Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
~16~3~
--1-
METI~OD OF I;IPRESSI~IG A~ID P~E~DItiG OUT A
SURFACE CHARGE O~l A MULTILAYERED DETECTOR STRUCTURE
TECEnNICAL FIEIJD
The present invention relates generally
to apparatus and methods used to obtain image
information from modulation of a uniform flux.
More specifically, the present invention relates
to methods and apparatus used to produce a radio-
graph by producing a latent image in a photocon-
ductor sandwich structure and then optically
reading out the latent image.
BACKGROUND ART
à. Facsimile and Vidicons
Facsimile systems modulate an electrical
signal in response to light that is reflected from
a small portion of an image. Facsimile requires
production of a real image.
Vidicon tubes store a latent image as a
charge distribution pattern in a semiconductor
target. This latent image i5 scanned by an elec-
tron beam and the current variation induced by the
different charges o~ separate portions of the
target comprises a video signal. Vidico..s require
use of a high vacuum and precise focusing of an
electron bezm that~must be shielded from external
electric and magnetic fields.
b. Conventional RadioqraPhic Methods
Latent images stored in silver halide
film or sele~~ium xeroradioyraphic plates must be
chemically or powder cloud developed to produce
real images. Use of calcium tungstate crystals or
a high atomic weight gas for image intensification
,~
~6Z3:~2
-2-
degrades image quality and still requires expo-
sures of from 1 to 5 R per typical clinical mammo-
gram. Powder cloud development of latent xero-
radiographic images requires high differential
charge densities on the xeroradiographic plate to
attract and hold powder particles prior to fusing.
This high charge differential is generated by
x-rays, or ions, impinging on the surface of a
charged selenium plate. The higher the differ_
ential charge needed to produce an image, the more
X-ray exposure is required.
The reader is directed to the following
publications for detailed discussion of this prior
art.
XERORADIOGRAPHY, J. W. Boag, PhYs. Med.
Biol., 1973, Vol. 18, No. 1, pp. 3-37; Principals
of RadioqraPhic ExPosure and Processin~, Arthur
W. Fuchs, 1958, Chapter 14, "X-Ray Intensifying
Screens", pp. 158-164.
U.S. Patent No. 3,860,817; Reducing
Patient X-Ray Dose During Fluoroscopy With an
Image System.
U.S. Patent No. 3,828,191, Gas Handling
System for Electronradiography Imaging Chamber.
U.S. Patent No. 3,308,233, Xerographic
Facsimile Printer Having Light Beam Scanning and
Electrical Charging With Transparent Conductive
Belt.
Recent research by the present inventors
and others indicates that a detailed latent image
can be created in a xeroradiographic plate by very
low levels (under 10 mR) of X-ray exposure. At
this exposure level latent image is present, but
its associated electrostatic field is not of
sufficient intensity to produce a real image by
the powder cloud method of image development. For
~1:62.~Z
-3-
details of this research see, Grant Application,
"Radiomammography With Less than 150 mR Per Expo-
sure", available from the Department of Experimen-
tal Radiology, M.D. Anderson Hospital, Houston,
Texas 77025.
c. Semiconductor Art
The prior art of reading charge storage
patterns out of multilayer semiconductor sandwich
structures lies largely in the area of electro-
nics, especially exotic computer memories.
Charge patterns on certain MOS struc-
tures can be "read out" using a photon beam. See,
Imaqin~ and Storaqe With a Uniform MOS Structure,
Applied Physics Letters, Vol. 11, Number 11, pp.
359~361. This technology functions by modifying a
depletion layer and then charging the layer to
saturation. The few microns thick depletion layer
is the only active structure.
Electric fields can also be impressed
across two separable photoconductive insulating
films in pressure contact that are precharged to
the same polarity. See, Increasinq the Sensitiv-
itY of XeroqraPhic Photoconductors, IBM Technical
Disclosure Bulletin, Vol. 6, No. 10, 1964, page
60.
IBM has also developed an exotic charge
storage beam addressable memory comprising a
semiconductor sandwich wherein the semiconductor
is totally insulated from both electrodes in the
sandwich. See, IBM Technical Disclosure Bulletin,
Vol. 9, No. 5, 1966, pp. 555-556. This device
allows data readout by shifting charge population
to one side or the other of the insulated semicon-
ductor.
Finally, some theoretical work has been
done on the behavior of light sensitive capaci-
llfi2~3Z
tors, see, AnalYsis and Performance of a LiqhtSensitive Capacitor, Proceedings of the IEEE,
April 1965, p. 378.
d. Objects of the Present Invention
It is an object of the present invention
to provide a method and apparatus capable of
replacing conventional photographic and radiogra-
phic films.
Another purpose of the present invention
is to provide an X-ray sensing system capa~le of
quickly producing radiographic images while expo-
sing the sensed patient or object to lower radi-
ation dosage than has been practical using prior
art systems.
A further purpose of the present inven-
tion is to provide an X-ray sensing system whose
output is an analog or digital video signal that
may be selectively displayed on a television
monitor, recorded on film, or directly stored or
processed in a computer for image enhancement or
pattern recognition.
Yét a further purpose of the present
invention is to provide a novel method and appa-
ratus for converting a charge distribution on a
semi-conducting surface to a modulated electric
signal.
Another purpose of the present invention
is to provide an apparatus and method that com-
bines the edge enhancement effect of xeroradio-
graphy with a low patient dose level.
Yet still another purpose of the present
invention is to provide a novel low noise method
and apparatus for reading out a latent image
stored as a pattern of electrical charges by
selectively discharging said charges in the pres-
ence of a reverse biased electric field.
11623~Z
--5--
Yet another purpose of the present
in'vention is to provide an X-ray sensing system
~hat is capable of directly timing X-ray exposure.
A final purpose of the present invention
is to provide an apparatus capable of converting a
latent i~age to a modulated electric signal that
is simple and inexpensive to build and operate.
D I SCLOSURE OF I NVENT I ON
The present invention is an X-ray film
replacement comprising a sandwich detector struc-
ture; a method of converting the latent radiogra-
phic image stored by such a detector to an elec-
tric video signal; an apparatus capable of practi-
cing the method and a diagnostic X-ray system
using the apparatus.
The detector sandwich comprises a con-
ductive back plate overlain by a photoconductive
layer, for example, a layer of amorphous selenium
having a high effective dark resistivity. This
high effective dark resistivity may be accom-
plished by forming a blocking diode layer between
the photoconductor and the conductive back plate.
The photoconductor is, in turn, overlaid by an
insulator which is covered by a transparent con-
ductive film. The structure may be shielded to
suppress external electrical noise.
The method of operating the detector
structure requires that the detector be charged to
a high potential, preferably while the photocon-
ductive layer is flooded with light to decrease
its resistance. After the light is extinguished,
the conductive back plate and transparent film are
shorted together, causing surface charge redistri-
bution. The detector is then reverse biased by
reversing the electric potential placed across it.
Subsequent X-ray exposure forms a latent image by
llfi~;~3Z
-6-
selectively discharging part of the detector's
surface charge. This discharge current may be
integrated to control the exposure.
The detector need not be flooded with
light, but w_thout the light the resistance of the
photoconductor will be higher and the detector
will take longer to charge. The latent image can,
also, be formed by exposing th,- uncharged detector
structure to an X-ray flux while its photoconduc-
tor layer is biased by an electric field.
The detector is read out by being raster
scanned by a beam of photons. These photons
create electron-hole pairs in ~he photoconductor,
which allow a portion of the the photoconductor's
surface charge to discharge through an external
electric circuit. The signal amplitude in this
external circuit at any given time is a functlon
of the intensity of the latent image in the por-
tion of the photoconductor being irradiated with
photons at that time, i.e., it is a video signal.
The apparatus for practicing the method
comprises means for charging the photoconductor,
means for exposing it to X-rays to form a latent
image and means for outputing this latent image as
a video signal.
Since the present invention is essen-
tially a replacement for X-ray film, it can be
used with any conventional source of radiation. A
typical diagnostic X-ray system using the present
invention also includes means for storing, trans-
ferring and processing the video signal to produce
clinical images useable by a radiologist.
