Fabián Segovia‐Miranda

NucVerse3D is a deep-learning framework for 3D nuclear instance segmentation across imaging modalities, achieving high accuracy and outperforming existing methods in dense and irregular settings.

TiMiGNet reconstructs 3D tissue microstructure across species using GANs, enabling high-resolution predictions without paired images.

This paper demonstrates how deep learning-based image restoration extends fluorescence microscopy's capabilities by improving resolution, reducing photon exposure, and increasing frame rates.