SlideShare une entreprise Scribd logo
1  sur  78
Télécharger pour lire hors ligne
MEDICAL IMAGE COMPUTING (CAP 5937)
LECTURE 7: Medical Image Segmentation (I)
(Radiology Applications of Segmentation, and Thresholding)
Dr. Ulas Bagci
HEC 221, Center for Research in Computer
Vision (CRCV), University of Central Florida
(UCF), Orlando, FL 32814.
bagci@ucf.edu or bagci@crcv.ucf.edu
1SPRING 2017
Outline
• Introduction to Medical Image Segmentation, type of
segmentation methods, and definitions
– Recognition & Delineation
• Simplest Segmentation Method(s): Thresholding
– Otsu Thresholding
– Parametric Method
– PET Image Thresholding Methods
• ITM (Iterative Thresholding Method)
2
Motivation for Image Segmentation
In the last 20 years the computer vision and medical imaging
communities have produced a number of useful algorithms for
localizing object boundaries in images.
3
Motivation for Image Segmentation
• Content based image retrieval
• Machine Vision
• Medical Imaging applications (tumor delineation,..)
• Object detection (face detection,…)
• 3D Reconstruction
• Object/Motion Tracking
• Object-based measurements such as size and shape
• Object recognition (face recognition,…)
• Fingerprint recognition,
• Video surveillance
• …
4
Segmentation Tools in RadiologyApplications
• 3D views to visualize structural information and spatial
anatomic relationships is a difficult task, which is usually
carried out in the clinician’s mind.
5
Segmentation Tools in RadiologyApplications
• 3D views to visualize structural information and spatial
anatomic relationships is a difficult task, which is usually
carried out in the clinician’s mind.
• Image-processing tools provide the surgeon with interactively
displayed 3D visual information.
6
Segmentation Tools in RadiologyApplications
7
Credit: Kaus, et al. Radiology 2001.
• Determination of the volumes of abdominal solid organs and focal lesions
has great potential importance (liver, spleen, …).
• Monitoring the response to therapy and the progression of neoplastic
disease and preoperative examination of living liver donors are the most
common clinical applications of volume determination.
8
Segmentation Tools in RadiologyApplications
(credit: Farraher, et al.
Radiology 2005)
Segmentation Tools in RadiologyApplications
• Gross Tumor Volume in CT/MRI
• Metabolic Tumor Volume in PET/SPECT/
– Surgery/Therapy Planning
• Planning Tumor Volume (PTV)
– Tumor characterization
• Texture Extraction requires
segmentation to be done
• Shape analysis
9
Segmentation Tools in RadiologyApplications
• There is a strong interest in automatic and reproducible
techniques for detection and quantification of vascular
disease
• A first step toward an effective vessel analysis tool is
segmentation of the vasculature.
10
axial coronal sagittal
Credit: Manniesing, et al,
Radiology 2008
MIP: maximum intensity
Projection image of cerebral vessels (in CTA)
Segmentation Tools in RadiologyApplications
• MR volumetry of the
hippocampus can help
distinguish patients with
AD (Alzheimer’s
Disease) from elderly
controls with a high
degree of accuracy
(80%–90%).
11
Segmentation Tools in RadiologyApplications
• MR volumetry of the
hippocampus can help
distinguish patients with
AD (Alzheimer’s
Disease) from elderly
controls with a high
degree of accuracy
(80%–90%).
12
hippocampus
amygdala
Credit: Colliot et al, Radiology 2008.
Image Segmentation
Definition: Partitioning a picture/image into distinctive subsets is
called segmentation.
13
Image Segmentation
Definition: Partitioning a picture/image into distinctive subsets is
called segmentation.
14
Segmentation of an image entails the division or
separation of the image
into regions of similar attribute.
Image Segmentation
Definition: Partitioning a picture/image into distinctive subsets is
called segmentation.
15
Segmentation of an image entails the division or
separation of the image
into regions of similar attribute.
The most basic attributes:
-intensity
-edges
-texture
-other features…
Image Segmentation
Definition: Partitioning a picture/image into distinctive subsets is
called segmentation.
16
Purpose: To extractobject information
and representthis as a
hard/fuzzygeometric
structure.
Recognition: Determiningthe object’s
whereaboutsin the scene.
(humans> computer)
Delineation: Determining the object’s
spatial extent and
compositionin the scene.
(computers > humans)
Recognition - Example
17
(slice credit: J. Kim et al,
Signal Processing 2007)
Model is induced No Model is induced
Approaches to Recognition
18
• Model-based
• Knowledge-based - Non-interactive
• Atlas-based
• Human-assisted - Interactive
Approaches to Recognition
19
• Model-based
• Knowledge-based - Non-interactive
• Atlas-based
• Human-assisted - Interactive
- They all originate from human knowledge.
- Their relative efficacy is unknown.
Approaches to Delineations
20
pI (purely image-based) approaches
• Rely mostlyon informationavailable in the given image
only.
• Recognition: manual
Approaches to Delineations
21
pI (purely image-based) approaches
• Rely mostlyon informationavailable in the given image
only.
• Recognition: manual
SM (shape model-based) approaches
• Employ models to codify object family shape info.
• Recognition: model-based/manual
Approaches to Delineations
22
pI (purely image-based) approaches
• Rely mostlyon informationavailable in the given image
only.
• Recognition: manual
SM (shape model-based) approaches
• Employ models to codify object family shape info.
• Recognition: model-based/manual
Hybrid approaches
• Combine among pI and SM approaches.
• Recognition: model-based, automatic.
Classification of Methods
23
Boundary-based (BpI):
• optimum boundary
• active boundary
• live wire
• level sets
Classification of Methods
24
Boundary-based (BpI):
• optimum boundary
• active boundary
• live wire
• level sets
Region-based (RpI):
• clustering – kNN, CM, FCM
• graph cut
• fuzzy connectedness
• MRF
• watershed
• optimum partitioning
• (Mumford-Shah)
Classification of Methods
25
Boundary-based (BpI):
• optimum boundary
• active boundary
• live wire
• level sets
Region-based (RpI):
• clustering – kNN, CM, FCM
• graph cut
• fuzzy connectedness
• MRF
• watershed
• optimum partitioning
• (Mumford-Shah)
SM Approaches
• manual tracing
• live wire
• active shape/appearance
• M-reps
• atlas-based
Classification of Methods
26
Boundary-based (BpI):
• optimum boundary
• active boundary
• live wire
• level sets
Region-based (RpI):
• clustering – kNN, CM, FCM
• graph cut
• fuzzy connectedness
• MRF
• watershed
• optimum partitioning
• (Mumford-Shah)
SM Approaches
• manual tracing
• live wire
• active shape/appearance
• M-reps
• atlas-based
Hybrid Approaches
• BpI + BpI
• RpI + RpI
• BpI + RpI
• BpI + SM
• RpI + SM
• SM + SM
Classification of Methods
27
pI Approaches
+ Where image info is good,
accuracy is good;
- Bad where it is poor/absent;
- Need recognition help;
+ Can determine degree of
match of model to image
well;
- Lack obj shape &
geographic info;
Classification of Methods
28
SMApproaches
- Even where image info is
good, accuracy suffers;
+ Where bad, model helps;
+ Can help in recognition;
- Need best match info;
+ Good models embody obj
shape & geographic info;
Purely Image Based Segmentation Methods
29
Thresholding – Simple Segmentation
• Image binarization
– mapping a scalar image I into a binary image J
30
J(x, y) =
(
0 if I(x, y) < T
1 otherwise.
Thresholding – Simple Segmentation
• Image binarization
– mapping a scalar image I into a binary image J
31
J(x, y) =
(
0 if I(x, y) < T
1 otherwise.
Thresholding – Simple Segmentation
32
Brighter objects
Darker objects
Thresholding – Simple Segmentation
33
Brighter objects
Darker objects
DIFFICULTIES
1. The valley may be so broad that
it is difficult to locate a
significant minimum
2. Number of minima due to type
of details in the image
3. Noise
4. No visible valley
5. Histogram may be multi-modal
Example: CT Scan
34
Example: CT Scan
35
Example: CT Scan
36
Example: CT Scan
37
Example: CT Scan
38
Thresholding Methods
• Huang
• Intermode
• Isodata
• Li
• MaxEntropy
• Mean
• MinError
• Otsu
• Percentile
• RenyiEntropy
• Moments
39
Thresholding Methods
• Huang
• Intermode
• Isodata
• Li
• MaxEntropy
• Mean
• MinError
• Otsu
• Percentile
• RenyiEntropy
• Moments
40
Thresholding Methods
PET Imaging
Fixed Thresholding
Adaptive Thresholding
Iterative Thresholding
41
• Huang
• Intermode
• Isodata
• Li
• MaxEntropy
• Mean
• MinError
• Otsu (non-parametric)
• Percentile
• RenyiEntropy
• Moments
Otsu Thresholding
• Definition: The method uses the grey-value histogram of the
given image I as input and aims at providing the best
threshold in the sense that the “overlap” between two
classes, set of object and background pixels, is minimized
(i.e., by finding the best balance).
42
Otsu Thresholding
