Machine learning-predicted insulin resistance is a risk factor for 12 types of cancer - PubMed
6 days ago
- #cancer risk
- #insulin resistance
- #machine learning
- Machine learning-predicted insulin resistance (AI-IR) is linked to an increased risk of 12 types of cancer.
- AI-IR outperforms BMI, metabolic syndrome (MetS), TG/HDL ratio, and TyG index in predicting diabetes incidence.
- Significant associations found between AI-IR and six cancers: uterine, kidney, esophagus, pancreas, colon, and breast.
- Nominal associations observed with six additional cancers: renal pelvis, small intestine, stomach, liver and gallbladder, leukemia, and bronchial and lung.
- Composite cancer risk increases with AI-IR, showing a hazard ratio of 1.25 (95% CI: 1.20-1.31).
- AI-IR is comparable to MetS and TG/HDL ratio in predicting composite cancers but better than BMI and TyG index.
- AI-IR is proposed as a robust metric for identifying high-risk individuals for diabetes and cancer screening.