Optical computed tomography (OCT) is often used for measuring thermophysical parameters. Unfortunately, viewing access in many tomographic experiments, such as in plasma physics, is extremely limited, which leads to a highly undetermined inversion problem. Four major approaches to this problem are compared: a novel algorithm based on maximum entropy (ME), and three traditional algorithms, namely the simultaneous iterative reconstruction technique (SIRT), algebraic reconstruction technique (ART), and multiplicative algebraic reconstruction technique (MART). A novel basis function was adopted to improve the performance of SIRT, ART, and MART. These algorithms were used to reconstruct phantom data with realistic levels of noise from a number of different imaging geometries. The phantoms, the imaging geometries, and the noise were chosen to simulate conditions encountered in typical OCT measurement. In most cases, the ME and MART algorithms give better reconstruction results than the other two algorithms for the situations studied here.