Transcript History of CT - Nuclear Medicine Review

HISTORY OF CT
Why CT???
1.
2.
3.
Deals with the issue of superimposition
of structures
Provides excellent low-contrast
resolution because of beam geometry
and sensitive detectors
Spiral/Helical volume data acquisition
leading to major imaging innovations
(MPR, 3D, etc.)
Basic Principles of CT

X-ray beam is passed at a cross section through a patients body. This
eliminates superimposition

The beam is finely collimated this reduces scatter and gives better
contrast resolution

The collimated beam pass thru the body , the body tissue absorbs the
beam.

The beam exits the body and strikes the detectors. The detectors are
quantitative and distinguish differences in tissue contrast

Detector converts photons to a analog signal

The ADC converts it to digital signal

Digital data is sent to CPU for reconstruction
History of CT
Johann Radon - Theory
was actually developed
in 1917
 Showed that an image
can be reconstructed
of a 2 or 3 dimensional
object from a large
number of projections
stemming from
different directions
◦ Referred to Astronomy
et al
History of CT
Godfrey Hounsfield
1967 – Researched x-ray
beam being passed
through object in all
directions while
obtaining
measurements of
transmission that
information about
internal structures can
be obtained and
presented in 3d
representation
History of CT


Through the 1960’s,
mathematicians
continued to investigate
possibility of image
reconstruction in
medicine
1963 – Alan Cormack
(r.) was able to apply
techniques in nuclear
medicine for such which
was then seen as the
solutions to the
mathematical problems
developed at the advent
of the CT scanner
History of CT
Original “CT” Scanner
- Did not utilize x-ray
- Gamma Source coupled
with a single crystal
detector
-
9 days to scan object
- Extremely low radiation
output
-
-
Computer utilized
processed 28,000
measurements in 2.5
hours
When decision was made
to replace radiation
source with x-ray tube
cut time down to 1 day…
History of CT

Early experiments of brain tissue in
conjunction with a radiologist showed the
ability to differentiate between tumor tissue
and gray and white brain matter
◦ Also able to differentiate details like ventricles
and pineal gland

1971 – First clinical prototype CT Brain
Scanner was installed in England under
direction of Dr. James Ambrose
◦ Processing time = 20 minutes
◦ Minicomputer introduction = 4.5 minutes
History of CT

1972 – First patient
scanned
◦ Suspected brain lesion
which turned out to be
a large cyst

1979 – Hounsfield and
Cormack receive
Nobel Prize in
Medicine for their
contributions to the
development of CT
History of CT
1974 – Dr. Robert
Ledley
◦
First whole body CT
scanner
1975 – Dynamic Spatial
Reconstructor (DSR)
◦ Image dynamics of organ
system with high spatial
resolution – Advanced
speed of scanner
1983 – Electron Beam
CT Scanner
introduced
◦ First cardiac imager
History of CT
1989 – First practical spiral CT
scanner introduced at RSNA
◦ Dr. Willi Kalender
◦ Single Slice spiral/helical
◦ Allowed volumetric scanning
which allowed scanning larger
volumes in less time
1998 – Multislice CT introduced
◦ 4 or more slices per revolution
2005 – Dual Source CT
◦
Developed by Siemens
◦ Advanced cardiac imaging by
utilizing 2 x-ray tubes with 2
detector arrays
 Between the beats of the heart…
AND ON, AND ON, AND
ON…
Generation 1
CT Generations
First-Generation Systems:
 Original scan geometry used by Hounsfield
 Set of parallel rays that generate a projection
profile
 Translate-Rotate
◦ Single collimated beam with one or two detectors
translate across the patient collecting readings
◦ After translation tube and detectors rotate 1 degree
and begin the process again
◦ Repeated for 180 degrees – AKA rectilinear pencil
beam scanning
◦ 4.5 – 5.5 minutes to produce scan
Generation 2
CT Generations
Second Generation System
 Still based on the original translate/rotate principle
 Introduced a detector array (approx. 30 detectors)
 Multiple pencil beams which resembled a small fan
◦ Ray now assumes divergence
◦ Resulted in different reconstruction computations

Rectilinear multiple pencil beam scanning
◦ After a translation, rotation is by larger increments over
180 degrees

Shorter scan times depending on number of
detectors
◦ 20 seconds to 3.5 minutes
Generation 3
CT Generations
Third Generation System
 Based on fan beam geometry rotating
continuously 360 degrees
 Curved detector array
◦ 30 – 40 degree arc

Continuously rotating fan beam scanning
◦ As tube and detectors rotate, projection
profiles are obtained for every fixed point

