GB2072839A

GB2072839A - Apparatus for Detecting, Localizing and Quantifying Coronary Stenoses from the Body Surface using Gamma-ray-emitting Particles and Coded Apertures

Info

Publication number: GB2072839A
Application number: GB8108830A
Authority: GB
Original assignee: Individual
Current assignee: Individual
Priority date: 1980-03-24
Filing date: 1981-03-20
Publication date: 1981-10-07
Also published as: IL62236A0; JPS579434A; FR2478826A1; DE3110516A1; FR2493148A3

Abstract

The extent, location, and severity of stenotic atherosclerotic lesions in the coronary arteries are determined by intravenous injection into the circulating blood of a number of small, discrete gamma-ray-emitting particles and by tracking the particles in three dimensions in the region of the heart using suitable gamma detectors (29), which have coded apertures 23 including regions having different transmission factors for the radiation and which are disposed external to the patient's chest. The resolution and counting rates of the gamma detectors permit the sequence of positions of each particle to be recorded as a function of time as each particle flows through a coronary vessel. Data analysis of these recordings of particle position provides information about the velocity of blood flow through the course of each coronary artery. This information is used to determine the course of the vessel of interest and the extent, severity, and location of stenotic lesions therealong. <IMAGE>

Description

SPECIFICATION Apparatus for Detecting Localizing and Quantifying Coronary Stanoses From the Body surface Using Gamma-ray-emitting Particles and Coded Apertures This invention is concerned with apparatus for detecting, localizing and quantifying coronary stenoses from the body surface using gamma-ray-emitting particles and coded apertures.

Coronary heart diseases (CHD) is the leading cause of death in the United States and the Western world. Coronary heart disease accounts for almost two-thirds of male deaths during that period of life #30--64 years) when responsibilities to family and society are the greatest. Approximately one-third of individuals dying of coronary heart disease succumb immediately after coronary occlusions; one-third die within a few hours; and only one-third receive the benefits of hospital therapy. Of all individuals sustaining their first myocardial infarction, more than half have had neither preceding signs nor symptoms of coronary heart disease, In recent years, coronary vein-graft by-pass surgery has been demonstrated to be an effective therapeutic modality of wide applicability.Because over 70% of all coronary artery stenoses occur in the first 4 cm. of the coronary arterial tree, a high percentage of all coronary lesions can be surgically by-passed with a low mortality and high patency rate.

Extensive epidemiological studies have delineated those factors (hypercholesterolemia, hypertension, obesity, and heavy smoking) that are statistically correlated with an increased incidence of coronary heart disease. But while a patient may be well advised to remove himself from the higher risk group by appropriately altering his manner of living, these indices cannot for any given patient furnish information concerning the existence of coronary stenotic lesions nor furnish a basis for clinical decisions regarding therapeutic intervention.

Electrocardiographic stress testing has been suggested as a means for screening individuals for significant coronary lesions. However, in a large prospective study in which subjects underwent repeated testing, the sensitivity of this test was found to be too low (only 30%) to be adequate.

At present, coronary arteriography yields more useful information about the state of the coronary arteries than any other technique. However, in a large cooperative study, the incidence of major complications was 2% and the mortality rate was 0.?.3%. In addition to the dangers, the procedure is painful, expensive, and time-consuming. For these reasons, coronary angiography is not performed upon asymptomatic individuals and is not suitable for screening large populations.

One embodiment of apparatus for detecting the extent, location, and severity of coronary stenotic lesions while operating external to the body is disclosed in U.S. Patent Specification Number 4,111,191.

The present invention provides apparatus for interacting with a number of particles which can be introduced into the blood of a subject and which can provide signals from within a vessel of interest of the subject that can be detected outside the body, the apparatus comprising: detector means including signal-selection means positionable near the body of the subject for detecting the signals from the number of particles at successive locations of each of the number of particles at successive times during the travel of each particle through a vessel of interest, said signalselection means including spaced regions thereof having different relative transmissiveness therethrough for signal from a particle within the body of the subject;; circuit means coupled to said detector means for providing output manifestations from successive positions of each of the number of particles at successive times during the travel thereof through the vessel of interest; and output means responsive to the output manifestations from said circuit means for providing an indication of the relative internal cross-sectional areas of the vessel of interest at successive positions along the course of the vessel of interest.

In apparatus as set forth in the last preceding paragraph, it is preferred that the signal-selection means is disposed intermediate the detector means and the body of the subject for intercepting signal from a particle therewithin.

In apparatus as set forth in either one of the last two immediately preceding paragraphs, it is preferred that the ratio of area of relatively high signal transmissiveness to the area of relatively low signal transmissiveness is within the range from about 1 to .45.

