This inventor holds 1 USPTO granted patent. Top assignee: Tsinghua University. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Haozhi Sha and Computational Electron Microscopy
Introduction
Haozhi Sha is an inventor based in Beijing, China, and is associated with Tsinghua University. His patent work addresses computational methods for reconstructing atomic spatial distributions and electron-beam characteristics from diffraction measurements.
Latest Patents
The provided information lists one patent for Haozhi Sha: “Method and Apparatus for Reconstructing Atomic Spatial Distribution and Electron Beam Function.” The disclosed method collects diffraction intensities as an electron beam scans a sample, then initializes a sample transmission function and an electron beam function. It uses these inputs to establish a forward propagation model with parameters to optimize and calculates a loss function. Gradients of the loss function guide updates to the parameters, and the process repeats until a termination condition is met. The method then outputs optimized sample transmission and electron beam functions. 1 patent
Career Highlights
Sha’s listed invention focuses on improving the reconstruction of sample structure and electron-beam properties through iterative optimization. The approach combines scanning measurements, a propagation model, and gradient-based parameter updates.
Collaborations
The provided information identifies Rong Yu and Wenfeng Yang as Sha’s coworkers. It does not specify their roles in the patent or the nature of their collaboration.
Conclusion
Haozhi Sha’s listed patent presents an iterative computational approach to reconstructing atomic spatial distributions and electron-beam functions from diffraction data, reflecting his association with research at Tsinghua University.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
