View
214
Download
0
Category
Tags:
Preview:
Citation preview
The content of these slides by John Galeotti, © 2008 to 2015 Carnegie Mellon University (CMU), was made possible in part by NIH NLM contract# HHSN276201000580P, and is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this
license, visit http://creativecommons.org/licenses/by/3.0/ or send a letter to Creative Commons, 171 2nd Street, Suite 300, San Francisco, California, 94105, USA. Permissions beyond the scope of this license may be available either from CMU or by emailing
itk@galeotti.net.The most recent version of these slides may be accessed online via http://itk.galeotti.net/
Methods In Medical Image Analysis
Spring 2015 BioE 2630 (Pitt) : 16-725 (CMU RI)
18-791 (CMU ECE) : 42-735 (CMU BME)
Dr. John Galeotti
2
What Are We Doing?
Theoretical & practical skills in medical image analysisImaging modalitiesSegmentationRegistrationImage understandingVisualization
Established methods and current researchFocus on understanding & using algorithms
3
Why Is Medical Image Analysis Special?
Because of the patientComputer Vision:
Good at detecting irregulars, e.g. on the factory floorBut no two patients are alike—everyone is “irregular”
Medicine is warRadiology is primarily for reconnaissance Surgeons are the marines Life/death decisions made on insufficient information
Success measured by patient recoveryYou’re not in “theory land” anymore
4
What Do I Mean by Analysis?
Different from “Image Processing”Results in identification, measurement, &/or judgment
Produces numbers, words, & actionsHoly Grail: complete image understanding automated within a computer to perform diagnosis & control robotic intervention
State of the art: segmentation & registration
5
SegmentationLabeling every voxelDiscrete vs. fuzzyHow good are such labels?
Gray matter (circuits) vs. white matter (cables).Tremendous oversimplification
Requires a model
6
Registration
Image to Imagesame vs. different imaging modalitysame vs. different patienttopological variation
Image to Modeldeformable models
Model to Modelmatching graphs
7
Visualization
Visualization used to mean to picture in the mind.Retina is a 2D deviceAnalysis needed to visualize surfacesDoctors prefer slices to renderingsVisualization is required to reach visual cortexComputers have an advantage over humans in 3D
8
Model of a Modern Radiologist
9
How Are We Going to Do This?
The Shadow ProgramObserve & interact with practicing radiologists and
pathologists at UPMCProject oriented
C++ &/or Python with ITK New ITKv4!
National Library of Medicine Insight ToolkitA software library developed by a consortium of
institutions including CMU and UPittOpen source Large online communitywww.itk.org
10
The Practice of Automated Medical Image Analysis
A collection of recipes, a box of toolsEquations that function: crafting human thought. ITK is a library, not a program.
Solutions:Computer programs (fully- and semi-automated).Very application-specific, no general solution. Supervision / apprenticeship of machines
11
Who Are We?
Personal introductionsNameAcademic Background (ECE, Biology, etc.)Research InterestWhy you’re here
Homework 1: email the TA/grader & myself the requested info about yourself, and a photo. (photo is optional, but requested; please crop to your head and shoulders) See website for HW1 details.
12
Syllabus
On the course websitehttp://www.cs.cmu.edu/~galeotti/methods_course/
PrerequisitesVector calculusBasic probabilityKnowledge of C++ and/or Python Including command-line usage and command-line
argument passing to your codeHelpful but not required:Knowledge of C++ templates & inheritance
13
Class Schedule
Comply with Pitt & CMU calendarsOnline and subject to changeBig picture:
Background & reviewFundamentalsSegmentation, registration, & other fun stuffMore advanced ITK programming constructsReview scientific papersStudent project presentations
14
Requirements and Grading
Attendance: Required (quizzes)Quizzes: 20%
Lowest 2 droppedHomework: 30%Shadow Program: 10%Final Project: 40%
15% presentation25% code
15
Textbooks
Required: Machine Vision, Wesley E. Snyder & Hairong Qi
Recommended: Insight into Images: Principles and Practice for Segmentation, Registration and Image Analysis, Terry S. Yoo (Editor)
Others (build your bookshelf)
16
Superior = headInferior = feet
Anterior = frontPosterior = back
Proximal = centralDistal = peripheral
Anatomical Axes
Recommended