PtychoShelves, a versatile high-level framework for high-performance analysis of ptychographic data

Area of Science:

• Microscopy
• Computational Imaging
• Materials Science

Background:

• Ptychography is a powerful technique for non-destructive, high-resolution imaging using electrons, X-rays, and light.
• It enables quantitative reconstruction of specimen transmissivity and recovery of the illuminating wavefront.
• Increasingly complex algorithms and faster microscopes generate large datasets, posing computational challenges.

Purpose of the Study:

• To introduce PtychoShelves, a novel software package designed for advanced ptychography data processing.
• To provide a modular and high-performance computing solution for handling large-scale ptychographic reconstructions.

Main Methods:

• Development of a modular software architecture for flexible data-processing pipeline customization.
• Implementation of high-performance computing capabilities utilizing both CPUs and GPUs for accelerated reconstructions.

Main Results:

• PtychoShelves facilitates efficient processing of complex ptychographic data.
• The software supports high-resolution quantitative imaging and wavefront recovery.
• Modularity allows for rapid adaptation and improvement of data-processing algorithms.

Conclusions:

• PtychoShelves addresses the growing computational demands of modern ptychography.
• The software package offers a scalable and efficient solution for researchers in quantitative microscopy.
• It enables faster and more accessible high-resolution imaging across various microscopy modalities.