Paul W. Doherty

A Wiener filter for nuclear medicine images.

To improve the quality of digital nuclear medicine images, we have developed a new implementation of the Wiener restoration filter. The Wiener filter uses as its optimality criterion the minimization of the mean-square error between the undistorted image of the object and the filtered image. In order to form this filter, the object and noise power spectrums are needed. The noise power spectrum for the count-dependent Poisson noise of nuclear medicine images is shown to have a constant average magnitude equal to the total count in the image. The object power spectrum is taken to be the image power spectrum minus the total count, except in the noise dominated region of the image power spectrum where a least-squares-fitted exponential is used. Processing time is kept to a clinically acceptable time frame through use of an array processor. Pronounced noise suppression and detail enhancement are noted with use of this filter with clinical images.

Use of indium-111-labeled OC-125 monoclonal antibody in the detection of ovarian cancer

Abstract This is a preliminary study to evaluate the utility of using the monoclonal antibody (CO-125) labeled with indium-111 to image recurrences of ovarian cancer. This technique has been investigated in 23 patients with ovarian cancer and the results have been compared with blood OC-125 levels, CT scans, and findings at second-look surgery. Following infusion of 1 mg of F(ab′) 2 fragments (1–2 mCi 111 In), quantitative SPECT and planar imaging was obtained daily for 72 hr along with analysis of serum. The nuclear medicine scans of the tumor site recurrences were technically excellent. When compared to second-look laparotomy, there were 2 true negatives, 2 false positives, 14 true positives, and 2 false negatives by nuclear imaging. CT scans correlated less well with surgery, but serum OC-125 levels correlate more closely with nuclear scans and second-look surgery. Those with multiple small metastatic implants showed a pattern of diffuse uptake which increased with time, whereas those with nodal or larger recurrences showed a more focal uptake. The combination of favorable biodistribution and positive images, especially in patients with normal antigen levels and negative CT scans, suggests a role for OC-125 monoclonal antibody imaging in their clinical management. However, further investigation is needed to determine whether nuclear scans can replace second-look surgery. If it can show that enough 111 In-labeled antibody accumulates in the tumor site to justify radioimmunotherapy, then 90 Y (a pure beta emitter) could be exchanged for 111 In. This is potentially a method of radioimmunotherapy for recurrent ovarian carcinoma.

Non‐invasive Evaluation of Ventricular Function and Volumes During Atrioventricular Sequential and Ventricular Pacing

Thirteen patients who all had previously inserted temporary or permanent pacemakers (6, VVI; 7, A‐V sequential) were studied by two‐dimensional echocardiography and radionuclide gated blood pool ventriculography (RVG) for non‐invasive evaluation of cardiac performance. Patients were paced in both the VVI mode and during sinus/atrial or A‐V sequential pacing. Although there was no objective change of the ejection fraction during V‐pacing and atrial/A‐V sequential pacing or sinus rhythm, as has been previously reported, A‐V sequential pacing did result in significant improvement in overall cardiac function and output as judged by radionuclide ventriculography and blood pressure response in most of our patients. An appropriately timed atrial contribution to ventricular systole resulted in improved ventricular function in those individuals with pre‐existing systolic or diastolic myocardiol dysfunction and/or sick sinus syndrome in whom pacemaker therapy was indicated. Radionuclide ventriculography appears to be a reliable, accurate, non‐invasive method that can be used to evaluate patients before implantation of a permanent ventricular or atrioventricular pacemaker in order to decide which pacing mode is best for that particular individual.

The effect of time and cholecystectomy on experimental biliary tree dilatation. A multi-imaging evaluation.

The changes of the biliary tree following distal bile duct obstruction and its release were confirmed by biliary scintigraphy and monitored by serial ultrasonography, computed tomography, and values of serum bilirubin and alkaline phosphatase in 14 mongrel dogs. The degree and rate of biliary dilatation were independent of cholecystectomy. The most rapid rate of extrahepatic dilatation occurred within the first 48 hours, while dilated intrahepatic ducts were first recognized three to six days after obstruction. Following release of the obstruction, the degree and rate of resolution of the biliary dilatation were independent of the duration of ligation (one vs. two weeks) and cholecystectomy. The dilatation resolved slowly. Dilated intrahepatic ducts were recognized for as long as eight to 13 days, while extrahepatic biliary dilatation was still present for 13 weeks, at which time the experiment was terminated. It is postulated that the extrahepatic biliary dilatation will approach a plateau approximately one month after total biliary obstruction. It appears that if the obstruction lasts more than one week, it results in irreversible damage to the elasticity of the extrahepatic ducts. Thus, after release of the obstruction, serial biliary imaging is indicated until a new baseline of the biliary tree diameter has been established.

Modifying constrained least-squares restoration for application to single photon emission computed tomography projection images

Image restoration methods have been shown to increase the contrast of nuclear medicine images by decreasing the effects of scatter and septal penetration. Image restoration can also reduce the high-frequency noise in the image. This study applies constrained least-squares (CLS) restoration to the projection images of single photon emission computed tomography (SPECT). In a previous study, it was noted that CLS restoration has the potential advantage of automatically adapting to the blurred object. This potential is confirmed using planar images. CLS restoration is then modified to improve its performance when applied to SPECT projection image sets. The modification was necessary because the Poisson noise in low count SPECT images causes considerable variation in the CLS filter. On phantom studies, count-dependent Metz restoration was slightly better than the modified CLS restoration method, according to measures of contrast and noise. However, CLS restoration was generally judged as yielding the best results when applied to clinical studies, apparently because of its ability to adapt to the image being restored.

