 |
Description  |
|
|
BACKGROUND OF THE INVENTION
The invention relates to apparatus and methods for improving the speed and
accuracy of biological x-ray radiology such as those used in mammography,
lithotripsy, and bone imaging, and to apparatus and techniques for
reducing the required x-ray dosage, improving contrast resolution,
improving the speed of the process, and improving the speed and
convenience of radiopaque needle localization and fine needle aspiration
biopsy techniques.
In conventional mammography, a woman sits in front of a mammography screen
and places her breast, designated by numeral 3 in FIG. 1, on a support 4
having thereon or therein a screen/film detector 4A that is sensitive to
x-rays. A breast compressor plate 2 that is transparent to x-rays presses
against the top of breast 3 to flatten it and prevent any movement of it
during the mammography process. An x-ray source 1 is turned on to produce
x-rays 5. The density of a tumor or microcalcification in the breast is
different than that of healthy breast tissue, and may appear as a lighter
or darker area on radiographic film 4A. Density variations which may be
indicative of lesions appear as variations in darkness of an image on
radiographic film 4A.
After the exposure of film 4A to the x-rays exiting from breast 3, film 4A
is developed using conventional processes, as indicated in block 6 of FIG.
1. This process typically requires 11/2 to 5 minutes. The film negative
obtained from the development process can be used in various ways. The
most common way of using the film negative is to simply place it on a
light box. The physician then can inspect it and hopefully visually
recognize any image density variations that suggest the presence of a
tumor or group of microcalcifications. Sometimes a digitizing machine is
utilized to scan and completely digitize the image appearing on the film
negative and store the digitized image data in a computer. The digitized
image usually is used for archival purposes and research purposes. (At the
present state of the art, the best image contrast resolution presently
available is 8 to 10 bits per pixel.) The computer then can perform
various known image processing operations on the data in order to identify
various features of the image. Arrow 10 in FIG. 1 designates storage of
digitized data, and block 11 represents processing of the image data. The
developed film negative typically is stored in a library archive after
light box viewing or digitization, as indicated in block 9 of FIG. 1.
There are a number of problems common to the above-described conventional
mammography techniques and other biological radiology processes. One
problem is that film development processing is much slower than is
desirable. Another problem with conventional mammography, lithotripsy, and
bone imaging is that x-ray dosages are higher than desirable. The human
body should be exposed to as few x-rays as is consistent with the
effectiveness of the particular diagnostic x-ray system. Unfortunately, it
sometimes is not known until after the step indicated in block 7 and/or 11
of FIG. 1 whether the film is underexposed or overexposed and whether more
x-rays of the woman's breast consequently are needed to accurately analyze
any possible tumors or microcalcifications. Furthermore, the 8-to-10 bit
intensity resolution per pixel contrast which is presently achievable is
not as high as desirable.
Another conventional mammography technique utilizes the apparatus and
method described above further augmented by a radiopaque needle
localization technique. A metal needle is inserted into a particular
location of the woman's breast as she stands in front of the mammography
screen. The needle is moved to various locations of the woman's breast on
a trial-and-error basis to "close in" on the abnormal density locations
indicated by the developed films. X-rays are taken and developed for each
insertion, until the tip of the needle is located precisely at the site of
a possible tumor or group of microcalcifications which have been visually
located in accordance with the procedure of block 7 in FIG. 1. Some prior
techniques feed the x-ray films to a film digitizer. A computer reads the
resulting digitized intensity values and computes how deep the needle tip
should be. In a procedure called stereotactic biopsy, two exposures are
taken using the same film. The x-ray source is tilted 45 degrees in
opposite directions for the two exposures. A computer or an operator
measures the displacement of a lesion on the film between the two
exposures and from that information computes the depth of the lesion
within the breast.
The procedure of obtaining a useable image and moving the needle on a
trial-and-error basis to locate it precisely at the site of a likely tumor
is time-consuming and very uncomfortable to the woman, who must remain
sitting at the mammography screen, without moving, until the needle is
properly located. After the needle has been properly located, the patient
then may go to surgery for a biopsy, wherein the surgeon follows an
incision along a needle to the tip of the needle and removes tissue
located thereat for analysis. Alternately, a radiopaque needle biopsy
aspiration technique can be performed while the breast remains compressed
by plate 2.