In apparatus as set forth in any one of the last three immediately preceding paragraphs, it is preferred that said detector means is arranged to produce electrical signals indicative of the location thereon at which a signal is received through the signal-selection means.

In apparatus as set forth in any one of the last four immediately preceding paragraphs, it is preferred that the signal-selection means includes an element of relatively low signal transmissiveness therethrough having at least one aperture therein of relatively higher signal transmissiveness therethrough.

In apparatus as set forth in the last preceding paragraph, it is preferred that said detector means is arranged to respond to signal received thereby from a particle within the body of the subject and to produce said electrical signals representative of the coordinates of the location thereon at which a signal from the body of the subject is received along a trajectory passing through said aperture.

In apparatus as set forth in the last preceding paragraph, it is preferred that the signal-selection means includes a substantially circular toroidal aperture of relatively high transmissiveness to such signal. Alternatively, the signal-selection means includes a plurality of apertures disposed at spaced locations along orthogonally oriented axes.

In apparatus as set forth in the last preceding paragraph but one, it is preferred that said detector means is arranged to produce said electrical signals in response to gamma radiation received thereby.

In apparatus as set forth in any one of the last eight immediately preceding paragraphs, it is preferred that said circuit means includes input means connected to receive electrocardiographic signals from the subject for controlling said circuit means to respond to detection by the detector means during a selected portion of the cardiac cycle of a signal from the body of the subject which is received along a trajectory of said signal that is transmitted through one of said apertures.

The present invention provides apparatus for interacting with a number of particles which can be introduced into the blood of a subject and which can produce radiation from within a vessel of interest that passes through the body walls of the subject from the location of the particle within the body, the apparatus comprising; detector means including signal-selection means having spaced regions of different relative transmissiveness therethrough of the radiation, said detector means being positionable about the body of the subject for receiving through one of said spaced regions the radiation emanating from the body of the subject to produce electrical signals in response thereto which are indicative of successive locations of each particle of the number of particles at successive times during travel of each particle through the vessel of interest; and circuit means including timing means coupled to said detector means for providing output manifestations indicative of the three-dimensional co-ordinates of successive positions of each particle.

of the number of particles which is within the vessel of interest and from which the detector means receives the radiation along a trajectorv path through said spaced regions during a time interval determined by said timing means.

In apparatus as set forth in the last preceding paragraph, it is preferred that said circuit means includes display apparatus responsive to said output manifestations for providing an output indication of sequential locations as a function of time of detections through said spaced regions to provide an indication of the course of the vessel of interest from the successive positions therealong of the detected particles.

In apparatus as set forth in the last preceding paragraph but ten, it is preferred that said circuit means includes data processing means responsive to said output manifestations for producing an indication of the frequencies of occurrence of such output manifestations within increments along at least one co-ordinate dimension that is representative of increments of length along the vessel of interest.

in apparatus as set forth in the last preceding paragraph, it is preferred that said data processing means is responsive to the output manifestations representative of detected signals passing in virtual straigh-line trajectories through the signal-selection means for identifying the virtual location of the signal source within the region of convergence of the straight-line trajectories.

In apparatus as set forth in the last preceding paragraph, it is preferred that said circuit means includes display apparatus responsive to the outputs from said signal-selection means for providing an output indication of sequential locations as a function of time of said regions of convergence to provide an indication of the relative internal cross-sectional areas of the vessel of interest at successive positions along the course of the vessel of interest.

In apparatus as set forth in the last preceding paragraph but one, it is preferred that said data processing means is responsive to the successive locations during each of said time intervals of said regions of convergence representing a particle for indicating the relative cross-sectional areas at various locations along a vessel of interest, thereby to determine the existence, severity and location of stenotic or dilated regions of said blood vessel.

In accordance with the illustrated embodiment of the present invention, stenotic atherosclerotic lesions of the coronary arteries are detected by injecting a number of gamma-ray-emitting particles into the circulating blood of a subject and detecting the emitted gamma ray through coded apertures to determine the velocity of blood flow through his coronary vessels.

Because of the high peripheral resistance of the myocardial vascular bed and the considerable range of auto-regulatory resistance changes available to the coronary circulation, coronary stenoses of 80%~90% are required to appreciably diminish the volume of coronary blood flow. This propensity of volumetric coronary blood flow to remain normal even in the presence of severe stenoses is responsible for the late occurrence or absence of anginal symptoms and diagnostic electrocardiogram patterns, even in the presence of coronary stenoses, and explains why measurement of volumetric blood flow provides only poor indications of coronary disease.