Characterization of indium-111 labeled recombinant tissue plasminogen activator for the imaging of thrombi

The in vitro functional properties of recombinant tissue plasminogen activator (rt-PA), its biodistribution in mice, and its pharmacokinetics and clot localization properties in dogs have been investigated after labeling rt-PA with 111In. The rt-PA was coupled with the bicyclic anhydride of DTPA using standard methodology. Amidolytic and fibrinolytic assays showed retention of protein activity when rt-PA was conjugated with an average of one DTPA group or less per molecule. Size exclusion HPLC showed each preparation to be radiochemically pure with 111In bound exclusively to the attached DTPA groups. Biodistribution in mice showed major accumulation of activity in the liver and kidneys. After administration of 0.5–1.0 mg of the labeled protein to dogs, blood activity decreased with a half time of approximately 5 min in agreement with previous reports of rapid blood clearance. Largely because of decreased blood levels, clot: blood ratios of labeled protein increased rapidly, in one study reaching 6.3 after 31 min, and satisfactory images of fibrin thrombi were obtained. The rt-PA may be labeled with 111In without destroying the ability of the protein to localize in clot and images of forming clot can be obtained with this agent within 1 h after administration.

Exercise testing and hemodynamic performance in healthy elderly persons

To determine the effect of age on cardiovascular performance, 39 healthy elderly men and women, 70 to 83 years old, underwent treadmill thallium-201 exercise perfusion imaging and radionuclide equilibrium angiography at rest and during supine bicycle exercise. Five volunteers who had a positive exercise thallium test response were excluded from the study. Radionuclide left ventricular ejection fraction, regional wall abnormalities, relative cardiac output, stroke volume, end-diastolic volume and end-systolic volume were measured. Seventy-four percent of the subjects maintained or increased their ejection fraction with exercise. With peak exercise, mean end-diastolic volume did not change, end-systolic volume decreased and cardiac output and stroke volume increased. Moreover, in 35% of the subjects, minor regional wall motion abnormalities developed during exercise. There was no significant difference in the response of men and women with regard to these variables. However, more women than men had difficulty performing bicycle ergometry because they had never bicycled before. Subjects who walked daily performed the exercise tests with less anxiety and with a smaller increase in heart rate and systolic blood pressure.

A preliminary pharmacokinetic study of 111In-labeled 260F9 anti-(breast cancer) antibody in patients

SummaryThe pharmacokinetics of 111In-labeled 260F9, a murine monoclonal antibody directed against a breast-cancer-associated antigen, was determined in seven patients with advanced breast cancer. Six patients were administered 1 mg antibody containing 1 mCi 111In. The seventh patient was administered 20 mg unlabeled antibody followed by 1 mg 111In-labeled antibody all via a peripheral vein. Immunoprecipitation, HPLC and SDS-PAGE gels demonstrated the stability of radiolabel on the antibody. The serum clearance of the radiolabel closely fits (r2>0.95) a two-compartment model for the first six patients. The apparent volume of distribution of the radiolabel approximated to the plasma volume (3 1) and its mean residence time was 23.7 h. The radiolabel had an average t1/2β of 22.9±12.21 h at the 1-mg dose. At the 20-mg dose one-compartment elimination kinetics were observed with the radiolabel and antibody showing similar mean residence times (36–41 h) and a t1/2β of 26–28 h. Whole-body imaging showed that the blood-pool:liver ratio of radioactivity increased fourfold (at 48 h postinfusion) at the higher dose and the percentage of the injected dose of radioactivity in the liver decreased from 25% to 8% (24 h postinfusion).In one patient 7–14 times more radioactivity was localized in a breast tumor than in fat (normal breast). Over the first 25 h an average (cumulative) 7.5% of the total dose was excreted in urine. A study of 260F9 in CDF-1 mice demonstrated that the radiolabel remained associated with the antibody in serum. The antibody, however, cleared 60-fold slower in mice than in patients and showed an increased mean residence time of 191 h. The disparity in the pharmacokinetics of the antibody seen in the mouse and in the clinic, points to the different behavior shown by murine monoclonal antibodies in humans. This points to the need for preliminary studies of antibodies in patients for preclinical evaluations of their effectiveness as drug-targeting agents.

A simple in vitro method of screening panels of monoclonal antibodies for tumor binding

We have developed a simple in vitro method of evaluating the relative binding properties of anti-tumor antibodies to human tumor and normal tissues. Cryopreserved surgical explants of tissues as 1 mm cubes are incubated in microtiter plate wells containing media and radiolabeled antibody. We show that the accumulation of antibody in tumor tissue is a specific process which may be reduced by preincubation with saturating levels of unlabeled specific antibody. Evaluation of 7 anti-breast and 4 anti-colorectal tumor antibodies against their respective tumor tissues showed good reproducibility of repeat measurements and up to a 100-fold difference in accumulation among different antibodies to the same tissue. Equivalent results were obtained with the same tissues employed fresh and after cryopreservation. Because of the simplicity of the assay, panels of antibodies may be screened against the large numbers of tumor and normal tissues required to identify superior antibodies for human trials.

A computer program for reconstruction of images from a scanning gamma camera

A program was written to convert the output of a scanning gamma camera into a whole body computer image which can then be used to study regional and whole body radiopharmaceutical localization. The electronically masked output of the camera is collected as a dynamic study on the computer with the time per frame set according to scan speed. The program overlays these frames onto a matrix in memory to form the image. The overlayed frame is incremented 1 pixel/frame in the direction of camera motion during camera scanning. The timing of camera motion is automatically determined by the program. Although written for a specific scanning gamma camera, the ideas inherent in the algorithm will allow the reconstruction of whole body images from other scanning cameras.