Medical Image Processing - PowerPoint PPT Presentation

1 / 19
About This Presentation
Title:

Medical Image Processing

Description:

The development of appropriate technology to extract clinically useful ... Wilhelm Konrad R ntgen. reported discovery of new 'rays' (Nobel Prize in physics in 1901) ... – PowerPoint PPT presentation

Number of Views:8805
Avg rating:5.0/5.0
Slides: 20
Provided by: coro2
Category:

less

Transcript and Presenter's Notes

Title: Medical Image Processing


1
Medical Image Processing
CVIP Lab
The University of Louisville
M. Sabry Hassouna Ph.D. Computer Vision Image
Processing Laboratory (CVIP) Louisville, Kentucky
2
Medical Imaging
  • The study of medical imaging is concerned with
  • Interaction of all forms of radiation with
    tissue.
  • The development of appropriate technology to
    extract clinically useful information, usually in
    the form of an image from the observed
    technology.

3
X-Ray
  • Image Formation
  • A beam of X-rays is directed through a patient
    onto a film.
  • The film provides a measure of the ray
    attenuation in tissue.

Nov 8, 1895 Wilhelm Konrad Röntgen reported
discovery of new rays (Nobel Prize in physics
in 1901). Jan 13, 1896 First clinical use of
X-rays by 2 British doctors to find a needle in
a hand.
Excellent for imaging bones. - No depth
information, bad for soft tissue, excessive
radiation
4
Sample X-Ray Slices
5
Computed Tomography (CT)
Image Formation The object is viewed from a
number of different angles and then a
cross-sectional image of it can be computed
(reconstructed).
Nov 8, 1895 G. Hounsfield (computer expert) and
A.M Cormack (physicist) (Nobel Prize in Medicine
in 1979).
Provide 3D anatomical information Preserves
topology (bones) - Excessive radiation - Not good
for all soft tissues
6
Sample CT Slices
7
CT Acquisition Techniques
3D Reconstruction
8
Magnetic Resonance Imaging (MRI)
Image Formation - Hydrogen nuclei (protons) under
a strong magnetic field spin in phase with one
another and align with the field. - Relaxed
protons induce a measurable radio signal.
1952 F. Bloch and E. Purcel, extended by R.
Ernst) (Bloch Purcel Nobel Prize in Physics in
1952) (Ernst Nobel Prize in Chemistry in 1991)
Main modality for image guided surgery.
Superb ability to discriminate between subtle
differences in tissue characteristics. Very
safe. - Less accurate for bone scanning.
9
Sample MRI Slices
10
Ultrasound (US)
Image Formation An ultrasonic energy is
propagated into the patient from a transducer
placed on the skin and back-scattered echo signal
is recorded by the same transducer.
1979 Samuel H. Maslak
Noninvasive Clean safe In-expensive -
Noisy - Gas filled and bony structures cannot be
imaged because they absorb ultrasound waves.
11
Positron Emission Tomography (PET)
Image Formation - Detection of radiation from the
emission of positrons.
1998-2001 Dr. David Townsend and Dr. Ron Nutt.
Valuable technique for some diseases and
disorders. Amount of radiation is small -
Invasive (inject radioactive material)
12
Magnetic Resonance Angiography (MRA)
Image Formation Imaging the blood vessels (moving
spins) using MRI scanner.
  • TOF (time-of-flight)
  • PCA (phase contract angiography)
  • (CEA) (contrast enhanced angiography)
  • (CTA) (computer tomography angiography)
  • (DSA) (digital subtraction angiography)

DSA
CTA
TOF
13
Medical Image Processing
14
Data Processing
1. Preprocessing Filtering, registration 2.
Detection Finding objects (nodules, polyps,
organs) 3. Segmentation Exact delimitation of
objects 4. Analysis Measurement (volume,
curvature) 5. Classification/diagnoses
15
Preprocessing
Original
Enhanced
16
Detection
Find location of objects of interest without
prior knowledge about their location/existence
Bones Organs Polyps in colon Nodules in
lungs
17
Segmentation
Exactly delimitate objects, once they are
detected. - Vessels - Liver - Cardiac imaging
(left ventricle) - Brain
18
Analysis
Measurement Volume - growth Vessel
stenosis Functional imaging Stroke
Cardiac perfusion Tumor perfusion Cardiac
function LV motion Injection fraction
19
Classification / Diagnoses
Comparison to developed atlases Use of
knowledge databases Classify as normal/abnormal
(brain) Classify lung nodules as
benign/malignant Determine cancer/non-cancer
Write a Comment
User Comments (0)
About PowerShow.com