ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20111

Medical Image Analysis and Processing

Introduction to Image Processing

1

• Our General Definition of image:– A physical property(ies) of an object.

• Not necessarily visible.

• Main Physical Property:– Electromagnetic Radiation:

• From Radio Waves to Cosmic rays

• A categorization:– Single Channel– Multi Channels

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20112

Medical Image Analysis and Processing

Introduction to Image Processing

2

The Electromagnetic SpectrumThe Electromagnetic Spectrum

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20113

Medical Image Analysis and Processing

Introduction to Image Processing

3

• Digital Image, Mathematical Definition:– I = f(x,y)– I: intensity (or color)– (x,y): Position or Coordination– When (x,y) and I are finite and discrete quantities -→ digital image

– pixels, picture elements, image elements

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20114

Medical Image Analysis and Processing

Introduction to Image Processing

4

• Image Representation:

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20115

Medical Image Analysis and Processing

Introduction to Image Processing

5

Bone ScanPET

Cygnus Loop Gamma Radiation from reactor valve

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20116

Medical Image Analysis and Processing

Introduction to Image Processing

6

Chest X-Ray

Angiography

CT

Circuit Board

Cygnus Loop

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20117

Medical Image Analysis and Processing

Introduction to Image Processing

7

UV imaging

Normal Corn Smut Corn

Cygnus Loop

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20118

Medical Image Analysis and Processing

Introduction to Image Processing

8

Taxon (Anti cancer)

CholesterolMicroprocessor

Nickel Oxide Thin Film

CD SurfaceSuperconductor

Light Microscopy

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 20119

Medical Image Analysis and Processing

Introduction to Image Processing

9

• CT

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201110

Medical Image Analysis and Processing

Introduction to Image Processing

10

• MRI

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201111

Medical Image Analysis and Processing

Introduction to Image Processing

11

• MRI

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201112

Medical Image Analysis and Processing

Introduction to Image Processing

12

• US

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201113

Medical Image Analysis and Processing

Introduction to Image Processing

13

• US

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201114

Medical Image Analysis and Processing

Introduction to Image Processing

14

• SPECT

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201115

Medical Image Analysis and Processing

Introduction to Image Processing

15

• PET

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201116

Medical Image Analysis and Processing

Introduction to Image Processing

16

• PET

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201117

Medical Image Analysis and Processing

Introduction to Image Processing

17

• PET-CT

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201118

Medical Image Analysis and Processing

Introduction to Image Processing

18

• MRA

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201119

Medical Image Analysis and Processing

Introduction to Image Processing

19

• A Sample of Multi Channels imaging:– Satellite imaging

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201120

Medical Image Analysis and Processing

Introduction to Image Processing

20

• MRI as a Multi Channels imaging modalities:

PD T1 T2

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201121

Medical Image Analysis and Processing

Introduction to Image Processing

21

• MRI as a Multi Channels imaging modalities:

PD weightedPD weighted T2 weightedT2 weighted

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201122

Medical Image Analysis and Processing

Introduction to Image Processing

22

• DIP applications:– Image Quality Enrichment – Data Redundancy Reduction– Automatic Detection– Machine Vision– Machine Recognition/Verification

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201123

Medical Image Analysis and Processing

Introduction to Image Processing

23

• DIP applications:– Image Enhancement, Denoising, Reconstruction.– Authentication (Biometrics):

• Face, Signature, Fingerprint, Palm, Gesture, Retina Iris.

– Robotic Production Line (Vision)– OCR (Optical Character Recognition)– Automatic Diagnosis (Medical, Industry, and etc.)– Image Compression (jpg, tiff, jp2, and etc.)

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201124

Medical Image Analysis and Processing

Introduction to Image Processing

24

Medical Ultrasound imaging

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201125

Medical Image Analysis and Processing

Introduction to Image Processing

25

• An Example of Image Processing Results

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201126

Medical Image Analysis and Processing

Introduction to Image Processing

26

• Image Sampling– How to determine the sampling rate?– Nyquist sampling theorem

• If input is a band-limited signal with maximum frequency ΩN

• The input can be uniquely determined if sampling rate ΩS > 2ΩN

– Nyquist frequency : ΩN

– Nyquist rate : ΩS

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201127

Medical Image Analysis and Processing

Introduction to Image Processing

27

• Image Quantization– L- level digital image of size MxN– Means: A digital image having:

• A spatial resolution MxN pixels• A gray-level resolution of L levels (0-L-1)

– Spatial resolution in real-world space

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201128

Medical Image Analysis and Processing

Introduction to Image Processing

28

• Image in Matrix Form:

f(0,0) f(0,1) … f(0,N-1)f(1,0) f(0,1) … f(1,N-1)

……

f(M-1,0) f(M-1,1) … f(M-1,N-1)MxN

bits to store the image = M x N x kgray level = L = 2k

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201129

Medical Image Analysis and Processing

Introduction to Image Processing

29

• L = 2k gray levels, gray scales [0,…,L-1]• The dynamic range of an image

– [min(image) max(image)]– If the dynamic range of an image spans a

significant portion of the gray scale → highcontrast

– Otherwise, low dynamic range results in a washed out gray look

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201130

Medical Image Analysis and Processing

Introduction to Image Processing

30

• Gray Levels in CT:– Gray levels in CT image represent attenuation

coefficient in each pixel.– Gray levels expressed in Hounsfield units (HU)

• Water: 0 HU • Air: -1000 HU • Bone: 400 - 3000 HU

– Maximum CT number is 2000-4000

Water

Water air

μ-μCT= ×1000 HUμ -μ

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201131

Medical Image Analysis and Processing

Introduction to Image Processing

31

• CT images displayed with suitable brightness and contrast.• Two important value:Window Level (WL) and Window Width

(WW) • WL is CT number of mid-grey • WW is number of HU from black to white

• Choice of WW and WL dictated by clinical need– -1000 HU– 0 HU– 4000+ HU

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201132

Medical Image Analysis and Processing

Introduction to Image Processing

32

• WL and WW effect:

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201133

Medical Image Analysis and Processing

Introduction to Image Processing

33

• Paradigm of image processing:– Low-level processing

• Inputs and outputs are images• Primitive operations: de-noise, enhancement,

sharpening, …– Mid-level processing

• Inputs are images, outputs are attributes extracted from images

• Segmentation, classification,…– High-level processing

• “Make sense” of an ensemble of recognized objects by machines

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201134

Medical Image Analysis and Processing

Introduction to Image Processing

34

• Matlab Image Processing Read/Write:– imformats– imfinfo, imread, imwrite– dicominfo, dicomread, dicomwrite– analyze75info, analyze75read (Mayo Clinic)– interfileinfo, interfileread

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201135

Medical Image Analysis and Processing

Introduction to Image Processing

35

• Matlab Image Processing Display:• image, imagesc, imshow, imtool, subimage• colorbar, montage

ee.sharif.edu/~maip

E. Fatemizadeh, Sharif University of Technology, 201136

Medical Image Analysis and Processing

Introduction to Image Processing

36

• Matlab Image Processing Type Conversion:• double, ind2gray, im2double• uint16, uint8, gray2ind