BRIEF DESCRIPTION OF DRAWINGS
FIGURE 1 is a greatly enlarged cross-
sectional view of a portion of one embodiment of
the multilayered photon detector apparatus taught
by the present invention; J
R "
1~;233~
--7--
FIGURE lA is a schematic electrical
diagram illustrating the electric circuit analog
of the structure shown in Figure l;
FIGURE 2 is a view of the structure
~hown in Fi~ure 1 being flooded with photons;
FIGURE 2A is a schematic showing the
electric circuit analog of Figure 2;
FIGURE 3 is a vie~ of the detector
~tructure after it is charged, flooded with pho-
ton~ and the surface charge has been redistri-
buted;
FIGURE 3A i3 ~he electrical schematic
analog of the detector a~ it is shown in Figure 3;
FIGURE 4 illustrates the detector struc-
ture of a preferred embodiment of ~he present
invention as it is used in an X-ray system;
FIG~E S is a greatly enlarged schematic
cross-sectional view of a portion of Figure 4
illustrating the effect of a modulated X-ray flux
on the detector;
FIGURE 6 is a schematic view of the
detector graphically illustrating the change in
detector surface charge caused by X-ray exposu~e;
FIGURE 7 is a schematic view of the
detector apparatus taught by the present invention
being read out by a thin, scanning photon beam, an
output wave forD of the video electric signal is
illustrated together with a schematic representa-
tion of the surface charqe of the detector;
FIGU~E 8 is a highly enlarged schematic
cross-sectional view of a small portion of the
detector structure shown in Figure 7 illustrating
the electron hole production mechanism in the
photoconductor layer used ~y the present inven-
tion;
li6233Z
-8-
FIGURE 9 shows a partially cut-away view
of an apparatus adapted to practice the preferred
embodiment of the present invention;
FIGURE 10 is a simplified diagram show-
ing the major parts of the apparatus shown in
Figure 9;
FIGURE 11 is a schematic block diagram
illustrating a diagnostic X-ray system constructed
according to a preferred embodiment of the present
invention;
FIGURE 12 shows a cross-sectional view
of a clinical apparatus for conducting mass chest
radiographic screening using the present inven-
tion;
FIGURE 13 is a highly enlarged schematic
cross-sectional view of an embodiment of the
present invention used in the apparatus shown in
Figure 12.
FIGURE 14 is a graph illustrating the
theoretical maximum charge obtainable from an
experimental detector structure per unit area of
the detector as a function of the voltage applied
across its photoconductive layer. Figure 14 also
shows a plot of specific points representing
experimental results from tests run on this ex-
perimental detector structure for a given mylar
thickness and selenium thickness.
FIGURE 15 is a graph illustrating the
voltage across the selenium layer of a theoretical
detector in kilovolts as a function of the value
of C2, in farads, for that detector.
FIGURE 16 is a graph illustrating the
charge, in coulombs collected by the experimental
detector structure per unit of exposure as a
function of total exposure. The values plotted in
Figure 16 were obtained using a Bakalite shutter;
l~`Z33~2
g
FIGURE 17 illustrates the same informa
tion as Figure 16, but the data was obtained using
a PVC shutter;
FIGURE 18 is a graph showing the effi-
~iency of the present invention, as expressed in
coulombs of charge obtained per roentgen of expo-
sure, plotted against supply voltage for various
levels o total exposure. This figure illustrates
the increased efficiency of the present invention
at low radiation levels;
FIGURE 19 is a graph that plots the
total charge collected by an experimental embodi-
ment of the present invention as a function of
applied voltage across its photoconductive layer.
Figure 19 illustrates increase in latitude of the
present invention's detector that is observed as
applied voltage across the photoconductor is
increased;
FIGURE 20 is a functional block diagram
of an experimental prototype system of the present
invention.
FIGURE 21 is a schematic diagram showing
a portion of the detector structure used as a
radiation exposure detector automatically timing
the exposure.
FIGURE 22 is an electrical schematic
showing the reversed biased mode of detector
operation.
FIGURE 23 is an electrical schematic of
the preamplifier used in experimental system 3.
FIGURE 24 is a schematic diagram of the
X-position interface module used in experimental
system 3.
FIGURE 25 is the characteristic curve of
the detector structure.
1162332
--10--
BEST MODE FOR CARRYING OUT TEIE INVENTION
In Figure 1, multilayered photon detec-
tor apparatus 10 has first electrode 12 conduc-
tively connected to a first conductive layer 14.
A photoconductive layer 16 is in physical and
electrical contact with plate 14 which may be made
of aluminum or any other conductor. An oxide
layer 11 may act as a blocking contact between
layer 16 and plate 14. Layers 14, 16, and 11 may
be provided by using a conventional xeroradiogra-
phic plate. Transparent insulating layer 18
overlays and may be integrally affixed to selenium
photoconductor layer 16. A transparent conductive
layer 20 may be integrally affixed to and overlies
insulator 18. These layers may be made ~y vapor
deposition or by adhesively bonding the individual
components together. Layer 20 is electrically
connected to lead 22.
DISCUSSION OF EXPERIMENTAL SYSTEM #l
The experimental plates used to test the
present invention as early as late-1975 were made
from con~entional xeroradiographic plates. These
plates are made for mamography by the Xerox Com-
pany. To make the present invention's detector, a
plate is first cut~to the correct size and then
covered with mylar. The conductive coating, which
is either Nesa glass or a few angstroms of gold,
is placed on the mylar.
In the preferred embodiment of the
present in~ention discussed in this section of the
specification, the aluminum backing plate 14 is
approximately 1/10 of an inch thick and ~he layer
of amorphous selenium 16 is approximately 150
microns thick. The physical properties of sele-
nium are listed in Table I. The transparent
insulator 18 is mylar. The transparent conductor
;~ frad~ ~vl a~C
3~3~
--11--
20 may be Nesa glass, a thin film of metal deposi-
ted directly on the transparent insulator 18, or a
plastic film with a conductive coating, i.e., gold
covered mylar. The entire structure 10 may be
made by depositing successive layers of selenium,
mylar and a thin film of metal onto an aluminum
plate. Assembly may be accomplished by vapor
deposition, sputtering, or any other technique
useful-to deposit even thickness films. This
technology is well developed in the art of semi-
conductor electronics and glass manufacturing.
The thickness of the selenium layer must
be selected to maximize quantum efficiency of the
detector. This optimum thic~ness will be a func-
tion of the photoconductor's characteristics, the
potential at which the detector is operated and
the energy of the X-ray photons or other particles
to which the detector is exposed that act to
discharge it.
Basically, the thicker the selenium, the
more it interacts with a given energy of exposing
radiation and the more electron-hole pairs a given
guantity of radiation produces. Conversely, as
the selenium layer is made thinner the electric
field acting on these electron-hole pairs becomes
stronger (the same potential over less distance).
Thus a very thin layer of selenium would not
interact with the exposing radiation enough to
produce electron-hole pairs while a very thick
layer of selenium would result in the production
of lots of electron-hole pairs, but they would all
recombine when the exposure stopped.
The optimum thickness of the photocon-
ductor layer of the detector will depend on the
characteristics of the photoconductor and the
energy of the X-ray photons it is designed to
~i~iX33~
-12-
sense. In the experimental embodiments of the
present invention applicants have found that from
one hundred microns to four hundred microns thick-
ness of amorphous selenium is optimum for the
X-ray photon energies commonly used in diagnostic
radiology.
There are several facts the inventors
discovered experimentally that should be men-
tioned.
When several experimental prototype
detector structures were built under laboratory
"clean room" conditions they worked poorly.
Regular aluminum backed selenium-xeroradiography
plates covered with gold covered mylar plastic and
assembled with optical cement under less than
ideal conditions work very well.
It appears that a very thin aluminum
oxide layer between the selenium and the aluminum
acts as a blocking contact, i.e., a diode, to
retain the positive charge on the surface of the
selenium. This bloc~ing layer has a blocking
potential that must be overcome for current to
flow through the contact.
Also, when the experimental detectors
built by the inventors were scanned fast, a higher
signal strength output was obtained than when they
were scanned slowly. Since the light intensity
was the same, the faster scan was predicted to
result in lower signal strength. Applicants
cannot yet explain this phenomena, but, by pulse
modulating their light source for between 2 nano-
seconds and 10 microseconds on and off, this
effect can be used to increase signal strength
from the present invention.
Finally, repeated use of the prototype
plate resulted in the development of "axtifacts",
", ~
~iZ33;i~
-13-
i.e., lines and trash on the image. Repeated use
also caused an overall loss in detector sensiti-
vity. It was accidentally discovered that heating
the prototype plate for a short time with a heat
gun used to shrink plastic tubing removed these
artifacts and otherwise generally restored the
detector's performance.
In Figure 1, Rl represents the resis-
tance of the photoconductive layer when it is
flooded with light. R2 represents the resistance
of the photoconductive layer when it is in the
dark. C2 represents the electrical capacity
manifested by the charge separation across the
transparent insulator 18. One charge polarity
resides on the transparent conductor 20 and the
other on the surface of selenium, which is in
immediate contact with insulator 18. Cl repre-
sents the electrical capacity of the detector
measured between the aluminum conductor 14 and the
selenium surface 16. The selenium is thus the
dielectric material of C1.
