Biomedical discoveries have been largely driven by advances in imaging, esp. microscopy at the cellular level, due to its unique capacity of capturing cell in action in the tissues. While most discoveries have been made by microscopy in dead, sliced, and labeled tissues, capturing biological processes alive and label-free is highly desirable as it adds the spatiotemporal dimension of information and mitigates the confirmation bias of existing labels. However, live-tissue label-free pathology is challenging because 1) intrinsic contrast is weak and few, 2) light diffracts and scatters, which limits imaging to shallow imaging depth, limited tissue contrast, and low throughput. In this talk, we will dive into a few mechanisms we can drastically increase the depth, contrast, and throughput of live-tissue pathology for biomedical applications, with a focus on metabolic imaging of living systems. We will also discuss their implications on potential breakthroughs in understanding and treatment of prevalent, but complex and poorly understood human diseases.