Share Email Print
cover

Proceedings Paper • new

Bayesian reconstruction of ultralow-dose CT images with texture prior from existing diagnostic full-dose CT database
Author(s): Yongfeng Gao; Jiaxing Tan; Hao Zhang; William H. Moore M.D.; Priya Bhattacharji; Amit Gupta; Haifang Li; Hongbing Lu; Jianhua Ma Sr.; Zhengrong Liang
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Markov random field (MRF) has been widely used to incorporate a priori knowledge as a penalty for regional smoothing in ultralow-dose computed tomography (ULdCT) image reconstruction, while the regional smoothing does not explicitly consider the tissue-specific textures. Our previous work showed the tissue-specific textures can be enhanced by extracting the tissue-specific MRF from the to-be-reconstructed subject’s previous full-dose CT (FdCT) scans. However, the same subject’s FdCT scans might not be available in some applications. To address this limitation, we have also investigated the feasibility of extracting the tissue-specific textures from an existing FdCT database instead of the to-be-reconstructed subject. This study aims to implement a machine learning strategy to realize the feasibility. Specifically, we trained a Random Forest (RF) model to learn the intrinsic relationship between the tissue textures and subjects’ physiological features. By learning this intrinsic correlation, this model can be used to identify one MRF candidate from the database as the prior knowledge for any subject’s current ULdCT image reconstruction. Besides the conventional physiological factors (like body mass index: BMI, gender, age), we further introduced another two features LungMark and BodyAngle to address the scanning position and angle. The experimental results showed that the BMI and LungMark are two features of the most importance for the classification. Our trained model can predict 0.99 precision at the recall rate of 2%, which means that for each subject, there will be 3390*0.02 = 67.8 valid MRF candidates in the database, where 3,390 is the total number of candidates in the database. Moreover, it showed that introducing the ULdCT texture prior into the RF model can increase the recall rate by 3% while the precision remaining 0.99.

Paper Details

Date Published: 1 March 2019
PDF: 7 pages
Proc. SPIE 10948, Medical Imaging 2019: Physics of Medical Imaging, 1094855 (1 March 2019); doi: 10.1117/12.2512787
Show Author Affiliations
Yongfeng Gao, Stony Brook Univ. (United States)
Jiaxing Tan, The Graduate Ctr., CUNY (United States)
Hao Zhang, Stanford Univ. (United States)
William H. Moore M.D., New York Univ. (United States)
Priya Bhattacharji, New York Univ. (United States)
Amit Gupta, State Univ. of New York (United States)
Haifang Li, State Univ. of New York (United States)
Hongbing Lu, Fourth Military Medical Univ. (China)
Jianhua Ma Sr., Southern Medical Univ. (China)
Zhengrong Liang, Stony Brook Univ. (United States)


Published in SPIE Proceedings Vol. 10948:
Medical Imaging 2019: Physics of Medical Imaging
Taly Gilat Schmidt; Guang-Hong Chen; Hilde Bosmans, Editor(s)

© SPIE. Terms of Use
Back to Top