However, this propensity of volumetric blood flow to remain normal even in the presence of severe stenosis furnishes a distinctive characteristic that blood flowing through a stenotic arterial segment must have high velocity. In fact, to maintain the constant volumetric flow rate, the average fluid velocity within a stenotic segment of artery must change in strict inverse proportionality to the change in cross-sectional areas from normal to stenotic blood vessel. As coronary stenosis becomes more severe the increments in blood flow velocity become progressively greater.

In accordance with the illustrated embodiment of the present invention, discrete gamma-rayemitting radioactive particles of sufficiently small size to pass through capillary beds are injected intravenously and become distributed in the circulating blood volume. The particles that appear in the region of the heart are tracked in three dimensions, for example, through coded apertures associated with gamma-ray cameras where each coded aperture includes radiation transparent and opaque portions. A pinhole collimator, a parallel-channel collimator, as well as the coded aperture used in the Ulustrated embodiment are all described below.The recordings of particle position as a function of time are analyzed and whenever a particle follows a path indicating that it is passing through a coronary artery, the velocity of blood as it flows through the artery is measured by timing the transit of the particle. From the accumulated data of multiple particle transits through the coronary circulation, a representation of the cross-section of the lumen of the coronary arterial system is constructed.

There now follows a detailed description which is to be read with reference to the accompanying drawings of apparatus according to the invention which has been selected for description to illustrate the invention by way of example.

In the accompanying drawings:- Figure 1 is a pictorial view of the human heart showing the principal coronary arteries.

Figure 2A is a pictorial and schematic diagram of the present invention using conventional detectors and paraliel-channel collimators for viewing the source particles from two different angles to allow the source particles to be tracked in three dimensions; Figure 2B is a pictorial and schematic diagram of the present invention using conventional detectors and focussing collimators in a camera system in which the magnification depends on the distance from the source to the collimator plane; Figure 3 is a pictorial and schematic diagram of the present invention using a conventional detector and a coded aperture containing a random array of square openings; Figure 4 is a pictorial and schematic diagram of the present invention using a conventional detector and a coded aperture screen containing a single annular opening;; Figure 5 is a process diagram showing data analysis which may be used with the embodiments illustrated in Figures 2, 3 and 4; and Figures 6A and 6B are graphs showing, respectively, the relative frequency of detection along the length of a coronary vessel and the correlation thereof with the cross-sectional area of the associated lumen.

Referring now to Figure 1, it should be noted that there are three main coronary arteries. These three vessels branch somewhat irregularly to form an average of ten secondary vessels, as shown.

Arteriosclerotic lesions are typically limited to the epicardial segments of the coronary vessels and rarely extend beyond the most proximal portions of the secondary vessels. The highest concentration of arteriosclerotic lesions is within the first 2 to 3 cm. of the left anterior descendens artery 3, but the lesions are otherwise rather randomly distributed in the proximal primary and secondary arteries.

Seventy percent of all arteriosclerotic coronary lesions are found within the proximal 4 cm. of the main coronary arteries.

The average velocity of blood flow through the epicardial coronary blood vessels is of the order of 30 cm./sec. A 50% stenosis is generally considered to be significant. To be useful, the system should be capable of discriminating between normal vessels and 50% stenotic lesions and should be capable of assessing additional significant decrements in the cross-sections of these vessels.

Typical blood flow velocities through stenoses of varying degrees are as follows:~ Degree of Stenosis Average Velocity 0% 30 cm/sec 50% 60cm/sec 60% 75 cm/sec 70% 100cm/sec 80% 1 50 cm/sec 90% 300cm/sec Since nominal resting coronary flow velocity is about 30 cm/sec, ideally the system should be able to differentiate a flow velocity of 60-75 cm/sec from 30 cm/sec in order to detect significant lesions, and to discriminate between velocities of 75, 100, 1 50 and 300 cm/sec in order to follow additional 10% increments in stenoses.

In accordance with the preferred embodiment, the present invention tracks a number of discrete moving gamma-ray-emitting sources present in the circulating blood. The system not only locates the position of the sources in three dimensions, but also locates them again and again at very short time intervals. The requirement as to how often a particle source must be located is determined from the following considerations.

As noted above, 30 cm/sec is about the nominal velocity of blood flow through a coronary artery for a subject at rest. Blood flow through a diseased arterial segment which is 80% stenotic will have a velocity five times this nominal value (150 cm/sec). Blood flowing through a segment of a blood vessel with a severe 90% stenosis will have a velocity of the order of 300 cm/sec. In order to measure a 90% stenosis that is 1 cm long, a particle moving through the stenosis should be detected at least a few times. At 300 cm/sec, only three milliseconds are required to pass through a 1 cm length of vessel.

As noted above, there are three main coronary arteries which branch into an average of ten secondary branches. If, for statistical purposes, it is desired that three velocity measurements be made through each of the ten secondary branches, then statistically a total of 3x 10=30 particle transmits through the coronary system would be required. This would furnish approximately 10 transits through each of the proximal principal coronary arteries where most of the atheromatous lesions are located.