It is known in lithotripsy, in which a kidney stone or gallstone is located
by ultrasound imaging techniques and high intensity sonic energy is then
focused on the kidney stone to shatter it, that it would be desirable to
obtain faster determination of the location of the kidney stone or
gallstone because the patient is maintained under anesthesia during the
procedure. Increase of risk to the patient could be reduced substantially
by reducing the time under anesthesia. Similarly, known radiological bone
imaging techniques utilize x-rays and CCD (charge-coupled device) cameras,
but the amount of time required to obtain images and is greater than
desirable, and the amount of x-ray dosage required is greater than
desirable.
CCD's have been used in radiological imaging, and are known to have
inherent resolution capable of matching or exceeding the resolution of
developed x-ray film. See "HIGH RESOLUTION DIGITAL RADIOGRAPHY UTILIZING
CCD PLANAR ARRAY", by Shaber, Lockard, and Boone, presented in December
1989, and published in SPIE, Volume 914, Medical Imaging II, page 262-269.
(The present assignee provided the cooled, slow scan CCD camera and a
three stage Peltier thermoelectric cooler for the CCD planar array
utilized in the camera for the reported experiments. However, the
apparatus and technique described fails to enable a user to reduce the
x-ray dosage to the minimum levels needed to obtain the high level of
spatial resolution reported and to overcome some of the disadvantages of
prior radiopaque needle localization techniques.
There is a need for a technique to rapidly obtain results of x-ray
mammography, lithotripsy, and bone imaging with spatial resolution and
contrast resolution equal to or better than those achievable by
conventional x-ray film negatives, while avoiding the time-consuming
techniques and patient discomfort of prior mammography techniques and
prior needle localization techniques, lithotripsy techniques, bone imaging
techniques, and radiation therapy techniques.
Cooled, slow scan, low noise CCD cameras have been used in astronomy, and
also have been used in various industrial radiography applications, but
with much higher dosages of x-rays than is permissible in mammography and
other medical applications. As used herein, the term "cooled, slow scan"
refers to CCD arrays in which the CCD array is cooled, for example, by
means of thermoelectric devices, to temperatures at which thermal noise is
reduced to levels at which diagnostic quality radiological images can be
produced. Cooled, slow scan CCD cameras presently marketed by the assignee
have CCD array operating temperatures of about -20 degrees Centigrade or
lower. The "slow scan" terminology refers to scanning rates that are
substantially less than conventional video scan rates of 30 frames per
second. The "slow scan" terminology also refers to devices in which double
correlated sampling and dual slope integration techniques are
electronically accomplished to minimize electronic noise. This produces
improved contrast resolution and increased signal-to-noise ratios.
However, it is quite problematical to determine whether such cooled, slow
scan CCD cameras and techniques are practical in particular medical
radiology applications such as in mammography, because medical radiology
applications require minimum possible x-ray dosages, very high contrast
resolution and spatial resolution. In some cases, short times for image
generation are desirable, especially for needle localization techniques.
Some radiological applications, for example, coronary angiography, require
fast frame rates, e.g. 30 frames per second. This is inconsistent with the
high signal-to-noise ratios needed to achieve the high resolutions,
because fast cooled, slow scan CCD camera scanning rates needed for fast
image generation result in low signal-to-noise ratios.
SUMMARY OF THE INVENTION
It is an object of the invention to reduce the x-ray dosage required in
biological x-ray radiology while achieving acceptable contrast resolution
and spatial resolution.
It is another object of the invention to achieve higher contrast resolution
than is presently achievable for known x-ray mammography, lithotripsy, or
bone imaging film techniques and known stimulated phosphor screen/cooled,
slow scan CCD camera techniques.
It is another object of the invention to achieve more accurate results in
less time and with less inconvenience and discomfort to the patient than
is achievable, for example, in conventional x-ray mammography, especially
when known radiopaque needle localization and/or known fine needle biopsy
aspiration techniques are used.