Figure lA shows an electrical schematic
illustrating an electronic analog of detector
sandwich structure 10 shown in Figure 1. Rl, R2,
Cl and ~2 are shown schematically connected in
Figure lA as electrical symbols. Switch 24, which
i~ shown open, is used to represent the photocon-
ductive nature of selenium layer 16 of Figure 1.
Selenium has a dark resistance of approximately
1016 ohm-centimeter. A portion of this resistance
is caused by a biocking contact formed at the
Se-Al interface. When photons strike the photo-
conductor its resistance is l~wered because the
photons create electron hole pairs that carry
current. This lower resistance is illustrated
schematically in the electrical analogs contained
~i~2332
-14-
in this specification by a closure of switch 24
leaving only residual resistance Rl, which repre-
sents the forward resistance of the blocking
contact in series with the resistance of the
selenium layer when it is flooded with photons.
For a current to flow when the photoconductor is
flooded with light the blocking potential of
blocking contact 11 must be overcome.
33~
--15--
TABLE I
PhYsical Properties of Selenium
(2,3)
Atomic Number 34
Density (g cm 3) 4.25
Dielectric Constant 6.6
Resistivity at 20C. (~ cm) lol3_1ol6
Thermal Conductivity at 20C.
(W -lK-l) 2 x 10 3
Optical Band Gap (2V) 2.3
Photo Response Edge (AE) 4600
K absorption Edge (KeV) 12.7
Mobility of Holes (cm2s lv 13 0.14
Mobility of Electrons (cm2s lv 1) 5 x 10 3
Energy for Production of Charge
Carriers (eV) 7*
_______________~_________
* W = 2.67 x Eg ~ 0.86 eV
in which Eg is the optical band gap
1~6Z332
-16-
Figure 2 illustrates detector 10 being
flooded with photons 26. Simultaneously a nega-
tive high voltage, i.e., 2,000 volts, is applied
to terminal 22 hence to transparent conductor 20.
Figure 2A, the electrical schematic
analog of Figure 2, shows switch 24 closed by
light 26. The -2,000 volt potential, Vs, is
impressed across electrodes 12 and 22.
Figure 3 shows the semiconductor detec-
tor sandwich 10 after light 26 has been exting-
uished and leads 12 and 22 shorted together. The
resulting positive surface charge on the amorphous
selenium layer is illustrated diagrammatically by
dotted line 28. This surface charge is uniformly
distributed over the entire surface.
Figure 3A is Figure 3's electrical
analog. The detector is in darkness, so switch 24
is illustrated as open. Diode 15 and battery 13
represent the blocking contact and potential,
respectively, of blocking junction 11. The elec-
tric charge originally present on capacitor C2 has
redistributed so that a portion of that charge is
present on capacitor Cl, the exact portion being
dependent upon the ratio of Cl to (Cl + C2).
Figure 4 shows the charged detector
illustrated in Figure 3 used as an image receptor
structure for an X-ray image. Uniform X-ray flux
30 is generated by a convenient source of radia-
tion, such as an X-ray tube, not shown. The
detector will work with any radiation source
capable of generating electron-hole pairs in the
photoconductor. This uniform photon flux encoun-
ters and interacts with object 32, which may be
any object of interest. Object 32 is placed
directly over detector 10. For simplicity, object
32 is represented as an oblate spheroid of uniform
density.
~ ,, .~
~1~233Z`
-17-
Figure 5 shows a greatly enlarged sche-
matic view of a portion of detector structure 10
of Figure 4. Modulated X-ray flux 34 from the
radiation that has passed through object 32,
creates electron-hole pairs 36 in selenium layer
16.
Figure 6 shows detector 10 of Figure 4
after the X-ray exposure has been completed. The
detector is now storing a latent image. The
modulated surface charge comprising the image is
schematically illustrated by a humped dotted line
38. This dotted line represents the change in
potential created by the electron-hole pairs
generated by the X-ray exposure.
Figure 7 shows detector 10 having a
latent image stored as a modulated surface charge
38. A thin beam of light 40 is shown scanning the
surface of photoconductive layer 16 in a regular
raster pattern. In the experimental system number
1 embodiment of the present invention, this thin
scanning light beam is produced by a ~e-Cd laser.
It should be understood that the photon
beam need not be coherent. It may be of any
frequency capable of creatinq electron-hole pairs
in photoconductive layer 16 of detector 10.
Electrode 41 is connected to electrical
ground and electrode 43 carries a video electric
signal whose wave form is a function of modulated
surface charge 38 in detector 10 scanned by light
beam 40.
Arrow 42 indicates the direction of
movement of scanning light beam 40. Output wave
form 44 indicates the voltage variation of the
output video signal obtained by said beam's scan-
ning the latent image.
~fi23~Z
-18-
Figure 8 is a schematic, highly enlarqed
cross-sectional view of a portion of the detector
structure being scanned by light beam 40 of Figure
7. Light beam 40 penetrates the transparent
conductor and transparent insulator to generate
electron-hole pairs 36 in that portion of selenium
layer 16 irradiated by the beam.
Functionally, detector 10 operates by
placing a uniform surface charge on selenium layer
16 and then selectively discharging part of this
surface charge by X-ray exposure to form a latent
image.
As shown in Figure 2, when light photons
flood detector 10, selenium photoconductor layer
16 becomes conductive. If, as shown, a -~,000
volt potential is applied between the aluminum
backplate (at ground) and the transparent conduc-
tor 20 ~at -2,000 volts potential), then the
detector will charge as a capacitor. Figure 2A
illustrates how this charge is applied to the
system. Light flood 26 increases conductivity in
the selenium layer 16. The applied voltage Vs is
-2,000 volts. This causes a charge to build up
across capacitor C2, which represents the capacity
between the surface of the selenium in contact
with the transparent insulator and the transparent
conductor of detector 19. Capacitor C2 thus
charges to potential Vs of 2,000 volts.
After capacitor C2 has charged to poten-
tial Vs~ the light flood is removed and detector
10 is stored in darkness. In the absence of
photons, amorphous selenium layer 16 becomes
nonconductive and its dark resistivity in syner-
gistic combination with blocking contact 11 pre-
vents the surface charge on the selenium from
leaking off. This condition is illustrated by
~i~Zh~32
--19--
Figures 3 and 3A. After detector 10 is in dark-
ness, terminals 22 and 12 are shorted together.
This causes the electric charge on capacitor C2 to
redistribute between capacitor C2 and Cl, each of
which is then charged to a potential of one-half
Vs, i.e., 1,000 volts if Cl=C2. Capacitor Cl
represents capacitance between the surface of the
photoconductive layer 16 and the aluminum backing
14. This 1,000 volt surface charge is distributed
uniformly over the selenium layer's surface. This
surface charge is maintained across the selenium
layer by the high dark resistivity of the selenium
in conjunction with the r.ow reverse biased block-
ing junction at the Al Se interface. The effec-
tive blocking potential of this junction has been
experimentally determined to be about 150 to 300
volts in the experimental systems discussed below.
Resistance Rl, representing resistance
of the selenium layer when it is exposed to light,
is relatively low compared to its dark resistance
R2.
Each photon, striking the photoconductor
depending on its energy, generates a specific
number of electron-hole pairs within photacon-
ductive layer 16 (see Table II). Each electron-
hole pair discharges a portion of the surface
charge at the spot it is generated.
Figure 4 illustrates how this effect is
used to impress a latent image on the detector.
Figure 4 is a cross-section schematic
view of the photon detector structure taught by
the present invention being used as the image
receptor in an X-ray system.
A uniform flux of X-ray photons 30 is
modulated, i.e., partially absorbed, by its pass-
age through an object being X-rayed 32. The
modulated X-ray flux then strikes detector 10.
332
-20-
Detector 10 has a 1,000 volt surface
charge, as was illustrated by charge 28 in Figure
3, impressed on selenium layer 16. As the X-ray
photons strike the selenium layer, they generate
electron hole pairs. For monochromatic X-rays the
number of electron hole pairs generated is a
direct function of the number of X-ray photons
striking the selenium detector layer.
Some of the X-ray pho~ons in uniform
flux 30 are absor~ed by object 32. Thus the
modulated flux striking photoconductive layer 16
contains information about the internal structure
of the object being x-rayed. This information is
contained in the number of photons striking each
portion of the detector.
Figure 5 illustrates how this modulated
X-ray flux creates a modulated surface charge on
the surface of the detector. Modulated X-ray flux
34 generates electron-hole pairs 36 in selenium
layer 16 of detector 10. The number of electron
hole pairs generated at each po1nt on the surface
of the detector is a function of the number of
X-ray photons impinging on the photoconductive
layer. Each electron hole pair 36 discharges a
portion of the surface charge impressed on the
photoconductive layer at the point where it is
generated. The greatest number of electron hole
pairs will be generated where the X-ray photons
from uniform flux 30 strike the detector without
any absorption from the object being x-rayed. A
lesser number of X-ray photons will strike the
photoconductive layer under the object being
X-rayed. The radio-opacity of the object being
X-rayed is thus reproduced in the surface charge
o the detector structure after X-ray exposure.