• Definition: The method uses the grey-value histogram of the
given image I as input and aims at providing the best
threshold in the sense that the “overlap” between two
classes, set of object and background pixels, is minimized
(i.e., by finding the best balance).
• Otsu’s algorithm selects a threshold that maximizes the
between-class variance . In the case of two classes,
43
2
b
2
b = P1(µ1 µ)2
+ P2(µ2 µ)2
= P1P2(µ1 µ2)2
Otsu Thresholding
• Definition: The method uses the grey-value histogram of the
given image I as input and aims at providing the best
threshold in the sense that the “overlap” between two
classes, set of object and background pixels, is minimized
(i.e., by finding the best balance).
• Otsu’s algorithm selects a threshold that maximizes the
between-class variance . In the case of two classes,
• where P1 and P2 denote class probabilities, and μi the means
of object and background classes.
44
2
b
2
b = P1(µ1 µ)2
+ P2(µ2 µ)2
= P1P2(µ1 µ2)2
Otsu Thresholding
• Definition: The method uses the grey-value histogram of the
given image I as input and aims at providing the best
threshold in the sense that the “overlap” between two
classes, set of object and background pixels, is minimized
(i.e., by finding the best balance).
45
P1 =
uX
ı=0
p(i)
P2 =
GmaxX
ı=u+1
p(i)
u
u
Otsu Thresholding
• Definition: The method uses the grey-value histogram of the
given image I as input and aims at providing the best
threshold in the sense that the “overlap” between two
classes, set of object and background pixels, is minimized
(i.e., by finding the best balance).
46
P1 =
uX
ı=0
p(i)
P2 =
GmaxX
ı=u+1
p(i)
µ1 =
uX
ı=0
ip(i)/P1
µ2 =
GmaxX
ı=u+1
ip(i)/P2
CLASS MEANS
Otsu Thresholding-Algorithm
47
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
Otsu Thresholding-Algorithm
48
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
2
b = P1(µ1 µ)2
+ P2(µ2 µ)2
= P1P2(µ1 µ2)2
Otsu Thresholding-Algorithm
49
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
Otsu Thresholding-Algorithm
50
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
Otsu Thresholding-Algorithm
51
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
Otsu Thresholding-Algorithm
52
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
Otsu Thresholding-Algorithm
53
cI (u) 1 cI(u)
P1 P2
c indicates cumulative histogram,and P1 and P2
can be approximated well with cumulative density function.
optimal
Parametric Method for Optimal Thresholding
• Assuming again a two-class problem and assuming that the
distribution of gray levels for each class can be modeled by a
normal distribution with mean and variance
54
Parametric Method for Optimal Thresholding
• Assuming again a two-class problem and assuming that the
distribution of gray levels for each class can be modeled by a
normal distribution with mean and variance
• the overall normalized intensity histogram can be written as
the following mixture probability density function:
55
Parametric Method for Optimal Thresholding
• Assuming again a two-class problem and assuming that the
distribution of gray levels for each class can be modeled by a
normal distribution with mean and variance
• the overall normalized intensity histogram can be written as
the following mixture probability density function:
where P1 and P2 are class probabilities. The optimal threshold
(T) can be found as solving the quadratic equation à
56
Parametric Method for Optimal Thresholding
57
Parametric Method for Optimal Thresholding
58
In case, variances of both classes are equal, then->
Parametric Method for Optimal Thresholding
59
In case, variances of both classes are equal, then->
Thresholding methods for PET Image
Segmentation
• Due to the nature of PET images (i.e., low resolution with high
contrast), thresholding-based methods are suitable
– because the local or global intensity histogram usually provides a
sufficient level of information for separating the foreground (object of
interest) from the background. (Foster, Bagci, et al., CBM 2014)
60
Thresholding methods for PET Image
Segmentation
• Due to the nature of PET images (i.e., low resolution with high
contrast), thresholding-based methods are suitable
– because the local or global intensity histogram usually provides a
sufficient level of information for separating the foreground (object of
interest) from the background. (Foster, Bagci, et al., CBM 2014)
61
Fixed
Thresholding
Adaptive
Thresholding
Iterative
Thresholding
Fixed Thresholding Methods
• Due to the nature of PET images (i.e., low resolution with high
contrast), thresholding-based methods are suitable
– because the local or global intensity histogram usually provides a
sufficient level of information for separating the foreground (object of
interest) from the background. (Foster, Bagci, et al., CBM 2014)
62
Thresholding methods for PET Image
Segmentation
• Due to the nature of PET images (i.e., low resolution with high
contrast), thresholding-based methods are suitable
– because the local or global intensity histogram usually provides a
sufficient level of information for separating the foreground (object of
interest) from the background. (Foster, Bagci, et al., CBM 2014)
63
Fixed
Thresholding
Adaptive
Thresholding
Iterative
Thresholding
Phantom
Based
Image Quality
metrics based
Adaptive Thresholding 64
Thresholding methods for PET Image
Segmentation
• Due to the nature of PET images (i.e., low resolution with high
contrast), thresholding-based methods are suitable
– because the local or global intensity histogram usually provides a
sufficient level of information for separating the foreground (object of
interest) from the background. (Foster, Bagci, et al., CBM 2014)
65
Fixed
Thresholding
Adaptive
Thresholding
Iterative
Thresholding
Phantom
Based
Image Quality
metrics based
Iterative Thresholding Method (ITM)
66
S/B: Source to background ratio.
The method is based on calibrated
threshold-volume curves at varying
S/B ratio acquired by phantom
measurements using spheres of known
volumes.
Iterative Thresholding Method (ITM)
67
S/B: Source to background ratio.
The method is based on calibrated
threshold-volume curves at varying
S/B ratio acquired by phantom
measurements using spheres of known
volumes.
Iterative Thresholding Method (ITM)
68
S/B: Source to background ratio.
The method is based on calibrated
threshold-volume curves at varying
S/B ratio acquired by phantom
measurements using spheres of known
volumes.
The measured S/B ratios of the
lesions are then estimated from
PET images, and their volumes are
iteratively calculated using the
calibrated S/B-threshold-volume curves
Iterative Thresholding Method (ITM)
69
S/B: Source to background ratio.
The method is based on calibrated
threshold-volume curves at varying
S/B ratio acquired by phantom
measurements using spheres of known
volumes.
The measured S/B ratios of the
lesions are then estimated from
PET images, and their volumes are
iteratively calculated using the
calibrated S/B-threshold-volume curves
The resulting PET volumes are then
compared with the known sphere volume
and CT volumes of tumors that served
as gold standards.
ITM Example Result on PET Images/Lung
70
Another Example for PET Thresholding
71
ITM for tumor segmentation/FDG PET
Another Example for PET Thresholding
72
Further Thresholding Example – CT Bones
73
Further Thresholding Example – CT Bones
74
Head-Neck CT – Thresholding for Skull
Modeling
75
(Slice Credit: P.Seutens)
Segmentation of the skull and the mandibula in CT images using thresholding.(a) Original CT
image of the head. (b) Result with a threshold value of 276 Hounsfield units. The segmented bony
structures are represented in color. (c) 3D rendering of the skull shows a congenital growth
deficiency of the mandibula in this 8-year-old patient. This information was used preoperatively to
plan a repositioning of the mandibula.
Multiple Thresholds – MRI Thresholding
76
Thresholding can be done interactively and separates the image into different
regions. Valleys in the histogram indicate potentially useful threshold values
Credit: Toeonies,K.
Summary of today’s lecture
• Introduction into the Medical Image Segmentation
• Recognition and Delineation concepts in Segmentation
• Simplest Segmentation method: Thresholding
– Otsu
– Parametric method for optimal thresholding
– PET Image thresholding
• ITM, fixed thresholding,etc.
77
Slide Credits and References
• Jayaram K. Udupa, MIPG of University of Pennsylvania, PA.
• P. Suetens, Fundamentals of Medical Imaging, Cambridge
Univ. Press.
• Foster, B., et al. CBM, Review paper, 2014.
• Kaus, et al. Radiology 2001.
• Toeonies, K., Medical Image Analysis.
• Farraher, et al., Radiology 2005
• Zaidi, H., Quantitative Analysis in Nuclear Medicine Imaging.
• Bailey et al. Positron Emission Tomography, Springer.
• Dawood, M., et al. Correction Techniques in Emission
Tomography
78

