Digital Inline Holographic Microscopy (DIHM) is a potent, non-invasive method for analyzing and characterizing biological tissues, including unstained Basal Cell Carcinoma (BCC) tissues. Digital inline hologram reconstruction artifacts often compromise the accuracy of quantitative information derived from complex data. This work proposes a Constrained Anisotropic Total Variation (CATV) technique to enhance holographic reconstruction quality, incorporating sparsity prior, support, and physical constraints. The Alternating Direction Method of Multipliers (ADMM) solver addresses this regularized inverse problem. The constrained compressed sensing framework offers twin-free reconstruction, noise robustness, and expedited convergence for the numerical reconstruction of complex-valued objects. The validation of the introduced approach involved the reconstruction of synthetic and experimental holograms. The artifact-free phase reconstruction of BCC holograms reveals insightful features such as refractive index variations, tumor islands, palisading, clefting, and mitotic figures, thereby advancing the understanding of BCC tissues and demonstrating the efficacy of the proposed methodology.