~1~233~
-21-
Figure 6 schematically illustrates the
detector structure's modulated surface charge
after the X-ray exposure described in connection
with Figures 4 and 5.
The surface charge is lower where no
object absorbed X-ray photons. The object being
X-rayed, which is assumed to be uniformly opaque
to X-rays, absorbs a portion of the uniform X-ray
flux. This results in the generation of fewer
electron hole pairs under the object and a higher
surface charge on the detector under the object
being X-rayed. This is represented by modulated
surface charge dotted line 38.
Figure 7 illustrates schematically the
readout method used to extract an electric video
signal from the modulated surface charge which
forms the latent image on a detector structure
taught by the present invention.
The detector structure is kept in dark-
ness and a thin beam of light, preferably gene-
rated by a helium-cadmium laser, is scanned in a
regular raster pattern across the surface of
photoconductive layer 16.
As the scanning light beam 40 moves in
direction illustrated by arrow 42, it produces a
small moving spot on the surface of photoconduc-
tive layer 16. The size of this spot determines
the resolution of the final image. It is there-
fore desirable that this spot size be kept small.
As laser beam 40 scans the surface of
photoconductive layer 16, it generates electron-
hole pairs as is illustrated in detail by Fiqure
8. These electron-hole pairs are mobile within
the photoconductive layer and discharge a portion
o the surface charge. In the preferred embodi-
ment of the present invention the electrons move
-22-
toward the positively charged selenium surface of
the detector sandwich structure 10. The voltage
drop produced across the resistor connected be-
tween ground connection 41, which is attached to
transparent conductor 20 and video cutput elec-
trode 43, which is attached to alumlnum conductive
backplate 14, will be a function of the intensity
of the surface charge at the point where the laser
beam generates the electron-hole pair.
Scanning laser beam 40 thus produces a
modulated electric signal 44 at output electrode
43. The voltage of this output electric signal
will be a function of the surface charge present
at the spot where the laser beam stri~es the
photoconductive layer of the detector. The cur-
rent present in the output circuit is a function
of the frequency and intensity of the scanning
light beam and is a further function of the speed
with which the light beam scans the surface of the
photoconductor. Video signal 44 can be electro-
nically processed to produce an image that repro-
duces the latent image found in the surface charge
o the photoconductive layer of the detector.
Depending on the intensity of the sur-
face charge, the beam scanning speed, and the
frequency and intensity of the light beam used,
the surface charge may be repetitively scanned
several times by the beam of light.
Table II lists the values for the thick-
ness of the selenium layer, the X-ray photon
energy, the fraction of this energy absorbed by
the selenium layer, quantum conversion between
X-ray photons and electron-hole pairs, dark resis-
tance of the selenium and the total receptor area
of detector 10 used in calculating the operating
parameters of a sample detector constructed ac-
~;lfi2332
-23-
cording to the preferred embodiment of the present
invention.
TABLE II
Assumed Values
Thickness of Selenium Layer 100 microns
Energy Radiation 21 keV
Absorbed Fraction of photons ~5%
Quantum Conversion 3000 electron/
photon
Surface Voltage 1000 V
Photon Flux/Roentgen 5 + 109 cm 2
Dark Resistivity 1ol6 n cm
Total Receptor Area 560 cm2
Table III lists the calculated capacity
of the detector sandwich structure, the charge
density present in this structure when the surface
potential is 1,000 volts, the number of elementary
charges present per square centimeter in the
structure and the dark resistance and dark current
of the detector at a thousand volts charge.
32
-24-
TABLE III
Calculated Values
Capacitance 5 8 1o~llF/ 2
Charge Density at 1000 Volts 5.8 x 10 8Coul/cm2
11 2
Number of Elementary Charge 3.6 x 10 el/cm
Dark Resistance 1014 Q /cm2
Dark Current at 1000 Volts 10 llA/Cm
Using the values from these tables it is
possible to calculate operating parameters for a
typical detector system constructed according to
the preferred embodiment of the present invention.
If detector 10 has an area of 560 syuare
centimeters and is exposed to uniform flux of 21
keV X-rays resulting in an exposure of Np X-ray
photons per unit area, then the total number of
electron hole pairs generated will be 3,000 times
Np. Information theory requires that three times
as many electron hole pairs be generated as are
present at the statistical noise level of the
system for the latent image to be detectable
There are approximately 5 times 10
photons per square centimeter per roentgen of
X-ray exposure at 21 Kev. Thus the minimum de-
tectable latent radiation exposure that will
produce a detectable latent image in the selenium
sandwich structure taught by the preferred embodi-
ment of the present invention (for a 560 square
centimeter detector) is approximately 23 micro
roentgens per exposure. This may be thought of as
-25~
the theoretical lower limit of exposure to gene-
rate a 10 line pair per millimeter image with the
present invention.
Turning now to a practical ex~mple, if
the X-ray exposure of a patient scan is 100 mR and
at the thinnest section of the X-rayed object 50%
of this radiation is absorbed by the sample, then
exposure at the detector at the brightest portion
of the image amounts to 50 mR while the lowest
theoretical detectable dose, as was discussed
above, amounts to 23~R. This theoretically yields
a latent image represented by modulation of the
surface charge present on the detector at a ratio
between the brightest and the darkest areas of the
image of approximately 2,000.
The example discussed above shows that
the detector structure taught by the present
in~ention is theoretically capable of recording an
image using a selenium photoconductive layer with
a resolution of 10 line pairs per millimeter and a
brightness range of over 2,000 after a total X-ray
exposure of only 100 mR. An example of how this
latent image can be read out of the detector
follows.
Figure 9 shows a partially cut-away view
of an apparatus capable of reading out the image
whose generation was discussed above.
Light-tight housing 46 contains a light
source 48 aligned so as to project a light beam 50
onto the front of a scanning means 52 which may be
a rotating multi-sided mirror. Mirror 52 is
mounted on an axis 54 which is operably attached
to a scanning motor 56. The mirror axis and scan-
ning motor are affixed to a platform 58. This
platform is mounted on an axis 60 which is ortho-
gonal to axis 54. Axis 60 is mounted at one end
li~2332
-26-
to first upright support 62 and its other end to
second upright support 64. The end of axis 60 is
swingably mounted in second upright 64 and is
operably attached to stepping motor 66.
The entire scanning means comprising the
mirror and associated positioning components is
available from Texas Medical Instruments, Inc.,
1210~ Radium, San Antonio, Texas 78~16.
Mirror 52 and its associated scanning
mechanism is located at one-end of housing 46. A
multilayered photon detector apparatus 10 is
removably inserted through a light-tight opening
68 into a holder at the other end of housing 46
such that deposited transparent conductive layer
20 faces toward mirror 52.
Light source 48 is preferably a Helium-
Cadmium laser such as a Liconix model #402 laser
with an optical modulator. This laser produces a
beam of intense light at approximately 4,~00
angstroms.
Scanning motor 56 rotates multi-sided
mirror 52 on axis 54 to cause light beam 50 to
scan horizontally across the surface of detector
10. This causes spot 70 to intersect the selenium
photoconductive layer as was discussed in connec-
tion with Figures 7 and 8 above. Each time spot
70 moves from the left to the right side of detec-
tor 10, stepping motor 66 moves platform 58 through
sufficient arc to deflect spot 70 vertically
1/20th of a millimeter. This stepping-scanning
function may be controlled mechanically or electri
cally. By convention, spot 70 begins scanning the
surface of detector 10 at its upper left hand
corner. After raster scanning the entire surface
of the detector, the scanning mechanism may be
116Z332
--27--
programmed either to turn off or to rescan the
plate.
Figure 10 is a simplified schematic view
of the scanning readout mechanism shown in Figure
9. Light beam S0 from laser 48 reflects off of
the polygonal surface of multi-sided mirror 52 as
that mirror rotates on axis 54. This causes light
beam 50 to generate flying spot 70 which moves
across the surface of detector structure 10. As
was discussed in connection with Figures 7 and 8,
above, this causes a video signal to be generated
across the detector structure.
Theoretically, photons from light source
48 have a wave length of approximately 4,400
angstroms. The quantum yield of these photons and
selenium approaches unity when field strength
nears 105 volts per centimeter. In a hundred and
f-fty micron selenium plate, such as is used in
the preferred embodiment of the present invention,
this corresponds to a thousand volts surface
potential.
It is possible to pulse the laser to
decrease the scanning time required to obtain an
image from the detector structure. In the present
example, the laser beam may be pulsed for a 100
nanoseconds period with 3 to 4 microsecond spacing
between pulses. This illuminates each image
element sufficiently to read out a latent image
stored in the detector. The practical advantage
of such a regime is to permit much shorter samp-
ling times. It may also result in higher signal
strengths.
