Techn. Fakultät Website deprecated and outdated. Click here for the new site. FAU-Logo

Yixing Huang M. Sc.

Researcher in the Precision Learning (PL) group at the Pattern Recognition Lab of the Friedrich-Alexander-Universität Erlangen-Nürnberg

Yixing Huang

Yixing Huang (黄益星)

  • 2016.6 –                PhD at Pattern Recognition Lab, Department of Computer Science 5, FAU, Erlangen, Germany
  • 2014.3 –  2016.5    Master of Medical Engineering, Department of Computer Science 5, FAU, Erlangen, Germany
  • 2013.7 –  2014.2    Research assistant at Dept. of Biomedical Engineering, College of Engineering, Peking University, China.
  • 2009.9 –  2013.7    Bachelor at Dept. of Biomedical Engineering, College of Engineering, Peking University, China.
  • 2006.9 –  2009.6    Rong Huai High School in Zhuji City, Zhejiang Province, China
  • 2003.9 –  2006.6    Rong Huai Middle School in Zhuji City, Zhejiang Province, China

PhD Research (FAU, Erlangen) -- Limited Angle Tomography

Thanks to the special collaboration program among Siemens Healthcare, Pattern Recognition Lab at FAU, and Peking University in China, I could already start my research on limited angle tomography when I was a master student. This research project is directly supported by Innovation Department, Advanced Therapies, Siemens Healthcare. During my PhD research, I have investigated analytic reconstruction algorithms using data consistency conditions as well as iterative reconstruction algorithms with total variation regularization. Besides, I also gained a lot of experience in the field of machine learning, including conventional machine learning like artificial neural networks and decision tree, and the cutting-edge deep learning technology like the U-Net and adversarial examples. 

Master's Thesis (FAU, Erlangen) -- Limited Angle Reconstruction Based on Data Consistency in Sinogram Domain

At this thesis, first we investigated a Fourier curve fitting method to restore the missing data for limited angle tomography. It works well for the Shepp-Logan phantom. However, it is suited for clinical data, which have complex strutures. In addition, we also investigated the application of data consistency conditions for missing data restoration. In the projection data of computed tomograpy, many constraints exist. They are typically mathematically expressed as consistency conditions, e.g., the well-know Helgason-Ludiwg consistency conditions. Our proposed algorithm can correct intensity bias and reduce most artifacts for the Shepp-Logan phantom and a brain image as well in parallel-beam limited angle tomography.

Research Assistant (PKU, Beijing) -- Geometry Calibration and FDK Reconstruction Algorithm Implementation for a Micro-CT System

During the gap semester waiting for the necessary documents to come to Germany, I worked as a research assistant at the multi-modality medical imaging lab at Department of Biomedical Engineering, Peking University, China. Collaborating with other colleagues, we set up a flat-panel cone-beam mico-CT system for the imaging of small animals. We designed a phantom for the geometry calibration of the micro-CT system, which is patented as CN103654730 A. In addition, I also implemented the FDK reconstruction algorithm.

Bachelor's Thesis (PKU, Beijing)——A New Fluorescence Molecular Tomography System with a Changeable LED Array Module

My Bachelor's thesis was the study the challenges of Light Emitting Diodes (LED) as illuminant source in Fluorescence Molecular Tomography (FMT). A fluorescence molecular imaging system was set up from pieces of optical parts under the supervision of Yichen Ding, a PhD candidate at PKU. I did a series of experiments, from phantom experiments to animal experiments, to demonstrate the efficiency of LED arrays as illuminant source in the FMT system.