methods in image analysis lecture 1 – the poet’s eye

26
1 Methods in Image Analysis Lecture 1 – The Poet’s Eye CMU Robotics Institute 16-725 U. Pitt Bioengineering 2630 Spring Term, 2004 George Stetten, M.D., Ph.D.

Upload: delora

Post on 23-Jan-2016

38 views

Category:

Documents


0 download

DESCRIPTION

Methods in Image Analysis Lecture 1 – The Poet’s Eye. George Stetten, M.D., Ph.D. CMU Robotics Institute 16-725 U. Pitt Bioengineering 2630 Spring Term, 2004. 16-725/BioE 2630 Course Structure. Review our website Very project oriented ITK community (coming to CMU, 6/5/03) - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Methods in Image Analysis Lecture 1 – The Poet’s Eye

1

Methods in Image AnalysisLecture 1 – The Poet’s Eye

CMU Robotics Institute 16-725

U. Pitt Bioengineering 2630

Spring Term, 2004

George Stetten, M.D., Ph.D.

Page 2: Methods in Image Analysis Lecture 1 – The Poet’s Eye

2

16-725/BioE 2630 Course Structure

• Review our website• Very project oriented• ITK community (coming to CMU, 6/5/03)• Strong computer vision community at CMU• Analogy to the factory floor

– Computer vision works there, discard irregulars

– In medicine, every patient is irregular

Page 3: Methods in Image Analysis Lecture 1 – The Poet’s Eye

3

Methods in Medical Image Analysis• There is a patient out there.• Medicine is a war:

Surgeons are the marines.Anesthesiologists are the air force.Psychiatrists are the diplomatsRadiologists (and Pathologists) are reconnaissance.

• Life-&-death decisions on insufficient information.• “Truth” is whether the patient recovers.• The “Practice” not the “Theory” of medicine.• Shadow Program

Page 4: Methods in Image Analysis Lecture 1 – The Poet’s Eye

4

Model of a Modern Radiologist

display of image and associated

information.

human observer

data acquisition and image

reconstruction

dictation, rounds,

procedures

automated and semi-automated image analysis

medical knowledge,patient history

Page 5: Methods in Image Analysis Lecture 1 – The Poet’s Eye

5

Methods in Medical Image Analysis

• Simultaneous measurements on a spatial grid.• Many modalities: mainly EM radiation and sound.

Page 6: Methods in Image Analysis Lecture 1 – The Poet’s Eye

6

“To invent you need a good imagination and a pile of junk.”

 

Thomas Edison

1879

Page 7: Methods in Image Analysis Lecture 1 – The Poet’s Eye

7

Bremsstrahlung

Electron rapidly decelerates at heavy metal target, giving off X-Rays.

Page 8: Methods in Image Analysis Lecture 1 – The Poet’s Eye

81896

Page 9: Methods in Image Analysis Lecture 1 – The Poet’s Eye

9

X-Ray and Fluoroscopic Images

Projection of X-Ray silhouette onto a piece of film or detector array, with intervening fluorescent screen.

Page 10: Methods in Image Analysis Lecture 1 – The Poet’s Eye

10

Computerized Tomography

From a series of projections, a tomographic image is reconstructed using Filtered Back Projection.

Page 11: Methods in Image Analysis Lecture 1 – The Poet’s Eye

11

Mass Spectrometer

Radioactive isotope separated by difference in inertia while bending in magnetic field.

Page 12: Methods in Image Analysis Lecture 1 – The Poet’s Eye

12

Nuclear Medicine

Gamma camera for creating image of radioactive target. Camera is rotated around patient in SPECT (Single Photon Emission Computed Tomography).

Page 13: Methods in Image Analysis Lecture 1 – The Poet’s Eye

13

Phased Array Ultrasound

Ultrasound beam formed and steered by controlling the delay between the elements of the transducer array.

Page 14: Methods in Image Analysis Lecture 1 – The Poet’s Eye

14

Real Time 3D Ultrasound

Page 15: Methods in Image Analysis Lecture 1 – The Poet’s Eye

15

Positron Emission Tomography

Positron-emitting organic compounds create pairs of high energy photons that are detected synchronously.

Page 16: Methods in Image Analysis Lecture 1 – The Poet’s Eye

16

Other Imaging Modalities

• MRI (will review later, with Fourier lecture)• OCT • Pathology (in addition to Radiology)• Other modalities coming down the pike

Page 17: Methods in Image Analysis Lecture 1 – The Poet’s Eye

17

Current Trends in Imaging

• 3D• Higher speed• Greater resolution• Measure function as well as structure• Combining modalities (including direct vision)

Page 18: Methods in Image Analysis Lecture 1 – The Poet’s Eye

18

The Gold Standard

• Dissection: – Medical School, Day 1: Meet the Cadaver.

– From Vesalius to the Visible Human

Page 19: Methods in Image Analysis Lecture 1 – The Poet’s Eye

19

Methods in Medical Image Analysis

• Different from “Image Processing”

• Results in identification, measurement, judgment

• Produces numbers, words, and actions

• People talk of “Computer Aided Diagnosis”

• Present state of art:

Segmentation / Registration

Page 20: Methods in Image Analysis Lecture 1 – The Poet’s Eye

20

Segmentation

• Labeling every voxel• Discrete vs. fuzzy• How good are such labels?

– Gray matter (circuits) vs. white matter (cables).

– Tremendous oversimplification

• Requires a model

Page 21: Methods in Image Analysis Lecture 1 – The Poet’s Eye

21

Registration

• Image to Image– same vs. different imaging modality– same vs. different patient– morphing Superman: topological variation

• Image to Model– deformable models

• Model to Model– matching graphs

Page 22: Methods in Image Analysis Lecture 1 – The Poet’s Eye

22

What are these models?

• Top-Down– Landmarks

– Preconceived, optimized to the humans

• Bottom-Up – Image primitives, features

– Unsupervised, optimized to the computer

• Internal Energy vs. Image Energy– Often viewed as a Bayesian compromise

– Perhaps it should, instead, be a consensus?

Page 23: Methods in Image Analysis Lecture 1 – The Poet’s Eye

23

"The poet's eye, in a fine frenzy rolling,

Doth glance from heaven to earth, and earth to heaven,

And as imagination bodies forth

The forms of things unknown, the poet's pen

Turns them to shapes, and gives to airy nothing

A local habitation and a name."  

King Theseus, from A Midsummer Night's Dream,by William Shakespeare

Page 24: Methods in Image Analysis Lecture 1 – The Poet’s Eye

24

What is Shape?

• Geometric relationships between boundary points• Curvature• Medialness (Symmetry)• Hierarchical structure• Difficult parameterize in 3D

Page 25: Methods in Image Analysis Lecture 1 – The Poet’s Eye

25

Analysis and Visualization

• Visualization used to mean to picture in the mind.• Retina is a 2D device• Analysis needed to visualize surfaces• Doctors prefer slices to renderings• Visualization is required to reach visual cortex• Computers may first beat humans in 3D

Page 26: Methods in Image Analysis Lecture 1 – The Poet’s Eye

26

Methods in Medical Image Analysis

• Computer programs (fully- and semi-automated). • Equations that function: crafting human thought.• Very application-specific, no general solution.• ITK is a library, not a program.• A kitchen of recipes, a “pile of junk.”• Supervision / Apprenticeship of machines