John Galeotti

I have 6+ years of experience integrating teaching and research. I taught my first graduate course in medical
image analysis algorithms back in Spring 2008, and my teaching has received two teaching awards and
external funding. I have conducted my teaching and my research in mutually beneficial ways, and the synergy
has paid off with better equipped students, new research collaborations, more productive students in my own
lab, and 2 years of combined research and teaching funding from the National Library of Medicine (NLM) at
the National Institutes of Health (NIH). My teaching focus on research and system development requires
that I teach not only the theory and application of current techniques, but also that I help my students gain a
deep understanding of the big-picture concepts and deficiencies in the field, so that they will become well
equipped leaders who can leverage the literature to develop new methods and systems that are actually
relevant to medicine and science. Another key component of my teaching is complexity management,
ranging from the breakdown of algorithms to the design of real-world systems. Together, these constructs
are critical to thinking like a scientist and engineer, and so I explicitly incorporate them as early as possible
into students’ lives, even at the pre-college level, such as when I teach robotics to underrepresented minorities
at Carnegie Mellon University (CMU) in the Summer Academy for Mathematics and Sciences. I am presently
advising/supervising the research of two PhD students (one of whom received an NSF GRF for her research
with me), one MS student, and one undergraduate researcher. This spring I am also the teaching mentor for
an assistant professor through the SYSU-CMU Joint Institute of Engineering. I maximize the public
availability of my teaching materials, and for my methods in medical image analysis class I have a complete set
of teaching materials available online (http://itk.galeotti.net), including lecture slides, assignments, and
professionally produced videos of all my 2012 lectures, which were deliverables for my funding from NLM.
Philosophy & Style: My pedagogical style makes extensive use of class discussions and project work to
foster a deep understanding of core concepts and the skills necessary to manage the complexity of modern
engineered systems. When I initially present important underlying concepts and theoretical relationships, I
prefer to utilize class discussion as the means of presenting the relevant underlying problem, with the goal
that students understand (and appreciate!) why they are being presented with yet another algorithm,
architecture, etc., before I ever put the details of it in front of them. I guide the discussion as necessary to
lead the students naturally from initial understanding of the key aspects of the underlying problem to a critical
analysis of potential solutions. My ultimate goal is for the students to pick up and run with the ideas,
independently re-deriving the overall theory of what I am about to present in detail, thereby walking students
through the process of thinking like an expert scientist or engineer. I typically use quizzes to prompt and
assess students’ initial comprehension of new material, with follow-on mini-project homework assignments
that are typically simplified real-world problems. A large, real-world final project gives students the
experience they need to translate their coursework into the real world, using theory, techniques, and tools to
identify, formulate, and solve scientific and engineering problems. I prefer projects with an element of
interdisciplinary teamwork and collaboration, but with clearly delineated individual goals and responsibilities
so that I can fairly evaluate each student. Projects provide a venue for students to begin looking up and
teaching themselves the more minute details for their chosen methods, helping jump-start the habits of
lifelong learning and literature search. Oral project presentations give students the opportunity (even if just
for 5 minutes) to organize, present, and teach the key components of their projects, providing them critical
experience with conference-style scientific or technical public speaking. I continually modify my course
materials to take into account student feedback, new opportunities, and my own perceptions of course
deficiencies, paying close attention to what students accomplish for their projects and what they report having
learned from them.
Teaching Experience: I teach the longest-standing course on medical image analysis based on the Insight
ToolKit (ITK), an open-source toolkit for segmentation, registration, and analysis funded by the NLM. The
cross-listed course not only elicits excellent student reviews, but I have also received an Outstanding
Teaching Certificate from the University of Pittsburgh (U. Pitt.) School of Engineering for both the 2010-
2011 and 2011-2012 terms, and I was selected to use this class as a mechanism for mentoring teaching at
CMU, starting with my first mentee this spring. My public website for this methods in medical image analysis
course is sufficiently popular to be one of the first results returned by Google for “medical image analysis,”
and it has been used as a basis for several similar classes around the world, including at NIH, the University
of Iowa, Old Dominion University, Ohio State, Bahcesehir University, Istanbul, University College London,
and Mayo Clinic College of Medicine. This graduate course is heavily based on projects that utilize ITK to
tackle cutting edge problems in medicine and biomedical research. Class exercises require students to design,
build, and run experiments in software to empirically optimize their projects’ algorithm architecture and
parameter tuning. The course repeatedly stresses the need to always consult the latest scientific literature
when seeking to build useful systems, and makes use of current papers for some of the lectures. Students
interact with practicing clinicians to learn contemporary practice, workflow, and limitations within the
medical and biological communities. As part of course development, I have made numerous contributions to
both ITK’s source code and to its teaching materials. In addition, the course has afforded me the
opportunity to be involved with numerous additional graduate research projects, helping students develop
novel methods ranging from segmenting cells in confocal microscopy to Diffusion tensor imaging (DTI)-
based neurosurgical path planning. I was also PI on a NLM contract to both update my publicly available
course materials to be consistent with the new version 4 of ITK and also to help add real-time video
capabilities to v4 of ITK, thereby keeping me actively integrated into its development loop.
I have 4 years of experience teaching a pre-college project-based course in robotics as part of CMU’s Summer
Academy for Mathematics + Science (SAMS). As a summer program for diversity, SAMS is targeted at
underrepresented minorities whom it seeks to prepare for admission to selective STEM programs at colleges
and universities. Working in teams and utilizing Lego Mindstorms and Robot C, students learn how to jointly
develop a robot platform capable of performing in "real-world" conditions. The teams custom design, build,
and program their robots to compete against each other, performing a variety of tasks including linefollowing, avoiding obstacles, and automatically coloring in figures with markers. When finished, the students
have experience with real-world C programming, an initial taste of complexity management for engineering,
and the confidence that they can build nontrivial functional systems.
Finally, I have helped advise three Ph.D. students (one at CMU Robotics and two at U. Pitt. Bioengineering),
three M.S. students, and multiple undergraduate researchers. The CMU Ph.D. student received an NSF
graduate research fellowship based on her research with me, and the U. Pitt. Ph.D. students (both
international) are producing work integral to some of my current grant proposals. I have also served on
qualifier and thesis committees for other students, primarily those who have taken my methods in medical
image analysis course.