Much faster
◦ Allowed for single breath hold scans
Generation 4
CT Generations
Fourth Generation System
 Two types of beam geometries
◦ Rotating fan beam within stationary ring of
detectors
 Tube is within stationary circular array which line 360
degrees of gantry
◦ Rotating fan beam outside nutating detector ring
 Tube rotates outside detector ring which tilts so that
beam strikes detectors on the far side which allowed
detectors nearest the tube to be outside the array
 No Longer manufactured…
Generation 5
CT Generations
Fifth Generation System
 Acquired scan data in milliseconds
 Electron beam CT Scanner (EBCT) and Dynamic
Spatial Reconstructor (DSR…and obsolete…)
◦ No Moving Parts…beam of electrons that scans
stationary tungsten rings
◦ No X-Ray Tube – Electron Gun in which electrons are
emitted in a beam which is accelerated, focused and
deflected at precise angles
◦ When beam collides with ring, x-ray is produced,
collimated into a fan beam through the patient
◦ Extremely fast reconstructions

Cardiac Scanner….Siemens Evolution
Generation 6
CT Generations
Sixth Generation
 Dual Source CT Scanner
 2 x-ray tubes with 2 sets of detectors offset 90
degrees
◦ Cardiac CT
◦ Better temporal resolutions
◦ Reduced CTA artifact
Seventh Generation
 Flat Panel CT – Still in prototype stages
◦ Similar to digital radiography systems in that utilize TFT
array as a detector
 Excellent spatial resolution but lack contrast resolution
 Angiography??
CT Physics
Lecture 2: Review of Basic Computing and
introduction to Digital image processing
COMPUTERS AND
DIGITAL IMAGING
The Computer

By definition – high speed electronic machine
utilized which accepts information in data format
through an input device and processes this
information with arithmetic and logic operations
from a program stored in memory
◦ Results can be displayed, stored, recorded or
transmitted

Introduced to radiology in 1955 in order to
calculate radiation dose distribution in cancer
patients
◦ Imaging Applications – Digital Imaging
◦ Non-imaging Applications – PACS , RIS
The Computer

Analog Computers
◦ handle data composed of continuously varied
electrical currents
 Analog watch – displays time with hands

Digital Computers
◦ Handle data composed of definite quantities of
current
 Digital watch – displays numerical readout
◦ All medical imaging achieved now with digital
The Computer

Hardware
◦ Physical components

Software
◦ Set of instructions upon which the computer
operates
 Computer Languages




Fortran – Formula Translations / Engineering
Basic – All purpose contains symbols and codes
Cobol – Buisness oriented
Pascal – High Level Math
The Computer
Computer Architecture
 General structure of a computer and includes all elements of hardware and
software – chips, circuitry, and systems software
Terminology
 Serial / Sequential Processing
◦ Data and Instructions is processed in the order in which items are stored – one
instruction at a time

Distributed Processing
◦ Information processed by several computers on a network – highly structured, free
exchange

Multitasking
◦ more than one task at a time

Multiprocessing
◦ two or more separate processors working differing sets of instructions

Parallel processing
◦ task distributed over multiple available processors carrying out at the same time

Pipelining
◦ fetching and decoding instructions in which at any time several programs instructions are
in varying stages
Components

Central Processor Unit – heart of computer, directs
information.. Capable of performing multiple tasks
(parallel processing)
◦
Consists of the control unit – tells computer how to carry out
software instructions
◦
Arithmetic/Logic Unit (ALU) – performs arithmetic or logic
calculations.
These are connected to the BUS 
Bus – conductors which connects various
components (provides path for the flow of electrical
signals)

2 basic types of internal memory
◦ RAM – Random Access Memory
 Temporary storage.
◦ ROM – Read Only Memory
 Contain data and programs to make computer
work.
◦ Basic Operating Instructions
Components
Hard Disk Drive – is a rewritable,
 Array Processor –
nonremovable storage system that must be
capable of storing a lot of data and transferring
◦ Primary data processing
data fast.
component. Has its own CPU and
uses CPU to perform
simultaneous mathematical
 Operating System is the primary software of
operations in a parallel fashion at
the CT computer. It controls the usage of
high speeds
computer hardware resources, such as available
memory, CPU time, and disk space.
◦ Is responsible for receiving the
scan data from the host computer,
 Common OS – Windows, MS-DOS, OS/2, and
performing all of the major
UNIX.
processing of the CT image, and
returning the reconstructed
image to the storage memory of
the host computer.

