Correlation of histologic, imaging, and artificial intelligence features in NAFLD patients, derived from Gd-EOB-DTPA-enhanced MRI: a proof-of-concept study.
Nina BastatiMatthias PerkoniggDaniel SobotkaSarah Poetter-LangRomana FragnerAndrea BeerAlina MessnerMartin WatzenboeckSvitlana PochepniaJakob KittingerAlexander HeroldAntonia KristicJacqueline C HodgeStefan TraussnigMichael TraunerAhmed Ba-SsalamahGeorg LangsPublished in: European radiology (2023)
• Unsupervised deep clustering (UDC) and MR-based parameters (FF and RLE) could independently distinguish simple steatosis from NASH in the derivation group. • On multivariate analysis, RLE could predict only fibrosis, and FF could predict only steatosis; however, UDC could predict all histologic NAFLD components in the derivation group. • The validation cohort confirmed the findings for the derivation group.
Keyphrases
- artificial intelligence
- machine learning
- end stage renal disease
- insulin resistance
- big data
- high fat diet
- contrast enhanced
- deep learning
- chronic kidney disease
- magnetic resonance imaging
- high resolution
- prognostic factors
- peritoneal dialysis
- computed tomography
- high fat diet induced
- adipose tissue
- single cell
- metabolic syndrome
- skeletal muscle
- patient reported