Pulmonary Nodule Malignancy Classification Using its Temporal Evolution with Two-Stream 3D Convolutional Neural NetworksDownload PDF

25 Jan 2020, 12:34 (modified: 22 Jul 2022, 19:50)MIDL 2020Readers: Everyone
Abstract: Nodule malignancy assessment is a complex, time-consuming and error-prone task. Current clinical practice requires measuring changes in size and density of the nodule at different time-points. State of the art solutions rely on 3D convolutional neural networks built on pulmonary nodules obtained from a single CT scan per patient. In this work, we propose a two-stream 3D convolutional neural network that predicts malignancy by jointly analyzing two pulmonary nodule volumes from the same patient taken at different time-points. Best results achieve 77% of F1-score in test with an increment of 9% and 12% of F1-score with respect to the same network trained with images from a single time-point.
Paper Type: both
TL;DR: A two-stream 3D CNN that predicts malignancy by jointly analyzing two pulmonary nodule volumes from the same patient taken at different time-points
Track: short paper
Keywords: Lung Cancer, Pulmonary Nodule Malignancy, Convolutional Neural Networks
Presentation Upload: zip
Presentation Upload Agreement: I agree that my presentation material (videos and slides) will be made public.
6 Replies

Loading