Contenu connexe

Tendances

Lec3: Pre-Processing Medical Images
Lec3: Pre-Processing Medical ImagesLec3: Pre-Processing Medical Images
Lec3: Pre-Processing Medical ImagesUlaş Bağcı
 
Lec10: Medical Image Segmentation as an Energy Minimization Problem
Lec10: Medical Image Segmentation as an Energy Minimization ProblemLec10: Medical Image Segmentation as an Energy Minimization Problem
Lec10: Medical Image Segmentation as an Energy Minimization ProblemUlaş Bağcı
 
BRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSING
BRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSINGBRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSING
BRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSINGDharshika Shreeganesh
 
Brain Tumor Segmentation using Enhanced U-Net Model with Empirical Analysis
Brain Tumor Segmentation using Enhanced U-Net Model with Empirical AnalysisBrain Tumor Segmentation using Enhanced U-Net Model with Empirical Analysis
Brain Tumor Segmentation using Enhanced U-Net Model with Empirical AnalysisMD Abdullah Al Nasim
 
Image segmentation ppt
Image segmentation pptImage segmentation ppt
Image segmentation pptGichelle Amon
 
MEDICAL IMAGE PROCESSING
MEDICAL IMAGE PROCESSING MEDICAL IMAGE PROCESSING
MEDICAL IMAGE PROCESSING MOUMITA GHOSH
 
Optic flow estimation with deep learning
Optic flow estimation with deep learningOptic flow estimation with deep learning
Optic flow estimation with deep learningYu Huang
 
Lecture 1 for Digital Image Processing (2nd Edition)
Lecture 1 for Digital Image Processing (2nd Edition)Lecture 1 for Digital Image Processing (2nd Edition)
Lecture 1 for Digital Image Processing (2nd Edition)Moe Moe Myint
 
Deep learning for medical imaging
Deep learning for medical imagingDeep learning for medical imaging
Deep learning for medical imaginggeetachauhan
 
Presentation on deformable model for medical image segmentation
Presentation on deformable model for medical image segmentationPresentation on deformable model for medical image segmentation
Presentation on deformable model for medical image segmentationSubhash Basistha
 
Brain Tumor Detection using CNN
Brain Tumor Detection using CNNBrain Tumor Detection using CNN
Brain Tumor Detection using CNNMohammadRakib8
 

Tendances (20)

Image segmentation
Image segmentation Image segmentation
Image segmentation
 
Final ppt
Final pptFinal ppt
Final ppt
 
Image segmentation
Image segmentationImage segmentation
Image segmentation
 
Image segmentation
Image segmentationImage segmentation
Image segmentation
 
Medical Image Processing
Medical Image ProcessingMedical Image Processing
Medical Image Processing
 
Lec3: Pre-Processing Medical Images
Lec3: Pre-Processing Medical ImagesLec3: Pre-Processing Medical Images
Lec3: Pre-Processing Medical Images
 