Figure 11 is a block diagram of a radio-
graphic X-ray system taught by the preferred
embodiment of the present invention. Laser 72
produces a small intense beam of light 74 which is
2332
-28-
expanded by optical objective 76 and lenses 78 and
80 into a reasonably wide parallel beam 82.
Focusing mirror 84 reflects beam 82 on the reflec-
tive surface 86 of scanning mechanism 88. Scan-
ning mechanism 88 is substantially the same as the
scanning mechanism discussed in connection with
Figure 9 above. This scanning mechanism may be
any apparatus capable of moving the laser beam
over the detector and film. For example it may be
a set of computer controlled mirrors or hologra-
phic optics.
Beam 82 projects a small flying spot 90
onto the surface of detector structure 10. Detec-
tor structure 10 was discussed in detail in con-
nection with Figures 1 through 8, above.
Detector structure 10 is electrically
connected by lines 92 and 94 to cooled amplifier
96. Amp~ifier 96 may be a low noise amplifier.
Amplifier 96 is connected by line 98 to lines 100
and 102 which electrically connect the output of
cooled amplifier 96 to signal processor 104 and
second signal processor 106, respectively. Signal
processor 104 is connected by line 108 to beam
modulator 110. Laser 112 produces an intense beam
of coherent light 114 which is modulated by beam
modulator 110 and spread by objective 116 in
lenses 118 and 120 to coherent modulated beam 122.
Modulated beam 122 is focused by focusing mirror
12~ onto a second mirror surface 126 of optical
scanner 88. This surface may be another surface
of the same scanner or may be a separate optical
scanning system.
Mechanical movement of scanning system
88 causes beam 122 to form flying spot 128 which
scans the surface of recording medium 13~. This
process allows the very low intensity latent image
Z332
-29-
on detector structure 10 to be electrically ampli-
fied by cooled amplifier 96 and signal processor
104 and then to be rewritten as an intensified
image on a photographic or xeroradiographic plate
130.
The electrical output of cooled ampli-
fier 96 also is fed to signal processor 10~ by
lines 98 and 102. The output of signal processor
106 is connected by line 132 to a digital computer
134. Computer 134 is used to digitally store and
manipuIate the information content imparted to it
by the electrical signal produced by cooled ampli-
fier 96 by way of signal processor 106. The
images may be stored on magnetic tape or disc
files and can be manipulated within the computer
by algorithms for image edge enhancement or pat-
tern recognition for automated diagnosis.
The output of signal processor 106 also
goes by line 136 to mass storage system 138.
Mass storage system 138 contains a high
resolution display tube 140 which produces an
analog image 142 which is focused by focusing
optics 144 onto film plane 146 of a mass film
storage system 148. This mass film storage system
may be a 35 or 70 mm cassette system.
The type of radiographic image proces-
sing and storage system described in connection
with Figure 11 is especially useful for inter-
facing mass radiographic data acquisition equip-
ment with the central computing facilities of a
large hospital complex. Digital storage of the
radiographic images allow the data to be accessed
by a remote radiologist with speed and precision.
Computer-based pattern recognition algorithms
allow inexpensive gross screening of large patient
populations for radiographic anomalies. The
Z332
-30-
system's ability to rewrite information obtained
at very low radiation dosages onto conventional
xeroradiographic plates or films permits the
system to protect the patient while interfacing
with presently existing radiographic data storage
formats.
As will be discussed further below, the
radiographic data comprising the latent image on
detector structure 10 may be read out in real time
to produce a video image which may be studied by
the radiologist.
Figure 12 shows a mass screening system
constructed according to the preferred embodiment
of the present invention. The embodiment shown in
Figure 12 could be used for real time read out of
chest films in a mass screening program. X-ray
source 150 produces a uniform flux of X~rays 152.
Uniform flux 152 passes through and is modulated
by patient 154. The modulated X-ray flux impinges
onto multi-layered photon detector apparatus 156.
Figure 13 shows a highly enlarged cross-
sectional view of a portion of detector sandwich
structure 156 used in Figure 12.
Detector structure 156 is like detector
structure 10 described in Figures 1 through 8,
above, except that it is positioned so the modu-
lated X-ray flux impinges on the selenium photo-
conductive layer after it passes through the
aluminum backplate. Additionally, a thin insula-
ting layer 158 has been added to the outer surface
of the aluminum support structure to electrically
insulate the aluminum layer from patient 154.
Detector sandwich structure 156 is
mounted within light-tight housing 160. Housing
160 also contains a means for scanning the detec-
tor structure with a thin beam of light, i.e.,
;233~
-31-
laser scanning system 162 which projects light
beam 164 in a raster pattern onto the back side,
i.e., the conductive transparent coating side, of
detector 156. Base 166 of housing 160 contains
control electronics 168 and signal electronics
170. The entire housing and base may be mounted
on legs 172 to raise detector structure 156 to
chest height of patient 154.
Control means 168 comprises electronics
necessary to control the scanning pattern of laser
scanner 162 so as to cause light beam 164 to scan
the photoconductive side of detector 156 in a
regular raster pattern. The control electronics
also controls X-ray source 150 and synchronizes
exposure and readout operations.
Signal acquisition means 170 comprises
electric amplifiers connected to the video output
of detector 156. Signal electronics 170 amplifies
the detector output and provides a video output
capable of operating video display means 174,
which may be a high resolution video display
monitor.
Operationally, the mass screening system
shown in Figure 12 is a small portable unit.
Patient 154 steps up to the unit and presses his
chest against thin insulating layer 158 over the
aluminum back plate of detector structure 156.
X-ray source 150 is then turned on by the control
means and irradiates the patient with under 100 mR
of X-ray radiation. As is shown in Figure 13,
this modulated X-ray flux penetrates thin insula-
ting layer 158 and the aluminum backplate of
detector 156. The X-ray flux then generates
electron hole pairs in the selenium detector
structure as was discussed in connection with
Figures 2 through 6, above. The effect of X-rays
33~
-32-
being absorbed in the selenium with consequent
neutralization of surface changes is a displace-
ment current that flows as C2 redistributes its
charge over Cl. When this current is integrated
over time it will reach a preset value. Control
circuitry will then terminate the X-ray exposure.
Use of such circuitry insures images of equivalent
quality regardless of the thickness of the patient.
The above device or operation is called "automatic
photo-timing" or "automatic exposure control".
Referring again to Figure 12, as soon as
the exposure is completed, control electronics 168
causes laser scanner 162 to scan light beam 164
across the surface of the selenium photoconductive
layer in a regular raster pattern as was described
in connection with Figures 7 through 10, above.
This causes detector 156 to produce a video signal
containing image information. This video signal
is fed ~o signal electronics 170 where it is
amplified to provide a ~ideo output to TV monitor
174.
Such a system may be used to make a
clinical radiograph of any body part, including
breast, chest, head, etc.
The detector itself, being a replacement
for film in a radiographic system, can be used
with intensifying screens. One embodiment of the
detector structure would use a phosphor, prefer-
ably a high Z rare earth phosphor, in place of
insulating layer 18 in Figure 1. The X-rays
striking the phosphor would generate light which
would create ion-hole pairs in pho~oconductive
layer 16.
1i~2332
-33-
DERIVATION OF THE CHARACTERISTIC CURVE OF
THE DETECTOR
This section refers to the derivation of
the curve shown in Figure 25.
W~len the experimental system is exposed
to light, a total charge f QL coulombs will flow
in the external circuit where
QL = 2VsC22 coul/pixel (1
Cl+C2
Exposure to X-radiation prior to such a light
exposure would simply reduce QL proportional to
the X-ray exposure dose.