Since coronary blood flow approximates only 5% of the cardiac output at rest, a given particle has only a .05 probability of entering the coronary circulation after a single pass through the heart. Thus, 20 circulations through the heart times 30 particle transits through the coronary system, or a total of 600 particle transits through the circulation would furnish the redundancy of coronary blood flow velocity measurements outlined above.

Since the mean circulation time is one minute or less, a single particle tracked in the circulation for 600 minutes would be suitable for the outlined redundancy, if its half-life were sufficiently long and if it continued to circulate for the ten-hour period. Of course, ten hours is an inconveniently long duration for a diagnostic measurement. Forty particles circulating for fifteen minutes would be much more convenient and would furnish a comparable 600 particle transit through the cirulatory system.

The number of particles required to attain 600 particle transits through the circulation is influenced by the possibility that the gamma-ray-emitting particles may be removed from the circulation by the Kupffer cells of the liver. The propensity of the liver to extract particles is a function of their size and surface characteristics, of the state of the reticuloendothelial system as influenced by pre-treatment and otherwise, and of other variables.

Liver blood flow is about 25% of cardiac output at rest. If pre-treatment is moderately successful, the particles will be extracted by the liver with about 50% efficiency, and the number of circulating particles will be reduced by a factor of 2 every 8 minutes.

If the half-life of the isotope used is longer than about 1 5 minutes, a reasonable program of particle administration might begin with an initial intravenous injection of 40 particles, followed by a continuous injection of 10 particles every 3 minutes. The number of particles in circulation would remain constant at about 40 throughout the 15 minute study, and a total of 87 particles would be administered.

If the half-life of the isotope used is very short, say 5 sec, a continuous intravenous administration of 40 source particles per minute would be required, since the source particles would only be useful on their first pass through the heart. (In a 60-sec interval between passes through the heart, the activity on each particle will decrease by a factor of 4000.) In this case, a total of 600 source particles would be administered.

The complexity of data processing is significantly influenced by the number of particles that must be simultaneously detected in the field of view that encompasses the heart. It is for this reason that the average number of circulating particles is kept at about 40 in the preceding examples. Assuming that 5% of the total blood volume is in the heart and that an additional 5% lies in the lungs and chest wall near the heart, an average of 4 source particles statistically will lie within the field of view that encompasses the heart.

General Operation In accordance with this invention, the location of all sources near the area of the heart is to be determined every 1 to 3 milliseconds with a spatial accuracy of + 1.5 mm.

For a number of years, the Anger Scintillation Camera has been used to image distributions of gamma-ray-emitting isotopes within the body (see, for example, U.S. Patent Specification No.

3,01 1,057). Commercially available cameras have spatial accuracies in the range from +3 mm to f5 mm when imaging 141 keV gamma rays. (The notation +3 mm used herein means that the line spread function has a half width at half maximum of 3 mm.) Conventional imaging procedures use a pinhole or multiple-channel collimator which performs a simple one-to-one mapping of the distribution of gamma-ray-emitting isotopes onto the scintillation camera. Unfortunately, this procedure does not provide three-dimensional information and the resulting image is only a two-dimensional projection of the distribution onto the plane of the scintillation camera. In order to track the gamma-ray-emitting sources in three dimensions, it is necessary that at least two imaging systems track the same source particles from different locations.

In the illustrated embodiment of Figure 2A, two separate camera systems 5 and 7 view the heart area from two different locations. These camera systems 5 and 7 may use pinhole, diverging, parallelchannel, or converging collimators 9 and 1 1.

In the illustrated embodiment of Figure 2B, it should be noted that when a pinhole or multichannel converging (i.e. focusing) collimator 13, 1 5 is used, the magnification of the image relative to the subject depends in a known way on the subject-to-camera distance. Thus, by positioning a variable magnification collimator camera system anterior, and another similar system posterior, to the heart area, a record of the apparent track and velocity of the source particle as seen by the two camera systems can be analyzed to determine both the true velocity of the particle and its track in three dimensions.

The geometrical transmission of commonly used multi-channel collimators having an accuracy of +5 mm at 141 keV is such that for every mCi of source strength about 10,000 gamma rays per second reach the scintillator crystal. Of these, about 8,000 are scattered in the patient and most can be rejected by the conventional pulse-height selection performed by the camera. The result is that for every mCi of source strength, about 2,000 unscattered gamma rays are detected each second and an equal number of scattered gamma rays pass the pulse-height threshold per second.In order for the source particle to signal its position every millisecond with an accuracy of +5 mm by producing a recorded dot, the strength of each source particle must be about 500,uCi. Although the accuracy of each particle detection does not meet the required +1.5 mm, an improvement in accuracy is possible when the positions of many dots are averaged. For example, in the case of 70% stenosis, the source particle moves 1 cm every 10 ms and it is feasible to average the positions of 1O dots to achieve the desired positional accuracy. In the case of 0% stenosis 30 dots are available every cm of travel.