Wiener Filter
Wiener FilterWiener Filter
Wiener Filter
 
Lec10: Medical Image Segmentation as an Energy Minimization Problem
Lec10: Medical Image Segmentation as an Energy Minimization ProblemLec10: Medical Image Segmentation as an Energy Minimization Problem
Lec10: Medical Image Segmentation as an Energy Minimization Problem
 
BRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSING
BRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSINGBRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSING
BRAIN TUMOR MRI IMAGE SEGMENTATION AND DETECTION IN IMAGE PROCESSING
 
Brain Tumor Segmentation using Enhanced U-Net Model with Empirical Analysis
Brain Tumor Segmentation using Enhanced U-Net Model with Empirical AnalysisBrain Tumor Segmentation using Enhanced U-Net Model with Empirical Analysis
Brain Tumor Segmentation using Enhanced U-Net Model with Empirical Analysis
 
Image segmentation ppt
Image segmentation pptImage segmentation ppt
Image segmentation ppt
 
Canny Edge Detection
Canny Edge DetectionCanny Edge Detection
Canny Edge Detection
 
MEDICAL IMAGE PROCESSING
MEDICAL IMAGE PROCESSING MEDICAL IMAGE PROCESSING
MEDICAL IMAGE PROCESSING
 
Optic flow estimation with deep learning
Optic flow estimation with deep learningOptic flow estimation with deep learning
Optic flow estimation with deep learning
 
Lecture 1 for Digital Image Processing (2nd Edition)
Lecture 1 for Digital Image Processing (2nd Edition)Lecture 1 for Digital Image Processing (2nd Edition)
Lecture 1 for Digital Image Processing (2nd Edition)
 
Digital Image Processing
Digital Image ProcessingDigital Image Processing
Digital Image Processing
 
Deep learning for medical imaging
Deep learning for medical imagingDeep learning for medical imaging
Deep learning for medical imaging
 
DIGITAL IMAGE PROCESSING
DIGITAL IMAGE PROCESSINGDIGITAL IMAGE PROCESSING
DIGITAL IMAGE PROCESSING
 
Presentation on deformable model for medical image segmentation
Presentation on deformable model for medical image segmentationPresentation on deformable model for medical image segmentation
Presentation on deformable model for medical image segmentation
 
Brain Tumor Detection using CNN
Brain Tumor Detection using CNNBrain Tumor Detection using CNN
Brain Tumor Detection using CNN
 

En vedette

Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)
Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)
Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)Ulaş Bağcı
 
Lec12: Shape Models and Medical Image Segmentation
Lec12: Shape Models and Medical Image SegmentationLec12: Shape Models and Medical Image Segmentation
Lec12: Shape Models and Medical Image SegmentationUlaş Bağcı
 
Lec6: Pre-Processing for Nuclear Medicine Images
Lec6: Pre-Processing for Nuclear Medicine ImagesLec6: Pre-Processing for Nuclear Medicine Images
Lec6: Pre-Processing for Nuclear Medicine ImagesUlaş Bağcı
 
Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)
Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)
Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)Ulaş Bağcı
 
Lec16: Medical Image Registration (Advanced): Deformable Registration
Lec16: Medical Image Registration (Advanced): Deformable RegistrationLec16: Medical Image Registration (Advanced): Deformable Registration
Lec16: Medical Image Registration (Advanced): Deformable RegistrationUlaş Bağcı
 
Lec11: Active Contour and Level Set for Medical Image Segmentation
Lec11: Active Contour and Level Set for Medical Image SegmentationLec11: Active Contour and Level Set for Medical Image Segmentation
Lec11: Active Contour and Level Set for Medical Image SegmentationUlaş Bağcı
 
Lec4: Pre-Processing Medical Images (II)
Lec4: Pre-Processing Medical Images (II)Lec4: Pre-Processing Medical Images (II)
Lec4: Pre-Processing Medical Images (II)Ulaş Bağcı
 
Lec15: Medical Image Registration (Introduction)
Lec15: Medical Image Registration (Introduction)Lec15: Medical Image Registration (Introduction)
Lec15: Medical Image Registration (Introduction)Ulaş Bağcı
 
Lec14: Evaluation Framework for Medical Image Segmentation
Lec14: Evaluation Framework for Medical Image SegmentationLec14: Evaluation Framework for Medical Image Segmentation
Lec14: Evaluation Framework for Medical Image SegmentationUlaş Bağcı
 
Lec13: Clustering Based Medical Image Segmentation Methods
Lec13: Clustering Based Medical Image Segmentation MethodsLec13: Clustering Based Medical Image Segmentation Methods
Lec13: Clustering Based Medical Image Segmentation MethodsUlaş Bağcı
 
Lec2: Digital Images and Medical Imaging Modalities
Lec2: Digital Images and Medical Imaging ModalitiesLec2: Digital Images and Medical Imaging Modalities
Lec2: Digital Images and Medical Imaging ModalitiesUlaş Bağcı
 

En vedette (11)

Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)
Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)
Lec5: Pre-Processing Medical Images (III) (MRI Intensity Standardization)
 
Lec12: Shape Models and Medical Image Segmentation
Lec12: Shape Models and Medical Image SegmentationLec12: Shape Models and Medical Image Segmentation
Lec12: Shape Models and Medical Image Segmentation
 
Lec6: Pre-Processing for Nuclear Medicine Images
Lec6: Pre-Processing for Nuclear Medicine ImagesLec6: Pre-Processing for Nuclear Medicine Images
Lec6: Pre-Processing for Nuclear Medicine Images
 
Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)
Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)
Lec9: Medical Image Segmentation (III) (Fuzzy Connected Image Segmentation)
 
Lec16: Medical Image Registration (Advanced): Deformable Registration
Lec16: Medical Image Registration (Advanced): Deformable RegistrationLec16: Medical Image Registration (Advanced): Deformable Registration
Lec16: Medical Image Registration (Advanced): Deformable Registration
 
Lec11: Active Contour and Level Set for Medical Image Segmentation
Lec11: Active Contour and Level Set for Medical Image SegmentationLec11: Active Contour and Level Set for Medical Image Segmentation
Lec11: Active Contour and Level Set for Medical Image Segmentation
 
Lec4: Pre-Processing Medical Images (II)
Lec4: Pre-Processing Medical Images (II)Lec4: Pre-Processing Medical Images (II)
Lec4: Pre-Processing Medical Images (II)
 
Lec15: Medical Image Registration (Introduction)
Lec15: Medical Image Registration (Introduction)Lec15: Medical Image Registration (Introduction)
Lec15: Medical Image Registration (Introduction)
 
Lec14: Evaluation Framework for Medical Image Segmentation
Lec14: Evaluation Framework for Medical Image SegmentationLec14: Evaluation Framework for Medical Image Segmentation
Lec14: Evaluation Framework for Medical Image Segmentation
 