One may compute the charges placed in
motion due to absorption of X-ray photons by first
defining the photogeneration efficiency after
Fender
~ = d . E (ieVon pair ~ ) (2)
where ~ is the energy deposited within the sele-
nium in ev due to absorbed photons, N is the
number of ions which neutralize the charge on the
surface of the absorber, and E is the average
eléctric field across the absorber. Also we note,
n = QxA and dn = A dQ (3)
where Qx is the surface charge neutralized in
coulombs per square centimeter, A the area of the
absorber in square centimeters and e the electron-
ic charge in coulombs/ion-pair. The energy absorbed
may be calculated as
"
11~233~
-34-
~ = fkXA and d~ =fkAdX (4)
where f is the fractlon of X-ray photons absorbed
in the selenium, k is the energy fluence in air
per roentgen exposure and X is the exposure in
roentgens. Also, since the electric field is that
which exists within the selenium,
E = Q 1 (5)
Cldl
where Q*l is the instantaneous charge existing
across the selenium, Cl the capacitance of the
~elenium layer, and dl the thickness of the sele-
nium. Substituting (3)(4) and (5) into (2) we
have
= fkeQ*l , dX
C1dl dQx (6)
From (1) we may express Q*l as
Q*l = 2VsClC2 _ QXC
Cl ~C2 Cl+C~
substituting in (6), we have the differential equation,
dQx + fke Qx ~ 2kfeC2Vs = O (7)
dX d~(Cl+C2 ~(Cl+c2) dl
the solution of which is
Qx = 2C2Vs (1 - exp ~ fkeX ~ ) (8)
~ ~1 dl ( C l+C2 ) J
33Z
-35-
if QR QX (C2 ) (9)
cl+c2
then the total charges placed in motion whenilluminating a pixel which has absorbed X roent-
gens of X-ray is,
QSig = QL OR ( 10 )
from (1) and (10),
Qsig = 2VsC2 - 2VsC2 (1 - exp ~ f~e ~) (11)
Qsig - ~2V c2 exp( - ~X)] (12)
Cl+C2
where ~ = fke
rldl ( Cl+C2 )
Equation (12) is of the form described by Boag in
which he states that the radiographic discharge of
electrostatic plates is exponential.
Figure 19 is a plot of QR versus Vs for
various doses of X-radiation. From such data we
may solve equation (8) for ~, the photogeneration
efficiency. From Figure 19 we also note the
emperical data predicts a loss of charge which may
be due to recombination; however, it may also be
indicative of deep hole traps within the selenium
that establish a residual potential below which
the system cannot be discharged.
-36-
DESCRIPTION AND DISCUSSION OF METHODS
OF CHARGING THE DETECTOR
Charqing Method #l
To review briefly, the semiconductor
sandwichlike structure, illustrated by Figures 1
and lA above, appears as a pair of capacitors in
series. C2 is the capacitor formed in the experi-
mental system by Nesa conductive glass and the
surface of the selenium proximate the Nesa. A
mylar insulating layer acts as a dielectric. Cl
is the capacitor formed by the surface of the
selenium and the aluminum substrate, the selenium
photoconductor acting as the dielectric. Because
photoconductors are not dielectrics when they are
illuminated, Cl will exist only when it is in the
dark. This ignores such things as deep hole
traps, dark currents, etc., but they will be
discussed in turn.
Referring to Figure 1, when a voltage is
applied between terminals 22 and 12, a current
will flow charging capacitors Cl and C2 to volt-
ages V1 and V2 respectively, where:
V5 = Vl + V2 (1)
In which case:
V2 = V5 Cl
Cl + C2 (2)
Where Vs is the supply voltage.
and V1 = V
(3)
If the selenium layer is illuminated
while there is a potential between terminals 22
11~2332
-37-
and 12, capacitor Cl becomes a conductor and
allows additional current to flow to C2. This
additional charge Ql is the charge initially
residing on Cl. That is, since
Ql = ClVl
(4)
substituting (3) into (4),
ClC2
Ql = Vs Cl+C2 (5)
The initial voltage across C2 is deter-
mined as in (2). The increase in voltage across
C2 due to Cl becoming conductive is:
Ql VsC2
V2 = Cl Cl+C2 (6)
thus the total voltage across C2 is now (2) plus (6) or
Cl vSc2
V2 = Vs Cl+C2 Cl~C2 = Vs (7)
Because of (l), V1 is now found to be
equal to zero. The charge across C2 is now the
total Q or
QT = VSC2
~8)
If the illumination is removed, this
voltage distribution will persist until the poten-
tial is removed and terminals 22 and 12 are con-
nected together. C2 will now partially discharge
into Cl resulting in an equal voltage appearing
....
~1~2332
-38-
across Cl and C2 since they are now essentially in
parallel. This voltage will be
V parallel = ~total
C total
-- V C = V = -V
s 2 1 2
Cl+C2
Note that Vl and V2 are equal, however,
of opposite polarity.
It will also be noted that no supply
voltage Vs is in the circuit during this redistri-
bution of charge. The above operation has resul-
ted in placing a charge on the surface of the
selenium much as is accomplished in the Xeroradio-
graphic procedure. In the present invention,
however, the charge has been applied to a selenium
surface that is physically inaccessible.
The system is now ready for exposure to
X-rays. An integrating ammeter or a coulomb meter
is placed in the circuit between Cl and R2 such
that the current, or net charge movement due to
radiation exposure may be measured. This flow of
current is proportional to exposure and thus can
be used to control the exposure. This is advan-
tageous because it allows the exposure time to be
controlled by the amount of radiation actually
reaching the detector.
Alternatively, the surface of capacitor
C2, which is a thin conductive film, may be etched
to form a small island as shown in Figure 21. In
Figure 21 coulomb meter 2101 may be connected
between island electrode 2103 and the aluminum
substrate 2105. During X-radiation exposure meter
2101 will accumulate a charge which may be used to
1~2.'33;~
-3g-
photo-time the exposure. This is advantageous
because regardless of the thickness of the patient,
the X-ray machine will stay turned on until the
correct amount of charge is collected by the
detector, thus insuring a properly exposed image.
Several of these isolated islands may be etched in
the plate and used as exposure meters.
When the detector is exposed to radia-
tion, electron-hole pairs are created within the
selenium which neutralize the charge across Cl.
As these charges are neutralized, the charge on C2
redistributes. This redistribution maintains the
two capacitors at the same potential. Thus the
total charge flowing through the external circuit,
i.e., through load resistor R2, is the initial
charge placed on C2 alone. From (9) we can obtain
the voltage across C2, i.e., the total charge on
C2 pxior to X-ray exposure:
Q2 = C2Vs C2
Cl+C2 ( 10 )
Equation (10) is the total charge that
will move if the system is not exposed to radia-
tisn. Should X-ray exposure occur to create a
latent image, a portion, Qx' of the charges on C
will be neutralized. The charge remaining on C2
will be:
Qsig ( QT QX) C2
Cl+C2 ( 11 )
Where QT = c2v5 as in (8). If QX = , then Equa-
tion (11) becomes (10).
i~i2`332
-40-
This charge can be accumulated by a
coulomb meter and would be a function of the
number of electron-hole pairs generated by the
X-rays absorbed in the selenium. When the detec-
tor is scanned by the laser beam, each pixel, i.e.
the area illuminated by the laser beam at rest, is
sequentially discharged. Functionally, Cl is made
conductive where the light shines. When Cl be-
comes conductive, the charge on C2 can flow through
the external circuit. The current which flows
through the external circuit is a function of the
latent image stored in the photoconductor.
As the laser beam scans across Cl, the
IR drop appearing across resistor R2 is a video
signal.
Charqinq Method #2
Consider the same sandwich-like detector
structure in a circuit configuration analogous to
the one shown in Figure 22.
Structurally, three-position switch 2201
is connected to one side of the plate structure
2203 illustrated as capacitor Cl and C2 in series.
One side of coulomb meter 2205 is connected to the
other side of plate structure 2203. The other
side of coulomb meter 2205 is connected to one
side of resistor 2207. The ot~er side of resistor
2207 is connected to the positive side of battery
2209; directly to terminal 2211 and to the nega-
tive side of battery 2213. The positive side of
battery 2213 is connected to terminal 1 of three-
position switch 2201. Terminal 2211 is connected
to terminal 2 of switch 2201. The negative side
of battery 2209 is connected to terminal 3 of
three-position switch 2201.
Three position switch 2201 is first
placed in the position 1. The voltages across C
32
-41-
and C2 are the same as shown in Figure 1 and lA.
If the selenium is now illuminated, Cl becomes
conductive as before and V2 = Vs as in (7). If
the illumination is now removed this voltage dis-
tribution again persists. Moving switch 2201 to
position 2 results in the charge on C2 being
distributed over both Cl and C2 as before with V
= -V2 as given in (9).
If switch 2201 is moved to position 3, a
voltage will exist across C2 which will be that
ratio of Vs due to capacitance Cl and c2 less the
voltage due to trapped charges (9), i.e.,
V =V C - V C
2 s 1 s 2
C1+C2 C1+C2 (12)
or
~Cl C2 ~
V ~ V ~Cl+C2 J
Cl+C2 2C2
= Vs Cl+C2
2C2 ~
= Vs ~ Cl+C2 J
2VsC2 (13)
V2= Vs Cl+C2
~.~332
-42-
since Vl + V2 = Vs
we find Vl 2Vs C2 (14)
Cl+C2
that is, the voltage across the selenium is now
twice that obtained in the other charging method
#1.
Now the total charge on C2 is from (13)
2VsC2 \
Q2 = C2 ~s ~ Cl+C2 J
2VsC2C2 (15)
= C2Vs ~ Cl+C2
Under these conditions if the ratio of
C2~C1+C2)is near unity, then V2 will be opposite
in polarity to Vl as well as at a lesser poten-
tial.
