TY - GEN
T1 - Diffuse optical tomography image reconstruction based on sparse recovery methods
AU - Sun, Hao Jan
AU - Hou, Hsiang Wen
AU - Chou, Chia Ching
AU - Fang, Wai-Chi
PY - 2015/12/4
Y1 - 2015/12/4
N2 - Diffuse Optical Tomography (DOT) is a non-invasive image detecting technique. It is used to assess spatial variation with absorption and scattering coefficients for tumor detection, distribution of oxygen concentration analysis, oxygenated hemoglobin concentration measurement or deoxygenated hemoglobin concentration measurement. In this paper, we use sparse recovery methods for DOT image reconstruction. Using the non-linear iterative method to reconstruct the DOT image can increase the resolution of each reconstructed layer. Sparse recovery methods use the p-norm regularization in the estimation problem with 0 < p <= 1. When the number of independent measurements is limited by nature, which is a typical case for diffuse optical tomographic image reconstruction, sparse recovery methods present good performance. According to the simulation results the reconstruction algorithm can parse the tumor clearly with p=0.6. The projection errors are 0.45357, 0.44588, 0.46781, 0.44109, and 0.47174(tumor / background) in each tumor position case using p=0.6. Since that the projection errors with p=0.6 are smaller than projection errors using p=1 or p=2, p=0.6 is selected to be the reconstructed parameter to have the best reconstructed results for the same iteration numbers. Simulation results show that the sparse recovery methods are good to improve the reconstructed image quality.
AB - Diffuse Optical Tomography (DOT) is a non-invasive image detecting technique. It is used to assess spatial variation with absorption and scattering coefficients for tumor detection, distribution of oxygen concentration analysis, oxygenated hemoglobin concentration measurement or deoxygenated hemoglobin concentration measurement. In this paper, we use sparse recovery methods for DOT image reconstruction. Using the non-linear iterative method to reconstruct the DOT image can increase the resolution of each reconstructed layer. Sparse recovery methods use the p-norm regularization in the estimation problem with 0 < p <= 1. When the number of independent measurements is limited by nature, which is a typical case for diffuse optical tomographic image reconstruction, sparse recovery methods present good performance. According to the simulation results the reconstruction algorithm can parse the tumor clearly with p=0.6. The projection errors are 0.45357, 0.44588, 0.46781, 0.44109, and 0.47174(tumor / background) in each tumor position case using p=0.6. Since that the projection errors with p=0.6 are smaller than projection errors using p=1 or p=2, p=0.6 is selected to be the reconstructed parameter to have the best reconstructed results for the same iteration numbers. Simulation results show that the sparse recovery methods are good to improve the reconstructed image quality.
KW - diffuse optical tomography (DOT)
KW - image reconstruction
KW - sparse recovery methods
UR - http://www.scopus.com/inward/record.url?scp=84962627670&partnerID=8YFLogxK
U2 - 10.1109/BioCAS.2015.7348439
DO - 10.1109/BioCAS.2015.7348439
M3 - Conference contribution
AN - SCOPUS:84962627670
T3 - IEEE Biomedical Circuits and Systems Conference: Engineering for Healthy Minds and Able Bodies, BioCAS 2015 - Proceedings
BT - IEEE Biomedical Circuits and Systems Conference
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th IEEE Biomedical Circuits and Systems Conference, BioCAS 2015
Y2 - 22 October 2015 through 24 October 2015
ER -