Applications of Deep Learning for the Automatic or Semi-automatic Detection of Lung Abnormalities

Principal Investigator

Name
Suryadipto Sarkar

Degrees
B.Tech (currently pursuing PhD)

Institution
Arizona State University

Position Title
Graduate Research Assistant

Email
ssarka34@asu.edu

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-679

Initial CDAS Request Approval
Jun 22, 2020

Title
Applications of Deep Learning for the Automatic or Semi-automatic Detection of Lung Abnormalities

Summary
I want to explore the applicability of deep learning in the effective segmentation of classification of lesions from lung images. I intend to first use pre-trained models like GoogLeNet, AlexNet, VGGNet et cetera, on the images. Then I intend to design my own CNN classifier to see if I can improve the classification accuracy any further.

Aims

- Applicability of deep learning in classification and segmentation of abnormalities in lung images
- Transfer learning
- Few-shot learning

Collaborators

None