If we assume C2 = 10 Cl, the ratio
C2~ 1+C2)= 10/11, also assuming Vs = lOOV, from
~14)
Vl = 2 x lOOV x 10/11
= 181.8 volts
and V2 = -81.8 volts
Thus the Q2 as shown in Equation (15)
represents a charge opposite to that on Cl and
that of the voltage supply Vs.
Upon exposure to radiation and/or subse-
quent read-out by the laser, the total Qsig moving
through the external circuit will be that residing
i23~
-43-
on C2 initially, plus that required to charge C2
to a voltage of Vs in the opposite polarity. Thus
Qsig Q2 QT
2VSC2C2 + C2VS
= -C2VS + Cl+C2
orQsig = 2Vs C2 C2
Cl+C2 (16)
If the detector has been exposed to X-rays then,
Qsig ~QT Q~ C2 (17)
Cl+C2
here, as before, QX are those charges neutralized
by the electron-hole created in the selenium by
absorbed X-ray photons.
In summary, use of charge method #2
results in doubling the voltage across the sele-
nium without doubling the voltage applied across
mylar capacitor C2. This allows capacitor C2 to
have a thinner dielectric and thus a higher capaci-
tance. Increasing the value of C2 relative to the
value of Cl brings the fraction C2 ~1+C2)nearer to
unity, which further increases the detector's
signal output. A further advantage of doubling
the voltage across the selenium is that the in-
creased electric field strength decreases electron-
hole pair recombination in the selenium and in-
creases the efficiency of ion-pair collection.
This increases the efficiency of the detector.
~1~2:~2
-44-
Increasing C2, or rather making Cl as
small as possible, has two beneficial results:
A. Reducing the capacity of Cl is
accomplished by increasing the thickness of
the selenium layer in the detector, which
increases the detector's X-ray absorption
efficiency.
B. Reducing Cl while using the highest
possible value of C2 reduces the total capaci-
tance of the detector, since Cl and C2 are in
series. This lower total capacitance in-
creases the scan speed that the invention can
achieve. This is important because the prior
art only teaches the segmenting of a plate
irlto small sections to avoid this large
capacitance.
CHARGING METHOD #3
It is also possible to use a detector
wherein the selenium layer has high effective dark
resistivity, but not a blocking contact, which is
not charged prior to x-ray exposure. That is, a
detector whose photoconductive layer has no sur-
face charge. In this method, an electric poten-
tial is placed across the previously uncharged
detector only during exposure to the x-rays. This
procedure will result in a modulated surface
charge conforming to the radio-opacity of the
object being x-rayed being placed on the photocon-
ductive layer. The modulated surface charge may
be read with a scanning laser beam as described
above.
EXPERIMENTAL SYSTEM #2
The ExPerimental ApParatus
To evaluate the operational character-
istics of the detector and readout means of the
present invention, a model detector as in Figure 1
33~
-45-
was built and mounted on an optical bench which
contained a He-Cd laser, collimating lens and
field stop. The C2 was formed of apprcximately 5
mil mylar plastic coated with a few angstroms of
gold, i.e., to where the surface resistance of the
mylar is equal to 20 ohms per square (layer 20,
Figure 1). The mylar was bonded to the selenium
layer with optical cement. The laser was pulsed
via a pulse generator. Charge method #2 was used
as described above. The detector was loaded with
various values of resistances in accordance with
the applied voltage. An amplifier with a gain of
1000 was used to condition the detector's output
signal prior to display on a storage oscilloscope.
A schematic of the amplifier is given in Figure
23. From photographs taken of the signals on the
oscilloscope's screen the total area under the
discharge curve was determined via planimeter to
estimate the charge set in motion by the laser.
This procedure was repeated after exposure to
measure exposure to X-ray radiation to determine
the number of coulombs of charge discharged by the
X-ray exposure.
Experimental Results
Experiments were designed to determine
if detector and readout means of the present
invention performed as predicted. Using Equation
(17), the total possible signal, in coulombs per
sguare centimeter, was calculated assuming Q = O,
Cl = 3.69 x 10 11 and C2 = 2.93 x 10 11 farads~
cm . Figure 14 is a graph of the number of cou-
lombs theoretically predicted by such a model as a
function of supply voltage applied across the
detector. Also shown on Figure 14 are several
experimentally measured values of charges col-
lected from experimental system ~2, which was
116Z33~:
-46-
constructed and tested in the experimental radio-
logy laboratory of M.D. Anderson Hospital in
Houston, Texas.
In experimental system #2 the selenium
thickness was measured as 150 microns. Five mil
mylar was employed as the second dielectric. All
charges calculated in Figure 14 assumed an active
area or pixel size of .3 cm2 and represe~t total
discharge of the pixel.
As can be seen from Figure 14, the
correlation of calculated and measured charge
collected is very good if one assumes a loss of
3.93 x 10 9 coulombs. This departure from the
theoretically obtained value may be due to incom-
plete discharge caused by the existence of a
blocking contact at the selenium/aluminum inter-
face, The fact that the emperical experimental
data does not extrapolate to zero may also imply
that as the plate nears total discharge, fewer
charges aré collected because more recombination
takes place at lower field strength.
Figure 15 is a graph illustrating the
electric field strength across the selenium layer
of the detector as a function of mylar thickness,
i.e. C2, for a given selenium thickness and vari-
ous supply voltages applied by charge method #2.
This graph is useful in predicting the dynamic
range of the system in coulombs for various combi-
nations of C2 and Vs~ In experimental system #2,
Cl is fixed at 3.69 x 10 1 farads/cm2 by the 150
micron thickness of selenium. Referring again to
Figure 15, if C2 is 8xlO 11 farads/cm2 and Vs at
2000V, we find the voltage applied to the selenium
will be 2500V, using charge method #2, and the
dynamic range of the detector will be lxlO 7
coul/cm2. This means that for each square centi-
~`~162~;~Z
-47-
meter of plate illuminated, a total of lxlO 7
coulombs will flow in the external circuit. X-ray
exposure to produce a latent image would decrease
this total.
The detector's sensitivity to X-ray
exposure is of great importance. That is, how
many coulombs of charge will be set in motion and
collected-per absorption of an X-ray photon.
Figure 16 is a graph showing the coulombs of
charge collected per roentgen exposure as a func-
tion of total exposure (measured in MAS (Milli-
amperes X seconds) applied to an X-ray tube).
The X-ray system employed with experi-
mental system #2 was a Siemens Mammomat designed
for operation with a xeroradiographic system. The
X-ray unit was operated at 44 kVp and had a half~
value-layer (HVL) of approximately 1.5 mm. of
aluminum. The receptor was housed in a cassette
made of 1/16th inch PVC and had a Bakelite shut-
ter.
Figure 16 shows that the sensitivity
(Q/roent.) of the present inyention is higher for
higher applied voltages and that the invention is
unexpectedly sensitive to low exposure levels.
The present invention's efficiency of collection,
expressed in coulombs per roentgen, is much higher
at low X-ray exposure levels than at high exposure
levels.
Figure 17 describes the same data as
Figure 16; however, the X-ray detector cassette is
equipped with a 1/16" thick PVC shutter. The high
attenuation of PVC to the X-rays employed in these
experiments was not appreciated at the time this
data was collected.
Figure 18 illustrates the efficiency of
the present invention in coulombs/roentgen, as a
~ a t ,'C
332
-48-
function of supply voltage. Again the system is
clearly more efficient at lower radiation levels.
Table IV simply presents the same data in tabular
form.
~2332
--49--
O~ ~~D ~ O C~l
u~
: ~ o ~ o
N ~ d' U'~ D
o~ I 0 0~ 0 1 ~
a~
In _ ~
~0 aJ I 00
c~J a c~J ~ u~ 1~
U~ 0, ~ C~
O ~ ~ ~ O U~
I~ r_ I~ LS~ ~ In
::- .~ C ~ L~
~ C~
l_
C 2~
~ <~
U~ _
O r I ~c~
O ~ C~J ~) Lf7 1~
C~l ~
~0 L" r ~1
F O + ~ + l + I
-- r~ C~J CO U~
cn ,-I~ ~ o c~l
g oO O g g
~ J DN ~7 r~ ~t ~
~C)
332
-50-
When a Bakelite shutter is used at 320
MAS and 2700V supply voltage, 93.2% of all avail-
able charges transfer. This represents near total
discharge. Further, in view of the variance shown
in Table IV in total charges available, the above
example does indeed represent total discharges
within experimental error.
Figure 19 illustrates total charge
collected as a function of voltage applied across
the detector. The curves diverge for the various
exposures ac- the applied voltage is increased.
