A Brief Introduction to Medical Imaging

Outline
• General Goals
• Linear Imaging Systems
• An Example, The Pin Hole Camera
• Radiations and Their Interactions with Matter
• Coherent vs. Incoherent Imaging
• Length Scales
• Contrasts
• Photon Intensity Tomography
• Magnetic Resonance Imaging
 Imaging Definitions

Object function - the real space description

of the actual object.
Resolution - the collected image is only an
approximation of the actual
object. The resolution
describes how accurate the
spatial mapping is?
Distortions - describes any important non-
linearities in the image. If
there are no distortions, then
the resolution is the same
everywhere.
Fuzziness - describes how well we have described the object we wish to image.
Contrast - describes how clearly we can differentiate various parts of the object
in the image.
Signal to Noise ratio
 Linear Imaging Systems
If the blurring of the object function that is introduced by the imaging processes is
spatially uniform, then the image may be described as a linear mapping of the
object function.
This mapping is, of course, at lower resolution; and the blurring is readily described
as a convolution of the object function with a Point Spread Function.
Image = object ⊗ Point Spread Function + noise
The noise is an important consideration since it limits the usefulness of
deconvolution procedures aimed at reversing the blurring effects of the image
measurement.
 An Example, the Pin-hole Camera
One of the most familiar imaging devices is a pin-hole camera.

a b

object function pin-hole image

source

The object is magnified and inverted. Magnification = -b/a.

 An Example, the Pin-hole Camera 2

Notice, however, that the object function is also blurred due to the
finite width of the pin-hole.
a b

object function pin-hole image

source

The extent of blurring is to multiply each element of the source by the

“source magnification factor” of (a+b)/a x diameter of the pin-hole.
 Distortions of a Pin-hole Camera

Even as simple a device as the pin-hole camera has distortions

1. Limited field of view due to the finite thickness of the screen.
a b

object function pin-hole image

source

As the object becomes too large, the ray approaches the pin-hole too
steeply to make it through.
 Distortions of a Pin-hole Camera 2

Also, as the object moves off the center line, the shadow on the
detector grows in area, (and the solid angle is decreased) so the image
intensity is reduced.
a b

object function pin-hole image

source
Spatial Frequencies

 Spatial Frequencies 2

When discussing linear imaging systems it is often useful to describe the

measurement in terms of a mapping of Fourier components of the object function.
 Transmission Tomography

source y r
Absorption imaging
s φ

φ
x
r
object detector

In absorption imaging, the integrated absorption along a column through the object is
measured. An array of detectors therefore measures a ‘shadow profile’.
Projection Imaging

Object
Projections
 Central Slice Theorem

Consider a 2-dimensional example of an emission imaging system. O(x,y) is the

object function, describing the source distribution. The projection data, is the line
integral along the projection direction.
P (0 o , y )= ∫ O( x, y)dx
The central slice theorem can be seen as a consequence of the separability of a 2-D
Fourier Transform.
o˜(kx , k y )= ∫ O(x, y)e −ikx x −iky y
e dxdy

The 1-D Transform of the projection is,

p˜ (ky ) = ∫ P(0 , y )⋅ e
o −iky y
dy
= ∫∫ O(x, y ) ⋅ e
−iky y
dxdy
= ∫∫ O(x, y ) ⋅ e
−iky y
⋅ e−i0x dxdy
= o˜ (0, k )
y

The one-dimensional Fourier transformation of a projection obtained at an angle

J, is the same as the radial slice taken through the two-dimensional Fourier
domain of the object at the same angle.
 Transmission Tomography

source y r
Absorption imaging
s φ

φ
x
r
object detector

In absorption imaging, the integrated absorption along a column through the object is
measured. An array of detectors therefore measures a ‘shadow profile’.
 Coherent vs. Incoherent Imaging

In both cases the image is the result of the scattering of a field by the object.
Incoherent - measure only the intensity fluctuations of this scatter. Usually
frequencies are too high to permit convenient measures of the phase. Examples,
light 1014 Hz
X-rays 1018 Hz
γ-rays 1020 Hz
A photograph is an incoherent image.
Coherent - measure both the intensity and the phase of the scattered field. This is
usually measured as a temporal evolution of the scattered field. The frequency of
radiation is normally quite low to permit an accurate measure of the phase (such as
microwaves). MRI is an example of coherent imaging.
----------------------------------------------------------------------------------------------------
Incoherent images are most readily measured by scanning a well collimated beam
across the sample and observing the attenuation of that beam (this may be
multiplexed with many detectors).
Coherent images permit the characterization of the entire sample at once and with
observation through a single detector element. A series of measurements are made
for fields of varying frequency or direction.
 X-ray Attenuation Coefficients

µ/ρ
(cm2/g)

2.5
2

1.0
BONE

0.5
0.4 MUSCLE
0.3
FAT
0.2
0.15
PHOTON
0.1 ENERGY
10 20 30 40 50 100 150 200 300 500 (kev)

