Data acquisition and processing is a critical issue for high-speed applications, especially in three-dimensional live cell imaging and analysis. This paper focuses on sparse-data sample rotation tomographic reconstruction and analysis with several noise-reduction techniques. For the sample rotation experiments, a live Candida rugosa sample is used and controlled by holographic optical tweezers, and the transmitted complex wavefronts of the sample are recorded with digital holographic microscopy. Three different cases of sample rotation tomography were reconstructed for dense angle with a step rotation at every 2°, and for sparse angles with step rotation at every 5° and 10°. The three cases of tomographic reconstruction performance are analyzed with consideration for data processing using four noise-reduction techniques. The experimental results demonstrate potential capability in retaining the tomographic image quality, even at the sparse angle reconstructions, with the help of noise-reduction techniques.