The aggregation of misfolded proteins into amyloid fibrils is a hallmark of several neurodegenerative diseases. Cryo-electron microscopy (cryo-EM) is a powerful imaging technique that enables the visualization of these fibrils at high resolution. In this project, we propose to develop a machine learning-based approach to de-noise cryo-EM refined maps of amyloid fibrils. Specifically, we will use MATLAB to implement a deep neural network that is capable of learning the underlying structure of the fibrils from noisy data. The network will be trained on a dataset of simulated noisy cryo-EM images of amyloid fibrils, along with their corresponding high-resolution maps. Overall, this project has the potential to significantly improve the accuracy and resolution of cryo-EM maps of amyloid fibrils, which could lead to a better understanding of the structural basis of neurodegenerative diseases. Additionally, the machine learning-based approach developed in this project can be extended to other cryo-EM imaging applications, where de-noising is a critical step for accurate reconstruction of the three-dimensional structure of biological macromolecules.