Digital Fundamentals

Operates on a binary number system
◦ Base 2 = 0, 1
 Yes, No system representing when current is present
Individual binary digit = bit
 Bit, like an atom, the smallest unit of storage
 A bit stores just a 0 or 1
 "In the computer it's all 0's and 1's" ... bits

How much exactly can one byte hold?
Digital Fundamentals

Individual binary digit = bit
◦
◦
◦
◦
4 binary bits (0.5 byte) = nibble
8 binary bits (1 byte) = byte = one addressable location in memory
16 binary bits (2 bytes) = word
32 binary bits (4 bytes) = double word
◦
◦
◦
◦
1 thousand bytes = 1 KB
1 million bytes = 1 MB
1 billion bytes = 1 GB
1 trillion bytes = 1 TB
Analog Image…
Analog images –
◦ Images that we, as humans, look at.
◦ Example photographs and all of our
medical images recorded on film or
displayed on various display devices, like
computer monitors.
◦ What we see in an analog image is
various levels of brightness (or film
density) and colors. It is generally
continuous and not broken into many
small individual pieces.
Digital Image

Numerical representations of images as 1
and 0…

Requires computer
Digital Image…
Digital Image

Numerical representations of images as 1
and 0…

Requires computer

A digital image is a matrix of many small
elements, or pixels.

Each pixel is represented by a numerical
value. In general, the pixel value is related
to the brightness or color that we will see
when the digital image is converted into
an analog image for display and viewing.

Generally, at the time of viewing, the
actual relationship between a pixel
numerical value and it's displayed
brightness is determined by the
adjustments of the window control as
discussed in other modules.
Analog to Digital Conversion
Converting an analog signal into “a sequence
of numbers having finite precision”
3 part process
 Sampling = conversion of continuous signal
into discrete signal from sampling stream at
certain increments
 Quantization = conversion of discrete signal
into a value
 Coding = assignment of a bit sequence to
the discrete output
Digital Imaging Systems
Generic Digital Imaging System
Components:
1. Data Acquisition
◦
Image Processing
2.
◦
3.
Attenuation Data
Input digital image to output digital image
utilizing binary
Image Display, Storage and
Communication
Digital Image Processing

CT based on a reconstruction process
where a digital image is changed into a
visible physical image

CT acquires images in the spatial location
domain
◦ Location of each number in an image
identified by x-y coordinate
Digital Imaging Characteristics

Characteristics:
◦
◦
◦
◦
Matrix
Pixels
Voxels
Bit Depth
Matrix

2 dimensional array of numbers consisting of
columns and rows defining small square
regions of “picture elements”

Diagnostic images generally are rectangular
in shape

When imaging a patient, the operator usually
selects the matrix size or aka FOV…

Standard CT Matrix is 512 * 512
Pixels

Smallest picture element

Generally square in shape and
measured in the X-Y
dimension

Contains a discrete value
representing a brightness level

Calculated using:
Pixel size = FOV / Matrix

DETERMINES
RESOLUTION: Larger
the matrix size, smaller
the pixel, better the
spatial resolution
Voxels
 Represents
the
volume of tissue
being imaged
 Measured
in the
Z dimension
3
dimensional
volume of tissue
Bit Depth

Determines shades of gray that a pixel
can take on

Number of bits per pixel

Uses the base 2 system

CT utilizes a bit depth of 12
◦ -1024 to 3071
Bit Depth
Image Digitization
Three distinct steps:
Scanning: Division of the picture into small
regions (pixels) onto a grid (matrix)
2. Sampling: Measures brightness of each pixel in
the image by utilizing a photomultiplier tube to
detect transmitted light
3. Quantization: Brightness value is assigned a
numerical value depending on strength which
could be positive or negative called a gray level.
Total number of gray levels is called gray scale.
1.
Spatial Resolution

The ability to see the difference in small objects that
are next to each other

Determined by the pixel size in the monitor matrix

Parameters that can affect spatial resolution:
◦
◦
◦
◦
◦
◦
Filters in high frequency regions
SFOV
Matrix size
Detector width and spacing
Number of projections
Focal spot size
PACS

(P)icture (A)rchiving and (C)ommunication (S)ystems
◦ Computer System which is used to capture, store,
distribute and then display medical images

Components –
◦
◦
◦
◦
◦
◦

Network Switches
PACS Controller with database image server
Short and Long term archives
RIS/PACS broker
Web server
Various displays
Integrated with both RIS and HIS
PACS
Communication Protocol Standards

HL-7
◦ Health Level 7: standard application protocol for use in
most HIS and RIS

DICOM
◦ Digital Imaging and Communications in Medicine:
developed by the ACR and NEMA (National Electrical
Manufacturers Association) – Standard for handling, storing
printing and transmitting information in medicine
◦ File format definitions and network communications
protocols
◦ Enables integration of scanners, servers, workstations,
printers and network hardware from multiple
manufactures into a PACS system
References
Image courtesy of Sprawls.com
 Stewart Bushong “Radiologic Science for
Technologists”
 Bushberg et al., “The Essential Physics of
Medical Imaging”
 Wikipedia
