A feasibility study on the improvement of CT image with the back propagation neural network

Shih-Chieh Lin; Tse-Li Wang

doi:10.1117/12.2015212

31 January 2013 A feasibility study on the improvement of CT image with the back propagation neural network

Shih-Chieh Lin, Tse-Li Wang

Proceedings Volume 8759, Eighth International Symposium on Precision Engineering Measurement and Instrumentation; 87590H (2013) https://doi.org/10.1117/12.2015212
Event: International Symposium on Precision Engineering Measurement and Instrumentation 2012, 2012, Chengdu, China

Abstract

It had been a major concern about multi-slice X-ray CT for its high radiation dose delivered to a patient. In order to reduce the radiation dose, one can either limit the dose per projection, or reduce the number of projections, or both. However, it was shown that artifact will appear when limited projections were used. In this study, the feasibility of using back propagation type artificial neural network to improve the image reconstructed using the filtered back projection is studied. Two networks were trained to reconstructed image by input information calculated using the filtered back projection method from 32, and 64 projections respectively. A series tests are also conducted to evaluate the performance of the trained networks. The results show that if information of 32 or 64 projections was used, the reconstructed images are generally improved by the use of the trained network.

Citation Download Citation

Shih-Chieh Lin and Tse-Li Wang "A feasibility study on the improvement of CT image with the back propagation neural network", Proc. SPIE 8759, Eighth International Symposium on Precision Engineering Measurement and Instrumentation, 87590H (31 January 2013); https://doi.org/10.1117/12.2015212

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available