Automatic Detection and Segmentation of Lung Lesions on CT scans using Deep Convolutional Neural Networks

11/15/2019

Lung cancer is the most pervasive and one of the deadliest types of cancer, leading to more than 2 million cases being diagnosed each year, and a mortality rate of 60%. CT screening trials have played a key role in improving early detection of lung cancer, which has shown to significantly improve patient survival. Apart from lesion detection, tumour segmentation is critical for developing radiomics signatures. In this work, we propose a novel hybrid approach for lung lesion detection and segmentation on CT scans, where the segmentation task is assisted by prior detection of regions containing lesions. For the detection task, we introduce a 2.5D residual deep CNN working in a sliding-window fashion, whereas segmentation is tackled by a modified residual U-Net with a weighted-dice plus cross-entropy loss. Experimental results on the LIDC-IDRI dataset and on the lung tumour task dataset within the Medical Segmentation Decathlon show competitive detection performance of the proposed approach (0.902 recall) and superior segmentation capabilities (0.709 dice score). Further validation of the models was also performed, with key components of both models tested through several ablation studies, in order to assess its contribution to the final models. These results confirm the high potential of simpler models, with lower hardware requirements, thus of more general applicability.

Authors in the community:

• 

  J

  João Francisco Lourenço Borges de Sá Carvalho

  ist181397

Supervisors of this institution:

• 

  M

  Mário Alexandre Teles de Figueiredo

  ist12403

industrial-biotechnology - Industrial Biotechnology

eng - English

Instituto Superior Técnico