In another embodiment, a pinhole collimator having i5 mm resolution at 140 keV has less than 50% of the transmission of the parallel channel collimator, and would require an even greater source activity. As discussed above, for long-lived isotopes, about 80 particles are needed to trace each coronary artery the desired number of times, and this requires a total activity of 40 mCi. In order to keep the dose to the liver within reasonable levels, it will be advantageous to pre-inject colloidal carbon particles to occupy the Kupffer cells of the liver and prevent the absorption of source particles. It is also advantageous to use short-lived isotopes, such as those produced by the decay of longer-lived isotopes.

One example is a 4.9 sec lrl9lm that emits only 129 keV gamma rays and is produced in the decay of 15-day 05191. The gamma ray energy is well matched to the camera and the absorption properties of lead, and the short half-life results in a very low patient dose. In a typical liver scan using 6-hr Tc99m sulfur colloid, the dose to the liver is 3 rads for a 3 mCi injection. Using 600 particles each containing 500 yCi of 4.9 sec lrl9lm, the corresponding dose would be less than 0.05 rad.

A more efficient approach for the tracking of a small number of discrete gamma-ray-emitting particles is the use of coded apertures according to the present invention. This allows the detection of a greater fraction of the gamma rays emitted by the source and also provides three-dimensional positional information. One embodiment of a coded aperture is shown in Figure 3 as a screen 21 containing a random pattern of square radiation-transmissive windows or openings 23. It is assumed that the gamma rays 27 emitted by the source particlesd 25 are stopped by the screen 21 unless they pass through one of the openings 23.

For a single, stationary source particle 25, the conventional gamma-ray image detector 29 will simply record a shadow pattern of the screen 21. This shadow pattern has several useful properties.

For a given distance from the screen 21 to the detector 29, the size of the pattern is uniquely related to the distance from the source particle 25 to the screen 21, and the position of the pattern is uniquely related to the position of the source particle 25 in a plane parallel to the screen 21. One means of determining the position of the source particle 25 in three dimensions is by assuming a large number of test positions (say, within a regular three-dimensional volume) and choosing the position that is most consistent with the observed pattern. This is done by effectively connecting lines from the test position to the point of detection on the detector 29 appearing within a given time interval and determining the fraction (Q) that passes through open portions 23 of the screen 21.The point in space with the highest figure of merit Q is the true source position. (The accuracy of this procedure is limited by statistical fluctuations.) For the case of several source particles 25, Q is maximized at each source particle position. If the square openings 23 in the screen 21 amount to a fraction F of its total area and if there is no scattering of gamma rays in the patient, then Q=F for test points far from the true source position, and Qo=1 for test points at the true source position. In the vicinity of the true source position, Q will vary between the two extreme values.

The Compton scattering of gamma rays will degrade the observed pattern and reduce IQ. When imaging the human heart using gamma rays of energies near 129 keV, it is found that of all the gamma rays passing the pulse-height threshold of the detector, a fraction S (approximately equal to 0.5) has scattered in the tissue of the patient. (Note that for every gamma ray accepted by the camera, approximately 1.5 have been rejected.) Considering Compton scattering, #o= 1-S(1-F) while Q=F, unaffected by scattering.

The presence of more than one source particle in the field of view further degrades the figure of merit so that Q0= [ 1-S(1-F)+(N-1)F ] /N at a source position and Q1=F away from all source positions, where N is the number of source particles in the field of view.

For a scattering probabilitv S=0.5 and for N=4 source particles, this becomes Q=(1 +7F)/8 and Q,=F. We wish to choose the screen transparency factor F (i.e., the ratio of square openings 23 to the total screen area) so that the minimum source activity is necessary to accurately detect the 4 source particles. The values of Q0 and Q, are subject to statistical fluctuations. A typical fluctuation in Q0 (actually, is standard deviation) is (QJ4ND)"2 where ND is the number of detected gamma rays per source particle.If it is required that QO and Q, differ by five standard deviations of Q: Q0#Q1=5(Q,,/4ND)112, or

