A machine learning tool to improve prediction of mediastinal lymph node metastases in non-small cell lung cancer using routinely obtainable [F]FDG-PET/CT parameters.

Background: In patients with non-small cell lung cancer (NSCLC), accuracy of [F]FDG-PET/CT for pretherapeutic lymph node (LN) staging is limited by false positive findings. Our aim was to evaluate machine learning with routinely obtainable variables to improve accuracy over standard visual image assessment.

Methods: Monocentric retrospective analysis of pretherapeutic [F]FDG-PET/CT in 491 consecutive patients with NSCLC using an analog PET/CT scanner (training + test cohort, n = 385) or digital scanner (validation, n = 106).

FDG-PET/CT for pretherapeutic lymph node staging in non-small cell lung cancer: A tailored approach to the ESTS/ESMO guideline workflow.

Objectives: In patients with NSCLC, current ESTS and ESMO guidelines recommend invasive lymph node (LN) staging with EBUS-TBNA even if FDG-PET/CT is negative for mediastinal LNs if at least one of three risk factors is present (cN1, non-peripheral primary or primary >3 cm). Modified workflows to avoid unnecessary invasive procedures were evaluated.

Materials And Methods: Monocentric retrospective analysis of pretherapeutic FDG-PET/CT in 247 patients with NSCLC (62 % male; age, 68 [43-88] years) using an analog or digital PET/CT scanner.