Identification of usual interstitial pneumonia pattern using RNA-Seq and machine learning: challenges and solutions

Synopsis

The methods paper behind the RNA-sequencing Envisia classifier, addressing limited sample size, heterogeneity and batch effects. The models classified usual interstitial pneumonia on transbronchial biopsy samples with a receiver-operating characteristic area under the curve of approximately 0.9 in cross-validation; the final model showed 70% sensitivity and 88% specificity in the test set. Kennedy is ninth of ten authors.

In the career genome

• VeracyteChapter 09 · Strand B · Research

  Veracyte, Inc.

  Percepta and Envisia

  Chief Scientific Officer and Chief Medical Officer, Veracyte

  APR 2015 – APR 2024