Lec13: Clustering Based Medical Image Segmentation Methods
Lec13: Clustering Based Medical Image Segmentation MethodsLec13: Clustering Based Medical Image Segmentation Methods
Lec13: Clustering Based Medical Image Segmentation Methods
 
Lec2: Digital Images and Medical Imaging Modalities
Lec2: Digital Images and Medical Imaging ModalitiesLec2: Digital Images and Medical Imaging Modalities
Lec2: Digital Images and Medical Imaging Modalities
 

Similaire à Lec7: Medical Image Segmentation (I) (Radiology Applications of Segmentation, and Thresholding)

A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...
A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...
A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...inventy
 
SIRG-BSU_3_used-important.pdf
SIRG-BSU_3_used-important.pdfSIRG-BSU_3_used-important.pdf
SIRG-BSU_3_used-important.pdfDrAhmedElngar
 
A Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation MethodologyA Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation Methodologyaciijournal
 
brain tumor detection by thresholding approach
brain tumor detection by thresholding approachbrain tumor detection by thresholding approach
brain tumor detection by thresholding approachSahil Prajapati
 
Image annotation - Segmentation & Annotation
Image annotation - Segmentation & AnnotationImage annotation - Segmentation & Annotation
Image annotation - Segmentation & AnnotationTaposh Roy
 
Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...
Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...
Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...Akshit Arora
 
Segmentation techniques for extraction and description of tumour region from ...
Segmentation techniques for extraction and description of tumour region from ...Segmentation techniques for extraction and description of tumour region from ...
Segmentation techniques for extraction and description of tumour region from ...Swarada Kanap
 
Intelligent computer aided diagnosis system for liver fibrosis
Intelligent computer aided diagnosis system for liver fibrosisIntelligent computer aided diagnosis system for liver fibrosis
Intelligent computer aided diagnosis system for liver fibrosisAboul Ella Hassanien
 
IRJET- Detection of Breast Asymmetry by Active Contour Segmentation Techn...
IRJET-  	  Detection of Breast Asymmetry by Active Contour Segmentation Techn...IRJET-  	  Detection of Breast Asymmetry by Active Contour Segmentation Techn...
IRJET- Detection of Breast Asymmetry by Active Contour Segmentation Techn...IRJET Journal
 
Survey on Brain MRI Segmentation Techniques
Survey on Brain MRI Segmentation TechniquesSurvey on Brain MRI Segmentation Techniques
Survey on Brain MRI Segmentation TechniquesEditor IJMTER
 
Welcome to International Journal of Engineering Research and Development (IJERD)
Welcome to International Journal of Engineering Research and Development (IJERD)Welcome to International Journal of Engineering Research and Development (IJERD)
Welcome to International Journal of Engineering Research and Development (IJERD)IJERD Editor
 
Hybrid Approach for Brain Tumour Detection in Image Segmentation
Hybrid Approach for Brain Tumour Detection in Image SegmentationHybrid Approach for Brain Tumour Detection in Image Segmentation
Hybrid Approach for Brain Tumour Detection in Image Segmentationijtsrd
 
Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)Vivek reddy
 
Introduction to medical image processing.pdf
Introduction to medical image processing.pdfIntroduction to medical image processing.pdf
Introduction to medical image processing.pdfASMZahidKausar
 
Texture Analysis As An Aid In CAD And Computational Logic
Texture Analysis As An Aid In CAD And Computational LogicTexture Analysis As An Aid In CAD And Computational Logic
Texture Analysis As An Aid In CAD And Computational Logiciosrjce
 

Similaire à Lec7: Medical Image Segmentation (I) (Radiology Applications of Segmentation, and Thresholding) (20)

A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...
A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...
A Dualistic Sub-Image Histogram Equalization Based Enhancement and Segmentati...
 
SIRG-BSU_3_used-important.pdf
SIRG-BSU_3_used-important.pdfSIRG-BSU_3_used-important.pdf
SIRG-BSU_3_used-important.pdf
 
A Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation MethodologyA Novel Efficient Medical Image Segmentation Methodology
A Novel Efficient Medical Image Segmentation Methodology
 
brain tumor detection by thresholding approach
brain tumor detection by thresholding approachbrain tumor detection by thresholding approach
brain tumor detection by thresholding approach
 
Ea4301770773
Ea4301770773Ea4301770773
Ea4301770773
 
Image annotation - Segmentation & Annotation
Image annotation - Segmentation & AnnotationImage annotation - Segmentation & Annotation
Image annotation - Segmentation & Annotation
 
Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...
Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...
Image Segmentation using Otsu's Method - Computer Graphics (UCS505) Project R...
 
Model
ModelModel
Model
 
Segmentation techniques for extraction and description of tumour region from ...
Segmentation techniques for extraction and description of tumour region from ...Segmentation techniques for extraction and description of tumour region from ...
Segmentation techniques for extraction and description of tumour region from ...
 
Intelligent computer aided diagnosis system for liver fibrosis
Intelligent computer aided diagnosis system for liver fibrosisIntelligent computer aided diagnosis system for liver fibrosis
Intelligent computer aided diagnosis system for liver fibrosis
 
radiotherapy.pptx
radiotherapy.pptxradiotherapy.pptx
radiotherapy.pptx
 
IRJET- Detection of Breast Asymmetry by Active Contour Segmentation Techn...
IRJET-  	  Detection of Breast Asymmetry by Active Contour Segmentation Techn...IRJET-  	  Detection of Breast Asymmetry by Active Contour Segmentation Techn...
IRJET- Detection of Breast Asymmetry by Active Contour Segmentation Techn...
 
Survey on Brain MRI Segmentation Techniques
Survey on Brain MRI Segmentation TechniquesSurvey on Brain MRI Segmentation Techniques
Survey on Brain MRI Segmentation Techniques
 
Welcome to International Journal of Engineering Research and Development (IJERD)
Welcome to International Journal of Engineering Research and Development (IJERD)Welcome to International Journal of Engineering Research and Development (IJERD)
Welcome to International Journal of Engineering Research and Development (IJERD)
 
Hybrid Approach for Brain Tumour Detection in Image Segmentation
Hybrid Approach for Brain Tumour Detection in Image SegmentationHybrid Approach for Brain Tumour Detection in Image Segmentation
Hybrid Approach for Brain Tumour Detection in Image Segmentation
 
Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)
 
Image Indexing and Retrieval
Image Indexing and RetrievalImage Indexing and Retrieval
Image Indexing and Retrieval
 
Introduction to medical image processing.pdf
Introduction to medical image processing.pdfIntroduction to medical image processing.pdf
Introduction to medical image processing.pdf
 
Texture Analysis As An Aid In CAD And Computational Logic
Texture Analysis As An Aid In CAD And Computational LogicTexture Analysis As An Aid In CAD And Computational Logic
Texture Analysis As An Aid In CAD And Computational Logic
 
A017350106
A017350106A017350106
A017350106
 

Dernier

Justdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts Service
Justdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts ServiceJustdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts Service
Justdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts Servicemonikaservice1
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .Poonam Aher Patil
 
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticsPulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticssakshisoni2385
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLkantirani197
 
Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Silpa
 
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate ProfessorThyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate Professormuralinath2
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learninglevieagacer
 
GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)Areesha Ahmad
 
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑Damini Dixit
 
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.Nitya salvi
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsSérgio Sacani
 
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxCOST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxFarihaAbdulRasheed
 
chemical bonding Essentials of Physical Chemistry2.pdf
chemical bonding Essentials of Physical Chemistry2.pdfchemical bonding Essentials of Physical Chemistry2.pdf
chemical bonding Essentials of Physical Chemistry2.pdfTukamushabaBismark
 
300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptx300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptxryanrooker
 
Forensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfForensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfrohankumarsinghrore1
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxseri bangash
 
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flypumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flyPRADYUMMAURYA1
 
Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.Silpa
 

Dernier (20)

Justdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts Service
Justdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts ServiceJustdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts Service
Justdial Call Girls In Indirapuram, Ghaziabad, 8800357707 Escorts Service
 
Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .
 
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceuticsPulmonary drug delivery system M.pharm -2nd sem P'ceutics
Pulmonary drug delivery system M.pharm -2nd sem P'ceutics
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
 
Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.Proteomics: types, protein profiling steps etc.
Proteomics: types, protein profiling steps etc.
 
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate ProfessorThyroid Physiology_Dr.E. Muralinath_ Associate Professor
Thyroid Physiology_Dr.E. Muralinath_ Associate Professor
 
module for grade 9 for distance learning
module for grade 9 for distance learningmodule for grade 9 for distance learning
module for grade 9 for distance learning
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)
 
Clean In Place(CIP).pptx .
Clean In Place(CIP).pptx                 .Clean In Place(CIP).pptx                 .
Clean In Place(CIP).pptx .
 
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
High Profile 🔝 8250077686 📞 Call Girls Service in GTB Nagar🍑
 
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
❤Jammu Kashmir Call Girls 8617697112 Personal Whatsapp Number 💦✅.
 
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune WaterworldsBiogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
Biogenic Sulfur Gases as Biosignatures on Temperate Sub-Neptune Waterworlds
 
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxCOST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
 
chemical bonding Essentials of Physical Chemistry2.pdf
chemical bonding Essentials of Physical Chemistry2.pdfchemical bonding Essentials of Physical Chemistry2.pdf
chemical bonding Essentials of Physical Chemistry2.pdf
 
300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptx300003-World Science Day For Peace And Development.pptx
300003-World Science Day For Peace And Development.pptx
 
Forensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdfForensic Biology & Its biological significance.pdf
Forensic Biology & Its biological significance.pdf
 
The Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptxThe Mariana Trench remarkable geological features on Earth.pptx
The Mariana Trench remarkable geological features on Earth.pptx
 
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit flypumpkin fruit fly, water melon fruit fly, cucumber fruit fly
pumpkin fruit fly, water melon fruit fly, cucumber fruit fly
 
Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.Molecular markers- RFLP, RAPD, AFLP, SNP etc.
Molecular markers- RFLP, RAPD, AFLP, SNP etc.
 

Lec7: Medical Image Segmentation (I) (Radiology Applications of Segmentation, and Thresholding)