It is another object of the invention to decrease the delay in obtaining
useable "film-quality" images in biological x-ray radiology procedures.
Briefly described, and in accordance with one embodiment thereof, the
invention provides a system for locating a region of varying density in
biological tissue, such as a breast, including a controlled x-ray source
for directing x-rays through the tissue, a phosphor screen positioned to
receive x-rays exiting from the tissue and producing light in response to
the exiting x-rays, a cooled, slow scan CCD camera, a variable
magnification optical system or a fiber optic reducer positioned to
receive light emitted by the phosphor screen and directing it to the
cooled, slow scan CCD camera, a high resolution monitor, and a control
system receiving signals from the cooled, slow scan CCD camera and
digitizing the signals to produce digital data representative of the
densities of various locations of the breast. The x-ray dosage is reduced
by repetitively reading a serial register in the cooled, slow scan CCD
camera during x-ray exposure and by determining in response to the
contents of the serial register when sufficient x-ray dosage has passed
through the tissue to produce an acceptable x-ray image and then
immediately terminating the x-ray dosage by shutting off the x-ray source.
The CCD array then is read out and the image is presented to a
radiologist. The control system in another embodiment in which the
variable magnification optical system is used, includes a processor
operated to determine a signal-to-noise ratio of light received by the
cooled, slow scan CCD camera from the variable magnification optical
system, to control the dosage of x-rays produced by the x-ray source in
response to the signal-to-noise ratios to determine locations of the
object at which the corresponding densities deviate by a prescribed
amount. Such locations then are digitally marked by storing addresses of
such locations. The variable magnification optical system then is operated
to focus and enlarge selected digitally marked locations of the object so
images thereof substantially fill a screen of the monitor. In one
described embodiment of the invention, either of the above systems is used
in conjunction with a radiopaque needle localization process to rapidly
and precisely identify locations of high density breast tissue which are
likely tumor and/or microcalcification sites, with minimum discomfort and
inconvenience to the patient. The x-ray dosage is reduced to the point
that only the radiopaque needle distinctly appears in the image during the
needle location procedures, thereby greatly reducing x-ray exposure of the
patient.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram of a conventional screen/film x-ray mammography
system using radiographic film to record x-ray patterns.
FIG. 2 is a diagram illustrating the CCD and phosphor screen digital
radiology apparatus of the present invention to achieve rapid, high
resolution mammography with low x-ray doses.
FIG. 2A is a diagram showing an optically zoomed high resolution image on
the monitor shown in FIG. 2.
FIG. 3 is a diagram including an automatic demagnification and refocusing
system, cooled, slow scan CCD camera, and camera scan electronics included
in the system of FIG. 2.
FIG. 4 is a diagram useful in explaining the operation of the apparatus of
FIG. 2.
FIG. 5 is a diagram of a system similar to that of FIG. 2, with a fiber
optic reducer replacing the lens coupling system of FIG. 2.
FIG. 6 is a diagram of a CCD dose measuring apparatus that can be used in
the embodiment of the invention shown in FIGS. 2 and 5, and also in other
dose measuring applications.
FIG. 7 is a diagram useful in describing application of the invention to
lithotripsy.
FIG. 8 is a flowchart useful in describing the operation of the embodiment
of FIG. 6.
DETAILED DESCRIPTION OF THE PRESENTLY PREFERRED EMBODIMENTS
FIG. 2 illustrates a high resolution x-ray system 50 that includes the
x-ray source 1 and a plate 2 compressing breast 3 to prevent any movement
thereof. A region 3A of high density breast tissue is shown. (Note that
object 3 can be tissue other than a breast, in which case plate 2 may not
be needed.) As in the conventional mammography x-ray apparatus of FIG. 1,
breast 3 is securely compressed between support 4 and plate 2. A phosphor
screen 15 is positioned in or on support 4 and receives the x-rays 5 that
have passed through and have been partially absorbed by breast 3 in
accordance with density of tissue therein. A suitable phosphor screen is
Model Min-R manufactured by Kodak, with an emission wavelength of 550
nanometers.