This requires ND=50 (1 +7F) (1-F)-2. The number of photons detected ND is related to the number Ns emitted by the source, the geometrical transmission of the screen (about 0.15 F if the screen subtends 2 steradian) and the probability of passing the pulse height selection of the detector (about 0.4): Ns=ND (0.060 F)-l Ns=833 (1 +7F) (1-F)-2 (F)-1 The following table shows the relationship between N9 and F:: F N9 Q Oi 0.5 30,400 0.56 0.50 0.4 22,400 0.48 0.40 0.3 17,900 0.39 0.30 0.25 16,500 0.34 0.25 0.20 15,800 0.30 0.20 0.175 15,600 0.28 0.175 0.15 16,000 0.26 0.15 0.10 17,600 0.21 0.10 0.05 25,400 0.17 0.05 Within a preferred range of F~.3 to F#.1,selecting F=0.175 requires that 15,600 gamma rays be emitted by each of the 4 source particles. In order to track the particles using 3 millisecond time frames, the activity is then 15,600 per framex333 frames/sec-5x 108 per sec or 140yCi per source particle. The gamma-ray detector would detect 216,000 gamma rays per second within its pulse height window. This is within current technology.

A more general analysis shows that the optimal value of F is not very sensitive to the number of source particles N, and the number of source emissions N9 rises with N.

Activity required per source particle for N F(optimal) Ns 3 millisecond frames 1 .218 7,700 70 jt4Ci 2 .214 10,400 95#Ci 3 .193 13,000 120,uCi 4 .176 15,600 140yCi 5 .164 18,000 160 4Ci 6 .153 20,400 180,uCi 10 .126 23,700 210 Ci

Referring now to Figure 4, there is shown a coded-aperture screen 31 with a single annular aperture 33. A single point source 35 projects a circular pattern 36 on the imaging detector 37, the size of the circle being uniquely, related to the distance from the point source 35 to the screen 31, and the position of the circle being uniquely related to the position of the source 35 in the plane parallel to the screen 31. If the circular aperture 33 in the screen 31 is, say 5 cm in diameter, and the aperture is 0.5 cm wide, then 4 particles in the heart area will produce 4 overlapping circles with little confusion.

For a source particle 10 cm below the screen 31, the geometrical transmission of this aperture is g=6.3x 1 0-3. The number of gamma rays N9 emitted by each source particle and the number (ND) passing the pulse-height threshold are related by: ND=N5(0.4) g=2.5x 1 0-3 Ns Requiring ND=60 detected gamma rays (half of which are scattered), we have No=24,000 emitted by each source particle. In order to track the particles using 3 millisecond time frames, the activity is then 24,000 per framex333 frames/sec=8.0x 106 per sec or 220 zbCi per source particle.

Data Analysis In accordance with the present invention, the system can be operated through conventional gates synchronized to the patient's heartbeat in order to observe sources only during the diastolic phase of the cardiac cycle when heart motion is least and the velocity of blood in the coronary arteries is greatest and most constant Cardiac diastole characteristically occupies 400-600 milliseconds, which is considerably longer than the transit time (30-300 milliseconds) required for a particle to ~flow through a 10 cm length of coronary artery.

The collimation, coding and detection techniques discussed above all serve to provide a list of times and the three-dimensional coordinates of each source particle at those times. The following table illustrates the behavior of source particles in the heart chambers and the contrasting behavior of source particles moving through the coronary arteries and into the myocardial capillary bed.

In Heart In Coronary In Mycardíál Chambers Arteries Capillary Bed Average speed (cm/sec)- < 30 30-300 < 1 Average duration in area listed 2 sec. .03-.3 sec 2 sec Type of trajectory random linear with basically occasional stationary turn Moves downward 5 to No Yes No 10 cm in < 0.3 sec, then stops for 2 sec.

Average number of 2 0.01 0.1 source particles present Average interval 1.5 sec 30 sec 30 sec between passes through area listed Basically stationary No No Yes over 2 or 3 successive cycles (all number approximate) These differences form the basis upon which the data is analyzed, as described with reference to Figure 5. First, the data pertaining to the coordinates and times of occurrence of detected rays may be stored on tape, disc, or the like, 41, 42 for subsequent processing. All source particles that have appeared in the myocardial capillary bed since the start of the study may then be identified 43. Each of these initially took some path through the coronary arterial system. As indicated in the preceding table, source particles in the myocardial capillary bed are very distinctive because of their very low speed relative to the heart wall.

Of the 900 diastolic periods occurring during the period of an examination, select only about 30 that are related to the appearance of a source particle in the myocardium 44. Thus, this first step rejects 97% of the diastolic periods that are of no further interest.

Second, examine the data (arranged in 3 millisecond frames and played in reverse) preceding the appearance of the source particle in the myocardium 45. The source particle will appear to be dislodged from the heart wall and move rapidly (30-300 cm/sec) and generally upward toward the aorta.

