Nathan R Huber

Title: Nathan R. Huber and Medical Image Enhancement

Introduction

Nathan R. Huber is an inventor based in Saint Peter, Minnesota. His patent work includes a method for training artificial intelligence systems to reduce noise and artifacts in medical images. He is associated with the Mayo Foundation for Medical Education and Research.

Latest Patents

Huber is named on 1 patent: “Generalizable Image-Based Training Framework for Artificial Intelligence-Based Noise and Artifact Reduction in Medical Images.” The invention describes a Generalized Noise and Artifact Reduction Network, or GARNET, for training a convolutional neural network or another suitable machine-learning system.

The method uses noise and artifact patterns derived from phantom images to synthetically corrupt medical images. The system is trained on paired corrupted and uncorrupted images, then applied to routine medical images to reduce noise and artifacts and improve image quality.

Career Highlights

Huber’s listed patent reflects work at the intersection of medical imaging and artificial intelligence. The approach aims to make image enhancement training more generalizable by using simulated examples to teach a neural network how to address both noise and artifacts.

Collaborations

The patent lists Shuai Leng and Cynthia H. McCollough among Huber’s coworkers. Their shared work connects the invention to medical imaging research at the Mayo Foundation for Medical Education and Research.

Conclusion

Nathan R. Huber’s patent describes an AI-based framework for improving medical images by training neural networks to reduce noise and artifacts.