  • 1. MEDICAL IMAGE COMPUTING (CAP 5937) LECTURE 7: Medical Image Segmentation (I) (Radiology Applications of Segmentation, and Thresholding) Dr. Ulas Bagci HEC 221, Center for Research in Computer Vision (CRCV), University of Central Florida (UCF), Orlando, FL 32814. bagci@ucf.edu or bagci@crcv.ucf.edu 1SPRING 2017
  • 2. Outline • Introduction to Medical Image Segmentation, type of segmentation methods, and definitions – Recognition & Delineation • Simplest Segmentation Method(s): Thresholding – Otsu Thresholding – Parametric Method – PET Image Thresholding Methods • ITM (Iterative Thresholding Method) 2
  • 3. Motivation for Image Segmentation In the last 20 years the computer vision and medical imaging communities have produced a number of useful algorithms for localizing object boundaries in images. 3
  • 4. Motivation for Image Segmentation • Content based image retrieval • Machine Vision • Medical Imaging applications (tumor delineation,..) • Object detection (face detection,…) • 3D Reconstruction • Object/Motion Tracking • Object-based measurements such as size and shape • Object recognition (face recognition,…) • Fingerprint recognition, • Video surveillance • … 4
  • 5. Segmentation Tools in RadiologyApplications • 3D views to visualize structural information and spatial anatomic relationships is a difficult task, which is usually carried out in the clinician’s mind. 5
  • 6. Segmentation Tools in RadiologyApplications • 3D views to visualize structural information and spatial anatomic relationships is a difficult task, which is usually carried out in the clinician’s mind. • Image-processing tools provide the surgeon with interactively displayed 3D visual information. 6
  • 7. Segmentation Tools in RadiologyApplications 7 Credit: Kaus, et al. Radiology 2001.
  • 8. • Determination of the volumes of abdominal solid organs and focal lesions has great potential importance (liver, spleen, …). • Monitoring the response to therapy and the progression of neoplastic disease and preoperative examination of living liver donors are the most common clinical applications of volume determination. 8 Segmentation Tools in RadiologyApplications (credit: Farraher, et al. Radiology 2005)
  • 9. Segmentation Tools in RadiologyApplications • Gross Tumor Volume in CT/MRI • Metabolic Tumor Volume in PET/SPECT/ – Surgery/Therapy Planning • Planning Tumor Volume (PTV) – Tumor characterization • Texture Extraction requires segmentation to be done • Shape analysis 9
  • 10. Segmentation Tools in RadiologyApplications • There is a strong interest in automatic and reproducible techniques for detection and quantification of vascular disease • A first step toward an effective vessel analysis tool is segmentation of the vasculature. 10 axial coronal sagittal Credit: Manniesing, et al, Radiology 2008 MIP: maximum intensity Projection image of cerebral vessels (in CTA)
  • 11. Segmentation Tools in RadiologyApplications • MR volumetry of the hippocampus can help distinguish patients with AD (Alzheimer’s Disease) from elderly controls with a high degree of accuracy (80%–90%). 11
  • 12. Segmentation Tools in RadiologyApplications • MR volumetry of the hippocampus can help distinguish patients with AD (Alzheimer’s Disease) from elderly controls with a high degree of accuracy (80%–90%). 12 hippocampus amygdala Credit: Colliot et al, Radiology 2008.
  • 13. Image Segmentation Definition: Partitioning a picture/image into distinctive subsets is called segmentation. 13
  • 14. Image Segmentation Definition: Partitioning a picture/image into distinctive subsets is called segmentation. 14 Segmentation of an image entails the division or separation of the image into regions of similar attribute.
  • 15. Image Segmentation Definition: Partitioning a picture/image into distinctive subsets is called segmentation. 15 Segmentation of an image entails the division or separation of the image into regions of similar attribute. The most basic attributes: -intensity -edges -texture -other features…
  • 16. Image Segmentation Definition: Partitioning a picture/image into distinctive subsets is called segmentation. 16 Purpose: To extractobject information and representthis as a hard/fuzzygeometric structure. Recognition: Determiningthe object’s whereaboutsin the scene. (humans> computer) Delineation: Determining the object’s spatial extent and compositionin the scene. (computers > humans)
  • 17. Recognition - Example 17 (slice credit: J. Kim et al, Signal Processing 2007) Model is induced No Model is induced
  • 18. Approaches to Recognition 18 • Model-based • Knowledge-based - Non-interactive • Atlas-based • Human-assisted - Interactive
  • 19. Approaches to Recognition 19 • Model-based • Knowledge-based - Non-interactive • Atlas-based • Human-assisted - Interactive - They all originate from human knowledge. - Their relative efficacy is unknown.
  • 20. Approaches to Delineations 20 pI (purely image-based) approaches • Rely mostlyon informationavailable in the given image only. • Recognition: manual
  • 21. Approaches to Delineations 21 pI (purely image-based) approaches • Rely mostlyon informationavailable in the given image only. • Recognition: manual SM (shape model-based) approaches • Employ models to codify object family shape info. • Recognition: model-based/manual
  • 22. Approaches to Delineations 22 pI (purely image-based) approaches • Rely mostlyon informationavailable in the given image only. • Recognition: manual SM (shape model-based) approaches • Employ models to codify object family shape info. • Recognition: model-based/manual Hybrid approaches • Combine among pI and SM approaches. • Recognition: model-based, automatic.
  • 23. Classification of Methods 23 Boundary-based (BpI): • optimum boundary • active boundary • live wire • level sets
  • 24. Classification of Methods 24 Boundary-based (BpI): • optimum boundary • active boundary • live wire • level sets Region-based (RpI): • clustering – kNN, CM, FCM • graph cut • fuzzy connectedness • MRF • watershed • optimum partitioning • (Mumford-Shah)
  • 25. Classification of Methods 25 Boundary-based (BpI): • optimum boundary • active boundary • live wire • level sets Region-based (RpI): • clustering – kNN, CM, FCM • graph cut • fuzzy connectedness • MRF • watershed • optimum partitioning • (Mumford-Shah) SM Approaches • manual tracing • live wire • active shape/appearance • M-reps • atlas-based
  • 26. Classification of Methods 26 Boundary-based (BpI): • optimum boundary • active boundary • live wire • level sets Region-based (RpI): • clustering – kNN, CM, FCM • graph cut • fuzzy connectedness • MRF • watershed • optimum partitioning • (Mumford-Shah) SM Approaches • manual tracing • live wire • active shape/appearance • M-reps • atlas-based Hybrid Approaches • BpI + BpI • RpI + RpI • BpI + RpI • BpI + SM • RpI + SM • SM + SM
  • 27. Classification of Methods 27 pI Approaches + Where image info is good, accuracy is good; - Bad where it is poor/absent; - Need recognition help; + Can determine degree of match of model to image well; - Lack obj shape & geographic info;
  • 28. Classification of Methods 28 SMApproaches - Even where image info is good, accuracy suffers; + Where bad, model helps; + Can help in recognition; - Need best match info; + Good models embody obj shape & geographic info;
  • 29. Purely Image Based Segmentation Methods 29
  • 30. Thresholding – Simple Segmentation • Image binarization – mapping a scalar image I into a binary image J 30 J(x, y) = ( 0 if I(x, y) < T 1 otherwise.
  • 31. Thresholding – Simple Segmentation • Image binarization – mapping a scalar image I into a binary image J 31 J(x, y) = ( 0 if I(x, y) < T 1 otherwise.
  • 32. Thresholding – Simple Segmentation 32 Brighter objects Darker objects
  • 33. Thresholding – Simple Segmentation 33 Brighter objects Darker objects DIFFICULTIES 1. The valley may be so broad that it is difficult to locate a significant minimum 2. Number of minima due to type of details in the image 3. Noise 4. No visible valley 5. Histogram may be multi-modal
  • 39. Thresholding Methods • Huang • Intermode • Isodata • Li • MaxEntropy • Mean • MinError • Otsu • Percentile • RenyiEntropy • Moments 39
  • 40. Thresholding Methods • Huang • Intermode • Isodata • Li • MaxEntropy • Mean • MinError • Otsu • Percentile • RenyiEntropy • Moments 40
  • 41. Thresholding Methods PET Imaging Fixed Thresholding Adaptive Thresholding Iterative Thresholding 41 • Huang • Intermode • Isodata • Li • MaxEntropy • Mean • MinError • Otsu (non-parametric) • Percentile • RenyiEntropy • Moments
  • 42. Otsu Thresholding • Definition: The method uses the grey-value histogram of the given image I as input and aims at providing the best threshold in the sense that the “overlap” between two classes, set of object and background pixels, is minimized (i.e., by finding the best balance). 42