The x-ray photons 5 emitted from the x-ray source 1 pass through breast 3
and impinge on phosphor screen 15. Phosphor screen 15 absorbs a certain
percentage of the x-ray photons. Each x-ray photon generates a somewhat
variable number of light photons 13 determined by the absorption
efficiency and the emission efficiency of the phosphor screen. In response
to the energy of the x-rays, phosphor screen 15 emits light, the downward
portion 13 of it being captured by lens 17. The brightness of the light
emitted from a particular point of phosphor screen 15 is proportional to
the intensity of the x-rays exiting from a corresponding point of breast
3. (Optionally, the light 13 can be intercepted by radiographic film 4,
which can be developed and used for visual inspection by a physician.)
It should be noted that presently available two-dimensional CCD arrays are
much smaller in size than the physical size of a horizontal cross-section
through a woman's breast. It therefore may be desirable to have an
automatic demagnification-and-focusing system 19 to demagnify and
accordingly refocus the image produced by phosphor screen 15 in response
to x-rays 5 to the size of the lens of cooled, slow scan CCD camera 20.
A suitable optical interface 18 is provided between the phosphorous screen
15 and cooled, slow scan CCD camera 20 to demagnify the image emitted by
the phosphor screen into the aperture of cooled, slow scan CCD camera 20.
Optical interface 18 includes lens 17, which is a 50 millimeter coupling
lens that may have an f number of 0.7. Optical interface 18 also includes
an automatic focusing system 19, subsequently described. Coupling lens 17
collects as many light photons as possible from phosphor screen 15 and
demagnifies the image onto the lens of cooled, slow scan CCD camera 20.
Cooled, slow scan CCD camera 20 can be a high sensitivity, slow scan,
cooled CCD camera, such as a Model CH250 marketed by the assignee, which
is positioned to receive the demagnified, focused image from optical
interface 18.
FIG. 3 shows part of optical interference 18 of FIG. 2 in more detail A
light-tight flexible bellows 38 in FIG. 3 prevents outside light from
entering lens 40 and passing into cooled, slow scan CCD camera 20A.
Automatic focusing system 19 is controlled by signals 23 (FIG. 2) from a
motion control system 22, which in turn is controlled by signals 27 from a
main control system 26. Main control system 26 includes a MACINTOSH
computer available from Apple Computer Company and NUBUS camera controller
software commercially available from the present assignee. Main control
system 26 digitizes the information received from cooled, slow scan CCD
camera 20, so that quantified information about the intensity of light
incident on each pixel is received. Main control system 26 communicates
via signals IA with x-ray source 1 and motion control system 22 to
instruct camera 20A to zoom to a particular distance as indicated by
arrows 42 in FIG. 3 so that the object of interest will cover the entire
screen of monitor 30 and obtain a high resolution image. Main control
system 26 also supplies control signals IA to the x-ray source 1 to allow
automatic corresponding reduction of the x-ray dosage to the lowest level
commensurate with the signal-to-noise ratio or the image resolution needed
for the present phase of operation of system 50.
Motion control system 22 is capable of varying the distance of cooled, slow
scan CCD camera 20 from the carbon fiber window 36 shown in FIG. 3, which
has a maximum field of view of 16 centimeters by 16 centimeters. Motion
control system 22 can vary the distance between the camera head and carbon
fiber window from 48 centimeters to 66 centimeters, producing variable
demagnification. Optical interface 40 can be achieved by the use of a
f0.7, 50 millimeter Nikkor lens system and appropriate lens extenders.
Blocks 20B and 20C in FIG. 3 designate Model CE200 scan electronics
available from the present assignee and capable of scanning at 500
kilopixels per second, and a NUBUS controller card also available from the
present assignee, respectively.