This can be interpreted as meaning that the lati-
tude of the detector, i.e. its ability to diffe-
rentiate exposures of various levels, increases as
applied voltage increases. Optimum latitude can
be determined as a trade off between the dynamic
range of the system and the signal to noise (S/N)
ratio of the detector output signal.
Estimates of the values of Cl and C2
have been made based on calculations of capaci-
tance assuming selenium thickness of 150 microns
and a diélectric constant of 6.3; mylar thickness
of 4.3 mils and dielectric constant of 3.5. The
experimental results allow certain conclusions to
be drawn about the present invention:
(1) Use of charge method #2 has allowed
use of increased voltage across the selenium
while maintaining a lower voltage across the
mylar.
(2) For a given thickness of selenium,
the dynamic range of the system may be varied
by varying the mylar thickness and applied
voltage.
(3) The system's dynamic range must
necessarily be greater than the range in
charges neutralized by radiation absorption,
33Z
-51-
i.e. the x-rays cannot be allowed to comple-
tely discharge any part of the detector.
This is necessary so as to maintain a field
across the selenlum even in areas of greate~t
radiation exposure. Doing so will maintain a
good collection efficiency.
It is empirically estimated that any one
pixel should not be discharged beyond 50% of its
initial charge. This, however, is only an esti-
mate and has yet to be derived theoretically or by
optimization procedures.
(4) Using the data collected, an ex-
ample of the operational characteristics of
the system can be computed as follows.
Assuming 44 kVp, 320 MAS x-ray source and a 6
cm. lucite phantom, the maximum exposure to
the receptor is 370 mR. When operating the
invention at 2700V applied, 6.0xlO 9 coulombs
are placed in motion due to irradiating an
area of .3 cm2. Given a load resistance of
looo n and readout time of 4xlO 6 sec. (cor-
responds to 125,000 hz band width), an ampli-
fier gain of lO00, and a pixel size of 50
micron, the output signal is:
-9 2 -5 2
EoUt ~ 6xlO coul/cm x 2.5 x 10 cm x 1000 x lO00 qain
4 l9-6 sec
= 4.8 x lO 2
E t = 48mv
System noise may be evaluated as ampli-
fier noise plus Nyquist noise of the input resis-
tance
3;~
-52-
E = Vr4kTR~F
= ~ 4x1.37x10-23x300xlOOOx1.5x105 = 1.57~v
Using an amplifier with noise figure of
.8 nano volt/ ~ and 1.5xlO5 hz,
En = .8xlO 9 (1. sxlo5 ) 1/2 = .8xlO 9x3.87x102
= .309~v
Hence, the amplifier noise is much less
than the Nyquist noise of the input resistor.
Therefore, signal to noise ratio is:
48.
S/N = 1 57 = 30.5:1
this dynamic range will produce an image with
approximately 9-10 shades of gray. Here a shade
of gray is defined as
increase in signal. That is,
Shades of Gray = 2 loa S~N
log 2
EXPERIMENTAL SYSTEM #3
Figure 20 is a block diagram that illu-
strates the major components of the third experi-
mental system built at the experimental radiology
department of M.D. Anderson Hospital in Houston,
Texas. This experimental system was built to test
the entire integrated system of the present inven-
tion and to provide a breadboard system for ini-
tial clinical evaluation. This system is fully
computerized.
As is shown in Figure 20, microcomputer
2007, which is a Southwest Technical Products
6800, controls laser blanker 2015; lfi-bit d/a
converter 2011; x-position interface module 2017;
11~;;~332
-53-
12-bit d/a converter 2009; and teletype 2005.
Teletype 2005 is a model 33ASR. Sixteen-bit d/a
converter 2011 is an Analog Devices 1136 and
12-bit d/a converter 2009 is an Analog Devices
1132. Laser blanking unit and laser 2015 are a
Liconix model 4110 helium/cadmium laser.
Laser 2015 is in a light-tight housing,
not shown, such as the one functionally illu-
strated in Figure 9, above.
The photon beam from laser 201~ scans
detector 2019. The output of detector 2019 is
connected to the input of preamp 2021. The output
of preamp 2021 is connected to scan rate converter
2003; a/d converter 2023; and write laser modula-
tor and laser 2025.
The output of a/d converter 2023 is an
input to microcomputer 2007. Teletype 2005 is in
two way communication with microcomputer 2007.
It would be appreciated by looking at
Figure 20 that laser 2015 is a read laser adapted
to scan detector 2019 and laser 2025 is a write
laser adapted to write information out on a film
or xerographic plate. This type of system is
generally described in connection with Figure 11,
above.
Both the read and write laser beams are
controlled by the microcomputer. The output of
12-bit d/a converter 2009 drives read laser x-posi
tion scanner, controller and mirror transducer
2013. It also drives write laser x-position
scanner, controller and mirror transducer 2027.
Similarly, 16-bit d/a converter 2011 drives both
read laser y-position scanner, controller and
mirror transducer 2029 and write laser y-position
scanner, controller and mirror transducer 2031.
It will easily be appreciated that the beams from
~ 1 5~4 ~æ
both the read laser and the write laser scan in
sychronization, although both of them need not be
turned on at the same time or modulated by the
same signal.
Functionally, the microprocesser accepts
instructions from the teletype. These instruc-
tions operate on a computer program. A copy of
this program's listing is included with this
application for the Examiner's convenience. It is
the inventors' request that it be inserted in the
record so it will be available to the public as
part of the File Wrapper of this application.
When the microprocessor addresses the
16-bit d/a converter and increments the number
stored in it, the y-position mirrors of both
lasers are deflected to a certain angle. In
experimental system #3 this angle is sufficient to
move the spot that is illuminated on detector 2019
50 microns. Thus the y-positioning of the laser
beam on the detector occurs in 50 micron incre-
ments. The program included with this application
allows specification of line sizing on the y-axis
to 50, 100, 200, or 400 microns. This translates
to a resolution in the image produced by the
system of 10, 5, 2.5 and 1.25 line pairs per
millimeter.
The 12-bit d/a converter is not directly
addressed by the computer. The computer addresses
a special interface that controls the 12-bit d/a
converter. This interface allows the computer to
specify the starting and ending positions of the
laser beam on the detector surface and along the
x-axis. A schematic of x-position interface
module 2017 is included as Figure 24. The compu-
ter then issues a start command to the interface
which enables a switch selectable clock to start
~3æ
-55-
incrementing the d/a to a specific start address.
The clock continues to increment the d/a until the
stop address is encountered, then the clock is
disabled by the interface. This x-position inter-
face accomplishes two objectives. First, the data
readout range can be varied. Secondly, different
size detector plates can be read out without
overscanning them. This is especially valuable on
an experimental system.
Computer 2007 also controls laser blank-
ing, which just means that it turns the laser on
and off. The computer blanks the laser after each
x-axis scan is completed. The laser is then moved
back across the detector and turned back on for
the next scan line.
To read out a plate using experimental
system #3:
(1) The user specifies plate size and
x-axis line spacing. The computer program
will ask for these values.
(2) The x-axis interface is given the
appropriate start and stop addresses by the
computer.
(3) The x-axis line spacing and total
number of required steps is determined by the
computer.
(4) The computer blanks the CRT and
puts the scan rate converter into the write
mode.
(5) The laser is turned on by the
computer.
(6) The computer instructs the inter-
face to turn on the clock.
(7) The x-axis mirror sweeps the laser
across the plate.
~332
-56-
(8) The computer turns off the laser.
(9) The computer increments the y-axis
d/a the required number of steps which moves
the laser beam through the specified line
spacing.
(lO) The computer then repeats steps 5-9
until the entire plate has been scanned.
(ll) The computer puts the scan rate
converter in the read mode and the image is
displayed.
The read laser x-position scan control-
ler and y-pos-ition scan controller both feed into
scan rate converter 2003. The scan rate converter
allows a relatively slow video image, such as that
formed by experimental system #3, to be interfaced
with a television display. Scan rate converter
2003 outputs a normal video signal to CRT display
2001, where the image may be viewed, moved about
on the screen, or expanded to examine detail.
The example and suggested systems illu-
strated and discussed in this specification are
intended only to teach those skilled in the art
the best way known to the inventors to make and
use their invention. Nothing in this specifica-
tion should be considered as limiting the scope of
the present invention. Many changes could be made
by those skilled in the art to produce equivalent
systems without departing from the invention. For
example, it is possible to use a photoconductor
other than selenium as part of the detector sand-
wich. Holographic optics could be used in the
scanning system in place of the mechanical systems
used in the example. It may be desirable to alter
the relative thicknesses of the various layers
making up the sandwich structure or to provide for
different freguency of light beams to read out the
"
-57-
radiographic image or to sense different x-ray
potentials. The present invention should only be
limited by the following claims and their equiva-
lents.
