Associate Professor Faculty of Engineering Multimedia University - - PowerPoint PPT Presentation

▶
associate professor
SMART_READER_LITE
LIVE PREVIEW

Associate Professor Faculty of Engineering Multimedia University - - PowerPoint PPT Presentation

Nov 22, 2022 32 likes •675 views

Mohammad Faizal Ahmad Fauzi, Ph.D. Associate Professor Faculty of Engineering Multimedia University Imaging Informatics Imaging basics Imaging modalities PACS and its core functions DICOM 20 mm Nodule 4 mm Nodule 5 mm

slide-1
SLIDE 1

Mohammad Faizal Ahmad Fauzi, Ph.D. Associate Professor Faculty of Engineering Multimedia University

slide-2
SLIDE 2

 Imaging Informatics  Imaging basics  Imaging modalities  PACS and its core functions  DICOM

slide-3
SLIDE 3
slide-4
SLIDE 4
slide-5
SLIDE 5
slide-6
SLIDE 6
slide-7
SLIDE 7
slide-8
SLIDE 8
slide-9
SLIDE 9
slide-10
SLIDE 10
slide-11
SLIDE 11
slide-12
SLIDE 12
slide-13
SLIDE 13

20 mm Nodule

slide-14
SLIDE 14

4 mm Nodule

slide-15
SLIDE 15

5 mm Nodule

slide-16
SLIDE 16
slide-17
SLIDE 17

Reader 1 Reader 1

slide-18
SLIDE 18

Reader 2 Reader 2

slide-19
SLIDE 19

Reader 3 Reader 3

slide-20
SLIDE 20

 Fatigue  Distraction  Emotional stress  Variation in reader  Satisfaction of Search

slide-21
SLIDE 21
slide-22
SLIDE 22

 Breast cancer is missed 10-30%

…

 by Expert Mammographers

slide-23
SLIDE 23
slide-24
SLIDE 24
slide-25
SLIDE 25
slide-26
SLIDE 26
slide-27
SLIDE 27

 Sensitivity of radiologists in detecting breast

cancer on mammograms can be improved by

15% through double reading.

slide-28
SLIDE 28

 Computer-aided

diagnosis:

  • a diagnosis made by

a physician using the

  • utput of a

computerized system

 Computerized

system

  • Automated image (or

data) analysis

slide-29
SLIDE 29

 Breast Cancer  Lung Cancer  Brain Cancer  Colon Cancer

slide-30
SLIDE 30

Find Six Differences

slide-31
SLIDE 31

Find Six Differences

slide-32
SLIDE 32
  • Microcalcifications
  • Masses
  • Solitary Pulmonary Nodules
  • Ground Glass Opacities
slide-33
SLIDE 33

Benign Malignant

slide-34
SLIDE 34

HR 2 (7/23/01) 5 10 15 20 25 30 35 40 5 10 15 20 25 30 35 40

slide-35
SLIDE 35

1 2 3 4 5 6 7 8 9 10 11 12

slide-36
SLIDE 36

10 20 30 40 50 10 20 30 40 50 60

slide-37
SLIDE 37

 Organ segmentation  Candidate detection/segmentation  Feature Extraction  Classification/clustering

slide-38
SLIDE 38

 Organ segmentation  Candidate detection/segmentation  Feature Extraction  Classification/clustering

slide-39
SLIDE 39
slide-40
SLIDE 40

 Segment Lung Regions within the CT slice  Detect left and right lungs

slide-41
SLIDE 41
slide-42
SLIDE 42

 Segmented lung region may exclude some

nodules adjacent to pleura

 Connect edge points of concave regions  Recover potential nodules adjacent to pleura

slide-43
SLIDE 43
slide-44
SLIDE 44
slide-45
SLIDE 45
slide-46
SLIDE 46

P1 P2 de d1 d2

slide-47
SLIDE 47

 Organ segmentation  Candidate detection/segmentation  Feature Extraction  Classification/clustering

slide-48
SLIDE 48

Pre-screening CNN Classifier Clustering

Mammogram Image Potential Signals Potential TP Signals Microcalcification Clusters

slide-49
SLIDE 49

Pre-screening CNN Classifier Clustering

Mammogram Image Potential Signals Potential TP Signals Microcalcification Clusters

slide-50
SLIDE 50
slide-51
SLIDE 51
slide-52
SLIDE 52
slide-53
SLIDE 53

 Identify high density regions within segmented

lung regions

 Segmentation by k-means clustering with two

classes:

  • nodule candidates
  • lung region
slide-54
SLIDE 54
slide-55
SLIDE 55

Thin long structure True nodule V-shaped structure

Identification of Blood Vessels

slide-56
SLIDE 56

 Organ segmentation  Candidate detection/segmentation  Feature Extraction  Classification/clustering

slide-57
SLIDE 57

 Thin long structures

  • Major-to-minor axis ratio of a fitted ellipse

 V-shaped structures

  • Rectangularity
slide-58
SLIDE 58

 Thin long structures  V-shaped structures a b

b a Rtl 

  • bject
  • f

Area rectangle

  • f

Area 

v

R

slide-59
SLIDE 59

 Organ segmentation  Candidate detection/segmentation  Feature Extraction  Classification/clustering

slide-60
SLIDE 60

FP ROI TP ROI

slide-61
SLIDE 61

Pre-screening CNN Classifier Clustering

Mammogram Image Potential Signals Potential TP Signals Microcalcification Clusters

slide-62
SLIDE 62

INPUT ROI

{

0: FP 1: TP

CNN Classifier

slide-63
SLIDE 63

Pre-screening CNN Classifier Clustering

Mammogram Image Potential Signals Potential TP Signals Microcalcification Clusters

slide-64
SLIDE 64

 Image

  • How to represent
  • How to generate it

 Imaging modalities

  • How to integrate
  • How to manage

 Image Analysis

  • Radiology
  • Big picture