In plain words: A three-step system scans 3D lung CAT images to find suspicious nodules, judges each one as cancerous or not, then combines those calls into a single cancer probability. In a 2017 lung cancer prediction contest, it finished 41st out of 1,972 teams.

Abstract · Deep Learning for Lung Cancer Detection: Tackling the Kaggle Data Science Bowl 2017 Challenge

We present a deep learning framework for computer-aided lung cancer diagnosis. Our multi-stage framework detects nodules in 3D lung CAT scans, determines if each nodule is malignant, and finally assigns a cancer probability based on these results. We discuss the challenges and advantages of our framework. In the Kaggle Data Science Bowl 2017, our framework ranked 41st out of 1972 teams.

Kingsley Kuan, Mathieu Ravaut, Gaurav Manek, Huiling Chen, Jie Lin, Babar Nazir, Cen Chen, Tse Chiang Howe, Zeng Zeng, Vijay Chandrasekhar
arXiv:1705.09435 · cs.CV · submitted May 26, 2017
abstract · pdf · html