Medical Imaging
& Applications
Concept, Processing, Analysis and Machine Vision
Course Content ?
Theme Elements
Introduction ● Human physiology
● Anatomy
● Imaging Systems and Modalities
● Healthcare Management
● Medical Communications
Radiology ● X-Ray Imaging
● Image ● CECT
Processing ● USG
● Machine Association ● MRI
Learning ● Nuclear Imaging
● Deep Learning
Pathology Digital Macro and Microscopy
● Computational Tissue Imaging
Models
Marks Distributions and Materials
➔ Project/Case Study - 50% (In the group of 2/3)
● State of the art 5%
● Objectives 5%
● Methodologies 10%
● Implementation/Results 15%
● Q/A 15%
➔ One Subjective Exams - 20%
➔ One Quiz (MCQ) - 15%
➔ Home Assignment - 15%
HealthCare Systems: Today & Trends
Anatomy OR Physiology …….. Difference?
Structural Organization
● Phase-Contrast Microscopy
● Confocal Microscopy
● Fluorescent Imaging
● Optical Imaging
● OCT
● CT
● MRI
● Ultrasonography
● Optical Imaging
● PET
● SPECT
● Microscopy
Clinical Workflow
Clinical Features/Parameters
—> Patients General History, Family data
—> Signs
—> Symptoms
—> Observations
—> Radiology
—> Endoscopy
—> Pathology
—> Biochemistry
—> Microbiology
—> Haematology
Diagnosis OR Prognosis…….. Difference ????
A diagnosis is the determination of the
cause of a person’s symptoms and
signs.
● Gathering information; symptoms and medical
history
● Performing tests and procedures; identify the
cause of the symptoms
● Confirming the diagnosis ; additional testing
or by observing the response to treatment.
A prognosis is a prediction of the course
of a person’s disease based on an
understanding of the natural history of
the disease and the patient’s current
state of health
Radiology vs Pathology vs Histology
Radiology
Methodology: Uses ionizing radiation to create images of body parts.
Purpose: Diagnoses body conditions, internal responses, cellular activities,
bone/joint functioning, and tissue inflammation.
Practitioner: Conducted by radiologists trained in operating ionizing
radiations and interpreting resulting images.
Pathology
Methodology: Involves physical examination of tissue samples,
cell samples, surgical specimens, and bodily fluids.
Purpose: Diagnoses diseases and guides treatment decisions
based on the examination of samples.
Practitioner: Performed by pathologists who specialize in various
subfields.
Treatment Management: Protocol
Analysis Altogether
Medical - Modalities- CT, MRI, USG….
Imaging - Organ appearance
- Tissue-energy interaction
Physics - Image formation
- Statistics
- Denoising, Segmentation,
Image - Feature Extraction
Processing - Classification, Visualization
- Prediction
Machine - Example based learning
Learning - Complex reasoning
Digital Image
Image
- A Multi-dimensional function of spatial coordinates
- 2D (x,y) >> Still Images
- 3D (x,y,z)/(x,y,t) >> CT/Video
Illumination
Reflectance
Digitization
- The continuous image on the image plane must be converted into
image points on a discrete grid. Furthermore, the intensity at each
point must be represented by a suitable finite number of gray values
(Quantization).
Digital Image
?
Why Digitization?
3
2
Electromagnetic Spectrum
Beyer filter arrays
( 50% G, 25% B, 25% R)
Image Formation: Acquisition
Light source strength and
direction
Surface geometry, nearby Image
Sensor property
environment and material representation
Image Formation: Color Model
● RGB (Red, Green, Blue)
● CMYK (Cyan, Magenta, Yellow, Black)
● HSV/HSL (Hue, Saturation, Value/Luminance)
● YCbCr ( Luminance, Blue Difference, Red Difference Chroma)
● XYZ (Y=Luma, X /Z = Chromatic components
● YUV (Y=Luma, U= blue projection(B-Y), V=Red Projection(R-Y)
● Lab (L=Luma, a=G to R and b = B to Y chromatic component)
Elements of Visual Perception
Ciliary
muscles
Three membranes:
(i) Cornea and Sclera
Spherical Diameter
(ii) Choroid of ~20 mm
a) Iris Diaphragm
b) Lens
c) Ciliary muscles
(iii) Retina
a) Cones
b) Rods
c) Blind spots
Elements of Visual Perception Protects posterior
surface
Ciliary
muscles Screen to form image by
Provides pigmentation lens
Contains photoreceptors
(Rods,Cones)
Protects anterior Dense with cones
surface Spatial vision
Made of concentric Devoid of receptors
fibrous cells Point where optic
nerve emerges
Visual axis
Spherical Diameter
of ~20 mm
Image Formation: Human Color Perception
The human eye has over 100 million
Scotopic Vision
rod cells.
The human eye only has about 6
million cones. Photopic Vision
Camera Model: Perspective Projection
Reconstruction
Self Study!!!!!!!!!!!
Image : Closed View
?