X-ray attenuation coefficients for muscle, fat, and bone, as a function of photon
energy.
 Binding energy (KeV) Characteristic
Binding energy (KeV) X-ray
66 KeV
≈0
λ4 0.6 KeV (N→M)
≈0
Photoelectron
0.6 λ3 4.4 KeV (M→L)
0.6
100 KeV 5 5
λ2 29 KeV (L→K)
incident Total 34 KeV
photon 34 34

K K
λ1 L L
M M
N N

Valence electrons

K
L Compton
M Electron (Ee-)

Incident
photon
(E0)
θ Angle of deflection
λ1 λ2
λ1 < λ2 Scattered
Photon (Esc)
 Attenuation Mechanisms

µ Dependence

Mechanism E Z Energy Range in

Soft Tissue
Simple Scatter 1/E Z2 1-20 keV
Photoelectric 1/E3 Z3 1-30 keV
Compton falls slowly with E independent 30 keV-20 MeV
Pair Production rises slowly with E Z2 above 20 MeV
 Attenuation Mechanisms 2

Attenuation
(log plot)

total
Compton

Compton
photoelectric
simple scatter pair

.01 .05 0.1 1 10 Photon energy (MeV)

.03 1.02 30 (log plot)
Attenuation mechanisms in water

The optimum photon energy is about 30 keV (tube voltage 80-100 kV) where the
photoelectric effect dominates. The Z3 dependence leads to good contrast:
Zfat 5.9
Zmuscles 7.4
Zbone 13.9
⇒ Photoelectric attenuation from bone is about 11x that due to soft tissue, which is
dominated by Compton scattering.
 Photon Intensity Tomography

X-ray CT SPECT PET

measuring X-ray source distribution of source distribution of

attenuation radio-pharmaceuticals, radio-pharmaceuticals,
gamma emitters positron emitters

anatomical Yes No No
information

beam definition collimators collimators coincidence detection

 Photon Intensity Tomography 1

source y r
Absorption imaging
s φ

φ
x
r
object detector

In absorption imaging, the integrated absorption along a column through the object is
measured. An array of detectors therefore measures a ‘shadow profile’.
 Photon Intensity Tomography 2

y r
Emission imaging
s φ

φ
x
r
object
detector

In emission imaging, the integrated emitter density is measured.

 Spatial and Temporal Limits in NMR

1m Space MRI absolute

medical spatial
measurements
Mini-Imaging
1cm
1mm
micro-imaging

NMR-microscopy

1µm

1nm
°
1A

1µs 1ms 1s
Time
 The Bloch Equations

d M
M x = ∆ω ⋅ M y − x
dt T2
d My
M y = −∆ω ⋅ M x + ω1 (t ) ⋅ M z −
dt T2
d M − M0
M z = −ω1 (t ) ⋅ M y − z
dt T1

ω1 is the strength of an applied external resonant radio-frequency field. ∆ω is the

precession frequency; it includes contributions from
• variations in magnetic field strength (inhomogeneities),
• applied magnetic field gradients,
• chemical shifts (screening of the nucleus by surrounding electrons),
• and coupling of spins to each other (the dynamics are more complicated
than indicated by the Bloch equations however).
 Spin Magnetization Gratings
Grating - “a system of equidistant and parallel lines… to produce spectra by
diffraction”.
Spin magnetization grating - a periodic modulation of the phase (or amplitude) of the
local spin magnetization vector superimposed on the spin density
Spin Magnetization gratings may be created by spin evolution in a linearly
increasing magnetic field. RF
G
k
k=0

u
u u
2π
k

Mx Mx Mx
My My My

This produces a grating as a linear phase ramp, since motions are torques.
Moire complex gratings are produced through a combination of RF and gradient
pulses. The spatial frequency distribution of these are describable by a distribution of
components, each at a given wave-number.
 Mz(t)
My(t)
time
Mx(t)

magnetic field strength

precession frequency

spatial offset

precession 2t
angle
t
t=0
spatial offset
 Mz(t)
My(t)
time
Mx(t)

Diagrams of the spin magnetization’s return to equilibrium after being aligned along
the x-axis. In both pictures the evolution of a single bulk magnetization vector is
being followed. The initial position is shown as the green vector at top, which spirals
into the z-axis, the red vector. In the figure on the right, the three individual
components of the magnetization are shown as a function of time. The NMR
experiment measures the two transverse components, Mx and My. There are three
motions, a precession about the z-axis, a decay of the transverse components and a
slower growth along z towards the static equilibrium value.
 G = 1 G/cm
1.0
attenuation D = 3×10-5 cm2/s
0.8
T2 = 0.1 s
0.6
0.4
0.2
k
500 1000 1500 2000 2500
k
t=
γG
k t
− −
γGT2
attenuation T2 = e ; e T2

k 3D k 2 Dt
− −
3γG
attenuation D=e ; e 3