Login / Signup

Noninvasive molecular subtyping of pediatric low-grade glioma with self-supervised transfer learning.

Divyanshu TakZezhong YeAnna ZapaischykovaAidan BoydSridhar VajapeyamRishi ChopraYining ZhaHasaan HayatSanjay P PrabhuKevin X LiuHesham ElhalawaniAli NabavidazehAriana FamiliarAdam ResnickSabine MuellerHugo J W L AertsPratiti BandopadhayayKeith LigonDaphne Haas-KoganTina PoussaintBenjamin H Kann
Published in: medRxiv : the preprint server for health sciences (2023)
An innovative training approach combining self-supervision and transfer learning ("TransferX") is developed to boost model performance in low data settings;TransferX enables the development of a scan-to-prediction pipeline for pediatric LGG mutational status (BRAF V600E, fusion, or wildtype) with high accuracy and mild performance degradation on external validation;An evaluation metric "COMDist" is proposed to increase interpretability and quantify the accuracy of the model's attention around the tumor.
Keyphrases
  • low grade
  • high grade
  • machine learning
  • computed tomography
  • working memory
  • big data
  • magnetic resonance imaging
  • young adults
  • single molecule
  • deep learning
  • data analysis
  • metastatic colorectal cancer
  • contrast enhanced