Lingsong Zhang

Lingsong Zhang

Lingsong Zhang; Associate Professor, Statistics (University of North Carolina at Chapel Hill; Statistics; 2007)

Holding a unique joint appointment between Purdue Statistics and Regenstrief Center for Healthcare Engineering, Dr. Lingsong Zhang combines his expertise in statistics methodology with enthusiasm for collaborative research in healthcare engineering.  His history in statistics began with a Bachelor of Science from Peking University in Statistics and a Master of Science degree from Tsinghua University.  He then continued with a PhD from the University of North Carolina, Chapel Hill in Statistics. 

When first introduced to the study of statistics, he fell in love with the real-life examples and applications of statistics and therefore decided to make it the focus of his education. One early example he remembers is that statisticians helped astronomers find the rings of Uranus.  His curiosity and enthusiasm for the applications of statistics motivated him to have experience in several domains including accounting, computer science, epidemiology, and healthcare.

After his time at UNC Chapel Hill studying under his advisor Professor J. S. Marron, he continued with a Postdoc at Harvard University.  His advisor there was Professor Xihong Lin, and they focused their research on developing machine learning methods and applying them to epidemiology; he applies these same methods in his research today.

His statistical expertise focuses on data exploration and classification, especially when data is complicated and/or has special structure. He has developed many tools to help with data visualization. These tools also help with the identification of outliers and revealing major modes of variations.

In addition, he focuses on scale-space inference, a novel approach for performing statistical analysis.  Instead of directly applying statistical methods to a data at the scale collected (e.g. time interval or measurement unit), the statistical analysis is performed on multiple scales and then all inference results are combined to draw a more comprehensive conclusion.  This multiple-scale approach has great potential to dramatically improve the power of many existing statistical methods. 

With respect to Dr. Zhang’s contributions in Healthcare Engineering, he has specific experience in developing new statistical methodology to address reducing hospital readmissions, experimental design, method validation, data integration, de-identification, and preprocessing data.  With his position at Regenstrief Center for Healthcare Engineering, he has collaborated with experts in nursing, industry engineering, animal studies, specialty doctors on diabetic management, lactation, kinship parenting, etc.

He is focused on creating statistical methods for real applications, and exploration of their theoretical and empirical properties.   As a part of his joint appointment, he enjoys the collaborative nature of his role.  His collaborations focus on a common goal working on it from different viewpoints, which often bring new thoughts and lead to innovative solutions.  Currently he is involved in the following collaborative projects:

  • Sleep and Obesity during Pregnancy: Impacts on Breastfeeding success among Indiana’s Rural Mothers- This is an AgSeed grant within Purdue University.  The PI is Professor Azza Ahmend of the Nursing School along with Professor Theresa Casey from the Department of Animal Study as a co-investigator; Dr. Zhang is assessing how a mother’s circadian rhythm during pregnancy impacts their postpartum breastfeeding.
  • Enhancing Kinship Families’ Health in Rural Counties through Extension Education- This is a grant from the US Department of Agriculture and the National Institute of Food and Agriculture.  The PI is Professor Karen Foli of the Nursing School; Dr. Zhang is a co-investigator.  They are in the data collection stage of the project focusing on developing curriculum to kinship families that try to improve health outcomes for those children.

Dr. Lingsong Zhang’s research program is focused on many aspects of statistical inference and methodology concentrating on real-world applications.  He welcomes new Purdue collaborations and can be reached via email: