Examining specific sub-cellular structures while minimizing cell perturbation is important in the life sciences. Fluorescence labeling and imaging is widely used for introducing specificity despite its perturbative and photo-toxic nature. With the advancement of deep learning, digital staining routines for label-free analysis have emerged as a replacement for fluorescence imaging. Nonetheless, digital staining of subcellular structures such as mitochondria is sub-optimal. This is because the models designed for computer vision are directly applied instead of optimizing them for the nature of microscopy data. We propose a new loss function with multiple thresholding steps to promote more effective learning for microscopy data. Throigh this, we demonstrate a deep learning approach to translate the label-free brightfield images of living cells into equivalent fluorescence microscopy images of mitochondria with an average structural similarity of 0.77, thus surpassing the state-of-the-art of 0.7x. Our results provide insightful examples of some unique opportunities generated by data-driven deep-learning enabled image translations.