  • 43. Otsu Thresholding • Definition: The method uses the grey-value histogram of the given image I as input and aims at providing the best threshold in the sense that the “overlap” between two classes, set of object and background pixels, is minimized (i.e., by finding the best balance). • Otsu’s algorithm selects a threshold that maximizes the between-class variance . In the case of two classes, 43 2 b 2 b = P1(µ1 µ)2 + P2(µ2 µ)2 = P1P2(µ1 µ2)2
  • 44. Otsu Thresholding • Definition: The method uses the grey-value histogram of the given image I as input and aims at providing the best threshold in the sense that the “overlap” between two classes, set of object and background pixels, is minimized (i.e., by finding the best balance). • Otsu’s algorithm selects a threshold that maximizes the between-class variance . In the case of two classes, • where P1 and P2 denote class probabilities, and μi the means of object and background classes. 44 2 b 2 b = P1(µ1 µ)2 + P2(µ2 µ)2 = P1P2(µ1 µ2)2
  • 45. Otsu Thresholding • Definition: The method uses the grey-value histogram of the given image I as input and aims at providing the best threshold in the sense that the “overlap” between two classes, set of object and background pixels, is minimized (i.e., by finding the best balance). 45 P1 = uX ı=0 p(i) P2 = GmaxX ı=u+1 p(i) u u
  • 46. Otsu Thresholding • Definition: The method uses the grey-value histogram of the given image I as input and aims at providing the best threshold in the sense that the “overlap” between two classes, set of object and background pixels, is minimized (i.e., by finding the best balance). 46 P1 = uX ı=0 p(i) P2 = GmaxX ı=u+1 p(i) µ1 = uX ı=0 ip(i)/P1 µ2 = GmaxX ı=u+1 ip(i)/P2 CLASS MEANS
  • 47. Otsu Thresholding-Algorithm 47 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function.
  • 48. Otsu Thresholding-Algorithm 48 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function. 2 b = P1(µ1 µ)2 + P2(µ2 µ)2 = P1P2(µ1 µ2)2
  • 49. Otsu Thresholding-Algorithm 49 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function.
  • 50. Otsu Thresholding-Algorithm 50 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function.
  • 51. Otsu Thresholding-Algorithm 51 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function.
  • 52. Otsu Thresholding-Algorithm 52 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function.
  • 53. Otsu Thresholding-Algorithm 53 cI (u) 1 cI(u) P1 P2 c indicates cumulative histogram,and P1 and P2 can be approximated well with cumulative density function. optimal
  • 54. Parametric Method for Optimal Thresholding • Assuming again a two-class problem and assuming that the distribution of gray levels for each class can be modeled by a normal distribution with mean and variance 54
  • 55. Parametric Method for Optimal Thresholding • Assuming again a two-class problem and assuming that the distribution of gray levels for each class can be modeled by a normal distribution with mean and variance • the overall normalized intensity histogram can be written as the following mixture probability density function: 55
  • 56. Parametric Method for Optimal Thresholding • Assuming again a two-class problem and assuming that the distribution of gray levels for each class can be modeled by a normal distribution with mean and variance • the overall normalized intensity histogram can be written as the following mixture probability density function: where P1 and P2 are class probabilities. The optimal threshold (T) can be found as solving the quadratic equation à 56
  • 57. Parametric Method for Optimal Thresholding 57
  • 58. Parametric Method for Optimal Thresholding 58 In case, variances of both classes are equal, then->
  • 59. Parametric Method for Optimal Thresholding 59 In case, variances of both classes are equal, then->
  • 60. Thresholding methods for PET Image Segmentation • Due to the nature of PET images (i.e., low resolution with high contrast), thresholding-based methods are suitable – because the local or global intensity histogram usually provides a sufficient level of information for separating the foreground (object of interest) from the background. (Foster, Bagci, et al., CBM 2014) 60
  • 61. Thresholding methods for PET Image Segmentation • Due to the nature of PET images (i.e., low resolution with high contrast), thresholding-based methods are suitable – because the local or global intensity histogram usually provides a sufficient level of information for separating the foreground (object of interest) from the background. (Foster, Bagci, et al., CBM 2014) 61 Fixed Thresholding Adaptive Thresholding Iterative Thresholding
  • 62. Fixed Thresholding Methods • Due to the nature of PET images (i.e., low resolution with high contrast), thresholding-based methods are suitable – because the local or global intensity histogram usually provides a sufficient level of information for separating the foreground (object of interest) from the background. (Foster, Bagci, et al., CBM 2014) 62
  • 63. Thresholding methods for PET Image Segmentation • Due to the nature of PET images (i.e., low resolution with high contrast), thresholding-based methods are suitable – because the local or global intensity histogram usually provides a sufficient level of information for separating the foreground (object of interest) from the background. (Foster, Bagci, et al., CBM 2014) 63 Fixed Thresholding Adaptive Thresholding Iterative Thresholding Phantom Based Image Quality metrics based
  • 65. Thresholding methods for PET Image Segmentation • Due to the nature of PET images (i.e., low resolution with high contrast), thresholding-based methods are suitable – because the local or global intensity histogram usually provides a sufficient level of information for separating the foreground (object of interest) from the background. (Foster, Bagci, et al., CBM 2014) 65 Fixed Thresholding Adaptive Thresholding Iterative Thresholding Phantom Based Image Quality metrics based
  • 66. Iterative Thresholding Method (ITM) 66 S/B: Source to background ratio. The method is based on calibrated threshold-volume curves at varying S/B ratio acquired by phantom measurements using spheres of known volumes.
  • 67. Iterative Thresholding Method (ITM) 67 S/B: Source to background ratio. The method is based on calibrated threshold-volume curves at varying S/B ratio acquired by phantom measurements using spheres of known volumes.
  • 68. Iterative Thresholding Method (ITM) 68 S/B: Source to background ratio. The method is based on calibrated threshold-volume curves at varying S/B ratio acquired by phantom measurements using spheres of known volumes. The measured S/B ratios of the lesions are then estimated from PET images, and their volumes are iteratively calculated using the calibrated S/B-threshold-volume curves
  • 69. Iterative Thresholding Method (ITM) 69 S/B: Source to background ratio. The method is based on calibrated threshold-volume curves at varying S/B ratio acquired by phantom measurements using spheres of known volumes. The measured S/B ratios of the lesions are then estimated from PET images, and their volumes are iteratively calculated using the calibrated S/B-threshold-volume curves The resulting PET volumes are then compared with the known sphere volume and CT volumes of tumors that served as gold standards.
  • 70. ITM Example Result on PET Images/Lung 70
  • 71. Another Example for PET Thresholding 71 ITM for tumor segmentation/FDG PET
  • 72. Another Example for PET Thresholding 72
  • 73. Further Thresholding Example – CT Bones 73
  • 74. Further Thresholding Example – CT Bones 74
  • 75. Head-Neck CT – Thresholding for Skull Modeling 75 (Slice Credit: P.Seutens) Segmentation of the skull and the mandibula in CT images using thresholding.(a) Original CT image of the head. (b) Result with a threshold value of 276 Hounsfield units. The segmented bony structures are represented in color. (c) 3D rendering of the skull shows a congenital growth deficiency of the mandibula in this 8-year-old patient. This information was used preoperatively to plan a repositioning of the mandibula.
  • 76. Multiple Thresholds – MRI Thresholding 76 Thresholding can be done interactively and separates the image into different regions. Valleys in the histogram indicate potentially useful threshold values Credit: Toeonies,K.
  • 77. Summary of today’s lecture • Introduction into the Medical Image Segmentation • Recognition and Delineation concepts in Segmentation • Simplest Segmentation method: Thresholding – Otsu – Parametric method for optimal thresholding – PET Image thresholding • ITM, fixed thresholding,etc. 77
  • 78. Slide Credits and References • Jayaram K. Udupa, MIPG of University of Pennsylvania, PA. • P. Suetens, Fundamentals of Medical Imaging, Cambridge Univ. Press. • Foster, B., et al. CBM, Review paper, 2014. • Kaus, et al. Radiology 2001. • Toeonies, K., Medical Image Analysis. • Farraher, et al., Radiology 2005 • Zaidi, H., Quantitative Analysis in Nuclear Medicine Imaging. • Bailey et al. Positron Emission Tomography, Springer. • Dawood, M., et al. Correction Techniques in Emission Tomography 78