K detected peripheral mononuclear cells. A has been created, which allows to be classified two groups: Group 1 – hyperactivation signaling, or without hyperactivation. most significant are levels, index, HbA1c. Conclusions. Discriminant proves importance comprehensive approach assessment taking compensation, sensitivity, cascade. confirms statistically influence hyperglycemia, hyperinsulinemia, resistance activating pathway.

Язык: Английский

---

Recent Progress in Saliva-Based Sensors for Continuous Monitoring of Heavy Metal Levels Linked with Diabetes and Obesity

Chemosensors, Год журнала: 2024, Номер 12(12), С. 269 - 269

Опубликована: Дек. 19, 2024

Sensors are versatile technologies that provide rapid and efficient diagnostic results, making them invaluable tools in public health for measuring monitoring community exposure to environmental contaminants. Heavy metals such as lead, mercury, cadmium, commonly found food water, can accumulate the body have toxic effects, contributing development of conditions like obesity diabetes. Traditional methods detecting these often require invasive blood samples; however, sensors utilize saliva, offering a noninvasive simplified approach screening. The use saliva fluid represents major advance population due its low cost, noninvasiveness, ease collection. Recent advances sensor technology enabled tests link heavy metal levels with risk developing Optimizing could facilitate identification individuals or groups at risk, enabling targeted, personalized preventive measures. hold promise diagnosing preventing metabolic diseases, providing valuable insights into between health.

Язык: Английский

---