A single three-dimensional picture (or at least two planar displays that are orthogonally oriented to present several views) representing the spatial distribution of all source-particle detections during the 3 millisecond frames of their rapid motion may then be displayed 46. The shape of the coronary arterial tree can best be seen from such a display and the frequency of detection within such time frames gives an indication of arterial cross section. The regions along any artery showing lower frequency of detection would thus be identified as prime candidates for arterial stenosis and further analysis.The lower frequency of detection is attributable to rapid motion of a source particle through a stenotic region such that the dot density of frequency of detection along the artery is thus directly proportional to the cross-sectional area of the arterial lumen, as illustrated in the graph of Figures 6A and 6B. For the larger coronary arteries, data from many passes can be added, resulting in a distribution with reduced statistical fluctuations. Although the information available at this point is most likely sufficient for clinical diagnosis of the existence, extent, and location of coronary heart disease, the following additional step of analysis may be used.

Third, the distributions obtained in the second step provide three-dimensional information regarding the detailed shape of the coronary arterial tree. This can be stored in computer memory as a list of interconnected straight line segments. For each source particle, the most likely route is determined. Thus, the average speed of a source particle along any line segment may be determined by taking the average of X,~X1~1 T1-T1#1 where Xis the source particle position (projected on a line segment) at time Ti. Note that it is necessary to use these averaging procedures due to the limited accuracy of the coordinates of individual dots.

The Particle Radiators Although 4.9 sec lrl9lm is referred to herein, severai other similar source particles such as Tc99m may also be used in accordance with the present invention. Several additionally useful particle sources and their properties are listed as follows:~ Parent Isotope Mo90 Ta183 05191 Hgl95m Mo99 Parent half-life 5.7 h 5.1 d 1 5 d 40 h 67 d Daughter Isotope Nb90m W183m lr191m Aurssm Tic99" Daughter half-life 24s 5.3 5 4.9 5 31 s 6.0 h Daughter emissions (keV) 122 106,160 129 261 144 f stablel f stablelstable Granddaughter 1 5 h Nb90 l W183 f l Ir191 j 183 d Au195 Tc99 Other decay schemes exist that involve much higher gamma ray energies, such as 68 min Gae8 (511 keV gammas) from 275 day Genes, But such source particles present the fundamental problem of designing a useful coded aperture plate for such penetrating radiation. For this reason, the range of gamma ray energies between 100 and 300 keV was concentrated on in the above discussion.

The circulating particles should be no larger than 6-8 microns so that they will freely pass through capillary beds, and each particle should have a specific activity of about 100 to 200 yCi. The particles may be small crystals of an insoluble salt containing the desired gamma-emitting isotope, or small particles of very absorbent material (such as molecular sieve) onto which the gamma-emitting Isotope has been attached.

The small crystals of insoluble salt have the advantage of very high specific activity, but such source particles must be screened for size before injection. An absorbent carrier has the advantage that all the source particles have uniform size and uniform activity, but the specific activity is lower. For short-lived isotopes this is not a serious problem. Typically it requires only one atom of a 5 sec isotope per 100,000 stable atoms to endow a 6 micron sphere with 100 ,uCi of activity.

The Detector The detector system according to the present invention includes one or more conventional gamma-ray detectors capable of determining the coordinates of gamma rays impinging on the plane of the detector, and one or more coded-aperture plates, which can be a pinhole collimator, or a multichannel collimator (parallel, converging or diverging channels), or a plate, as illustrated in Figure 3 or 4, having one or more zones that are substantially opaque to the gamma rays of interest and one or more zones that are substantially transparent to those gamma rays, or the like. A coded aperture plate 21, 31 is placed between the volume of interest in which the source particles are located and a detector 29, 37 and serves to map the location of the source particles onto the detector in a known way.The detector may be a conventional, commercially-available Anger Scintillation Camera which includes a single scintillation crystal that is viewed by a pluraiity of photomultipliers with circuitry to determine center of light intensity, or may be a plurality of scintillation crystals viewed by a plurality of photomultipliers of the type available from Baird Atomic Inc.

Also, the present invention may use such other detectors with improved spatial accuracy as the Germanium camera having a mosaic of Germanium crystals that are read out by charge amplifiers, or pressurized gas or liquid-filled wire chambers in which gamma rays interact in the pressurized gas or liquid and in which the resulting electronic signal on the wires is amp;ified and read out.

Therefore, the apparatus of the present invention detects stenotic regions in coronary arteries by tracking source particles of gamma rays from outside the patient's body. Coded aperture plates associated with gamma-ray detectors permit the coordinates of detected gamma rays to be produced in convenient form suitable for data reduction and analysis. By selecting coordinate data within successive time periods as being indicative of the course of a source particle through a coronary artery of a patient, it is possible to determine the cross-sections of the artery along its length. In the regions therealong where the frequency of detection of source particles per time interval is low, the velocity of blood and source particles flowing therethrough is high, and the higher velocity blood flow is indicative of stenosis in such region.