No of Gray Levels : L=2K
Image Properties : Resolution
Spatial Resolution
Gray level/Intensity Resolution
Image Properties : Resolution
Image Properties : Resolution
Image Properties : Sharpness, Contrast, Brightness
Sharpness
Sharpness describes the clarity of detail in a photo
Contrast
The range of brightness, from lightest to darkest, in an image
Image Properties : Sharpness, Contrast, Brightness
Sharpness
Contrast
Sharpness describes the clarity of detail in The range of brightness, from lightest to darkest, in an
a photo image
Brightness
Brightness refers to the
overall lightness or darkness
of the image
Image Properties : Sharpness, Contrast, Brightness
a) The brightness of the original image
was set at 0, with the lowest and highest
brightness being −100 and 100,
respectively.
b) The contrast of the original image
was set at 0, with the lowest and highest
contrast being −100 and 100,
respectively.
Image Components : Shape, Size, Texture,Context
Shape
Contour based
The shape of an object refers to its Region based
physical structure and can be
represented by the boundary,
region, moment, etc.
The size of objects in an image
is a function of scale.
Tone refers to the relative
Texture brightness or color of
objects in an image.
Texture refers to the arrangement Refers to repetition of basic texture elements
and frequency of tonal variation in called texels. A texel contains several pixels,
particular areas of an image. whose placement could be periodic or random
Image Components : Shape, Size, Texture,Context
Context
vs.
Content
Context can be usually obtained from its Content features are based on either foreground,
annotations or descriptions. background information or their combination.
Statistical Parameters : Mean, Median, Moment
Mean The sum of all pixel values divided by
the total number of pixels
Median Arrange the data points from
smallest to largest and replace the
center pixel value by middle value of
series
Moment A weighted average of the intensities
of an image's pixels.
Statistical Parameters : STD, Skewness, Kurtosis
Standard
Deviation
Standard deviation measures the
dispersion of a dataset relative to
its mean.
Image Operations ????
? ?
HOW ?
?
Image Operations ????
Convolution and Correlation
Image Operations : Convolution and Correlation
Correlation is used to determine the
degree of similarity between two signal
Correlation
Convolution
(180° flip)
Convolution is used to merge the effect
of two signal to get third signal
Correlation
Convolution
● Filtering, Detection
● Template Matching, Feature Matching
1
2
Image Histogram
Graphical
representation of an
image
x>>> Range
y>>>Freq./Prob. of Occ.
Imaging File Format : Significance and Uses
● Image Format describes how data related to the image will be stored.
● Data can be stored in compressed, Uncompressed, or vector format.
1. TIFF(.tif, .tiff) Tagged Image File Format.
2. GIF (.gif) GIF or Graphics Interchange Format files are used for web graphics.
3. PNG (.png) PNG or Portable Network Graphics files
4. JPEG (.jpg, .jpeg) Joint Photographic Experts Group
5. RAW
6. PDF Lempel-Ziv-Welch Algorithm DICOM
Wavelet (Digital Imaging and Communications in Medicine)
DCT NIFTI
(Neuroimaging Informatics Technology Initiative)
Huffman
MHA
Karhunen-Loève Transform (KLT) (MetaImage Medical Format)
Imaging File Format : Significance and Uses
DICOM
(Digital Imaging and
Communications in Medicine)
NIfTI
(Neuroimaging Informatics
Technology Initiative)
MHA
(MetaImage Medical Format)
Machine Vision In HealthCare
Medical Imaging
● Visualization of internal body organs, tissues or cavities using specialized
instruments
● Structural and functional imaging of human body for understanding
○ Body anatomy
○ Physiological processes
○ Function of organs
○ Behavior of organ in abnormal physiological condition
● Different Modalities are available to capture several views of the body organ
● in order to diagnose, monitor and treat medical conditions.
Most common modalities are: X-ray, CT, PET, MRI, Ultrasound and SPECT
Medical Imaging : Planes
1. Axial Plane (Transverse/Horizontal Plane)
2. Sagittal Plane (Longitudinal Plane)
3. Coronal Plane (Frontal Plane)
A-P : Anterior- Posterior (Front/Back)
I-S : Inferior - Superior (Bottom/Top)
R-L : Right - Left
Modalities Visualizing + Quantification
Image Quality Processing
Resolution Analysis
X-Rays PET
CT SPECT
MRI fMRI
USG Angiography
Optical Imaging
Mammography Fluoroscopy
Thermal Imaging
Morphology Function Blood Flow
Organ level Physiology Vascular function
Tissue level Biochemistry
Cellular level Endoscopy ?
Modalities: Imaging types
Image Communication Protocols
PACS
● PACS, or Picture Archiving and Communication System, is a medical imaging technology that
provides economical storage and convenient access to images from multiple modalities.
● Medical imaging equipment, including CT, MRI, X-ray, ultrasound and
more |Acquisition|
● DICOM mobile devices or workstations for viewing, re-processing and
interpreting medical images |Display|
● It has printers and archives for storage and retrieval of images,
related documentation and reports |Storage and Report taking|
● A secure network for the distribution and exchange of patient
examining data |Server and Network|
