Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 30, 2026Journal of Clinical Oncology

Machine learning model for lung cancer risk stratification using routine complete blood count exams.

View Full Paper
Ask AI
Bookmark
Share

Authors

DADaniella AraújoMicrosoft (Brazil)BRBruno Aragão RochaFleury S.A. (Brazil)SLSuzylaine da Silva LimaMicrosoft (Brazil)

Discussion

Loading...

Member takes

Overview

Retrospective study develops machine learning tool for lung cancer risk assessment in older adults, suggesting increased screening efficiency.

Key Points

  • This research aims to create a machine learning model for lung cancer risk stratification using routine complete blood count tests.
  • Retrospective analysis of CBC tests from 53,093 individuals aged 50 and older
  • Used high-risk CT findings (n=1,178) for model training and biopsy-confirmed cases (n=141) for final evaluation
  • A ridge regression model was trained using selected CBC features
  • Model achieved an AUC of 0.71 (95% CI: 0.70–0.71) for overall population
  • In smokers, model performance was comparable with an AUC of 0.68 (95% CI: 0.67-0.68)
  • CBC parameters such as neutrophil count and RDW showed significant differences between high-risk cases and low-risk controls (p < 0.001)

Cite This Study

Araújo et al. (2026) studied this question.

synapsesocial.com/papers/6a1a827f0307b7850943426ahttps://doi.org/10.1200/jco.2026.44.16_suppl.e20005
View Full Paper
Ask AI
Bookmark
Share