---

Michael I. Miller Director of the Center for Imaging Science The Herschel and Ruth Seder Professor of Biomedical Engineering Professor of Electrical and Computer Engineering Affiliations: Dept of Biomedical Engineering, Johns Hopkins University, Dept of Electrical and Computer Engineering, Johns Hopkins University Dept of Applied Mathematics and Statistics, Johns Hopkins University Dept of Computer Science, Johns Hopkins University Publications Curriculum Vitae

---

Center for Imaging Science 301 Clark Hall Johns Hopkins University 3400 N. Charles Street Baltimore, MD 21218

Email address: mim@cis.jhu.edu Office: Clark Hall 301 Phone: (410) 516-3826 Fax: (410) 516-4594

---

An Interview With Dr. Michael Miller at In-Cites.Com:

In this interview, in-cites correspondent Gary Taubes speaks with Dr. Michael I. Miller of Johns Hopkins University in Baltimore, Maryland, about his highly cited work in computational anatomy. In a recent update of ISI Essential Science Indicators, Dr. Miller garnered the highest increase in total citations in the field of Engineering. Dr. Miller is Director of the Center for Imaging Science as well as Professor of Biomedical Engineering and Electrical and Computer Engineering.

---

Current Graduate Students

• Dmitri Bitouk
• Marc Vaillant
• Can Ceritoglu
• Nayoung Lee
• Jun Ma
• Anqi Qiu
• Hasan Cetingul
• Jing Yu

---

Research Interests

• Pattern Theory
• Computational Linguistics
• Computational Neuroscience

---

Research Projects

Image Understanding/Automated Target Recognition

Prof. Michael Miller, Prof. Donald Snyder , Prof. Joseph O'Sullivan, Prof. Ulf Grenander

A major thrust of our work in image understanding/automated target recognition centers on representation of rigid bodies and their invariants via the study of Lie groups, and their actions on the rigid bodies. We have been focusing on air base targets imaged with high resolution radar and optical sensors, and ground targets imaged with FLIR sensors.

Computational Anatomy

Prof. Michael Miller, Prof. Ulf Grenander

The study of human anatomy presents the most difficult challenges to the understudy of typicality and variablity. While Biological shapes are highly structured, they are not rigid. We have been using Grenander's deformable anatomical templates for the representation of the typicality and for the representation of the variablilty. For this, complex anatomical templates (human and macaque brains) are annotated with coordinate systems defined within them. High-dimensional vector fields applied to these coordinate systems carry the templates with all of its geometry into the target. This allows for understanding modulo individual variation.

---

Last modified: Monday, 22-May-2006 10:26:18 EDT