逆問題解法による光CTのシミュレーション

概要

論文の詳細を見る
Optical computed tomography is a medical imaging modality in which red and near infrared light are used to probe biological tissues. Images are obtained by mapping the absorption coefficient of the tissue. Realization of clinically viable optical computed tomography depends on the development of reliable and fast methods of obtaining inverse solutions of the photon diffusion equation which describes the photon migration in biological tissues. In this paper we use a linearization and regularization technique and a finite element method to invert the diffusion approximation, and we propose the use of a zooming method as a way to improve the resolution of the images. We compare the performance of three regularization methods, i. e., the truncated singular value decomposition, Tikhonov regularization, and Phillips-Twomey regularization, The usefulness of the zooming method is demonstrated.
一般社団法人日本機械学会の論文
1996-03-25