Claims

1. Apparatus for interacting with a number of particles which can be introduced into the blood of a subject and which can provide signals from within a vessel of interest of the subject that can be detected outside the body, the apparatus comprising: detector means including signal-selection means positionable near the body of the subject for detecting the signals from the number of particles at successive locations of each of the number of particles at successive times during the travel of each particle through a vessel of interest, said signal selection means including spaced regions thereof having different relative transmissiveness therethrough for signal from a particle within the body of the subject;; circuit means coupled to said detector means for providing output manifestations from successive positions of each of the number of particles at successive times during the travel thereof through the vessel of interest; and output means responsive to the output manifestations from said circuit means for providing an indication of the relative internal cross-sectional areas of the vessel of interest at successive positions along the course of the vessel of interest.

2. Apparatus according to claim 1 wherein the signal-selection means is disposed intermediate the detector means and the body of the subject for intercepting signal from a particle therewithin.

3. Apparatus according to either one of the preceding claims wherein the ratio of area of relatively high signal transmissiveness to the area of relatively low signal transmissiveness is within the range from about 1.1 to .45.

4. Apparatus according to any one of the preceding claims wherein said detector means is arranged to produce electrical signal indicative of the location thereon at which a signal is received through the signal selection means.

5. Apparatus according to any one of the preceding claims wherein the signal-selection means includes an element of relatively low signal transmissiveness therethrough having at least one aperture therein of relatively higher signal transmissiveness therethrough.

6. Apparatus according to claim 5 wherein said detector means is arranged to respond to signal received thereby from a particle within the body of the subject and to produce said electrical signals representative of the coordinates of the location thereon at which a signal from the body of the subject is received along a trajectory passing through said aperture.

7. Apparatus according to claim 6 wherein the signal-selection means includes a substantially circular toroidal aperture of relatively high transmissiveness to such signal.

8. Apparatus according to claim 6 wherein the signal-selection means includes a plurality of apertures disposed at spaced locations along orthogonally oriented axes.

9. Apparatus according to claim 6 wherein said detector means is arranged to produce said electrical signals in response to gamma radiation received thereby.

10. Apparatus according to any one of the preceding claims wherein said circuit means includes input means connected to receive electrocardiographic signals from the subject for controlling said circuit means to respond to detection by the detector means during a selected portion of the cardiac cycle of a signal from the body of the subject which is received along a trajectory of said signal that is transmitted through one of said apertures.

11. Apparatus for interacting with a number of particles which can be introduced into the blood of a subject and which can produce radiation from within a vessel of interest that passes through the body walls of the subject from the location of the particle within the body, the apparatus comprising: detector means including signal-selection means having spaced regions of different relative transmissiveness therethrough of the radiation, said detector means being positionable about the body of the subject for receiving through one of said spaced regions the radiation emanating from the body .of the subject to produce electrical signals in response thereto which are indicative of successive locations of each particle of the number of particles at successive times during travel of each particle through the vessel of interest; and circuit means including timing means coupled to said detector means for providing output manifestations indicative of the three-dimensional coordinates of successive positions of each particle of the number of particles which is within the vessel of interest and from the detector means receives the radiation along a trajectory path through said spaced regions during a time interval determined by said timing means.

12. Apparatus according to claim 11 wherein said circuit means includes display apparatus responsive to said outPut manifestations for providing an output indication of sequential locations as a function of time of detections through said spaced regions to provide an indication of the course of the vessel of interest from the successive positions therealong of the detected particles.

13. Apparatus according to claim 1 wherein said circuit means includes data processing means responsive to said output manifestations for producing an indication of the frequencies of occurrence of such output manifestations within increments along at least one coordinate dimension that is representative of increments of length along the vessel of interest.

14. Apparatus according to claim 13 wherein said data processing means is responsive to the output manifestations representative of detected signals passing in virtual straight-line trajectories through the signal selection means for identifying the virtual location of the signal source within the region of convergence of the straight-line trajectories.

15. Apparatus according to claim 14 wherein said circuit means includes display apparatus responsive to the outputs from said signal-selection means for providing an output indication of sequential locations as a function of time of said regions of convergence to provide an indication of the relative internal cross-sectional areas of the vessel of interest at successive positions along the course of the vessel of interest.

16. Apparatus according to claim 14 wherein said data processing means is responsive to the successive locations during each of said time intervals of said regions of convergence representing a particle for indicating the relative cross-sectional areas at various locations along a vessel of interest, thereby to determine the existence, severity and location of stenotic or dilated regions of said blood vessel.