High‐Fidelity Computational Microscopy via Feature‐Domain Phase Retrieval (Adv. Sci. 21/2025)

High-Fidelity Computational Microscopy In article number 2413975 by Liangcai Cao and co-workers, a feature-domain phase retrieval framework is presented for high-fidelity computational microscopy. The quantitative phase imaging can be realized with enhanced accuracy and robustness. By leveraging feature pyramid extracted from the image, the proposed approach achieves improved noise suppression and resolution, making it ideal for biomedical applications such as cell morphology analysis and high-resolution microscopy.