Cooled, slow scan CCD camera 20 sends CCD signals 24 (FIG. 2) in a standard
format to control system 26. In response to CCD signals 24, main control
system 26 produces digital signals representing the light image produced
by phosphor screen 15 and representing the image of breast 3 and transmits
them to an archiving and communication system 33 for storage. Signals 28
from main control system 26 are standard video signals that are sent to a
conventional high resolution monitor 30, which displays an image 3' of
breast 3, with a darker image 3A' of the high density tissue 3A. A
radiologist manipulates a conventional mouse 31 to move a cursor on
monitor 30, and produces signals 28 that inform main control system 26 of
the location of the high density tissue 3A, thereby "digitally marking"
that location.
A control signal (not shown) initiated by the radiologist allows
amplification of image 3' with very high resolution by causing automation
focusing system 19 to focus on the digitally marked portion of breast 3,
producing the amplified high resolution image 3A' as shown in FIG. 2A. To
achieve digital marking, the image data is acquired and stored in the
following format: each pixel of the CCD detector is identified by its X
and Y coordinates, followed by the corresponding intensity level at that
pixel. The X and Y coordinates are calibrated to identify locations on the
phosphor screen. When the radiologist moves the cursor to identify a
particular area within the image displayed on monitor 30, the coordinates
of the corresponding area of breast 3 defined by the radiologist are
stored in a computer. Then motion control system 22 is activated, and
cooled, slow scan CCD camera 20 is electronically positioned and focused
by means of automatic focusing system 19 to limit the field of view to the
digitally marked region of interest.
Cooled, slow scan CCD camera 20 has associated with it conventional camera
electronics that need not be described in detail, except to note that the
electrical signals produced in response to light that is emitted by
phosphorous screen 15, demagnified by automatic focusing system 19 into
the lens of cooled, slow scan CCD camera 20, and impinging upon individual
CCD cells thereof are converted to suitable digital data having a
resolution of 12 or more digital bits per pixel. As previously indicated,
this high sensitivity equals or exceeds the sensitivity of presently
available radiographic film. In cooled, slow scan CCD camera 20, the CCD
detector is cooled and operated at a temperature of about 0 to -50 degrees
Centigrade, which limits thermal noise. Cooled, slow scan CCD camera 20
employs "slow-scan" electronics (i.e., substantially slower scan rate than
standard video scan rates) that limit electrical noise. High
signal-to-noise ratio thereby is achieved, resulting in high contrast
resolution which is desirable for mammography and other biological
radiology imaging applications.
In accordance with the above-described embodiment of the present invention,
an initial digital image is obtained for the entire breast 3, by operating
x-ray source 1 in response to main control system 26 to produce a
sufficiently high x-ray dose to obtain an initial diagnostic quality image
of the entire breast. That image is processed and stored in a computer
memory in main control system 26. The stored image is immediately
evaluated by a radiologist. The possible locations of tumors or
microcalcification are "digitally marked" in the computer memory so that
they can be instantly relocated.
Then, the x-ray dosage is reduced considerably in accordance with the
technique subsequently described with reference to FIG. 4. In the
embodiment of FIG. 2, the automatic focusing system then collimates the
reduced x-ray dose on a much smaller region of breast 3 which corresponds
to one of the digitally marked locations. This procedure is repeated for
all of the other digitally marked high density locations of the breast,
and results in a much lower overall x-ray dosage to the patient because
high doses are applied only to the high density tissue regions of the
breast, which high density tissue regions constitute a very small
percentage of the total breast tissue.
Variable demagnification and automatic refocusing by the system 19 is very
important in order to enable the physician to minimize the amount of
exposure of the woman's breast to x-rays by the above procedure of first
taking an x-ray of the entire breast, and after initial analysis, focusing
the analysis on more magnified subsequent enlarged images of likely tumor
or microcalcification sites, with minimum acceptable x-ray dosages. To
understand how the x-ray dosage is reduced from an initial level in order
to view a digitally marked region of breast 3, it would be helpful to
refer to FIG. 4.
In FIG. 4, numeral 51 designates a field of view essentially fully occupied
by breast 3. Field of view 51 is square, having a dimension of L.sub.f per
side. L.sub.f may be 16 centimeters. Numeral 52 designates a digitally
marked square region within field of view 51 having s | | |
