An iterative technique to segment PET lesions using a Monte Carlo based mathematical model.

PURPOSE: The need for an accurate lesion segmentation tool in 18FDG PET is a prerequisite for the estimation of lesion response to therapy, for radionuclide dosimetry, and for the application of 18FDG PET to radiotherapy planning. In this work, the authors have developed an iterative method based on a mathematical fit deduced from Monte Carlo simulations to estimate tumor segmentation thresholds. METHODS: The GATE software, a GEANT4 based Monte Carlo tool, was used to model the GE Advance PET scanner geometry. Spheres ranging between 1 and 6 cm in diameters were simulated in a 10 cm high and 11 cm in diameter cylinder. The spheres were filled with water-equivalent density and simulated in both water and lung equivalent background. The simulations were performed with an infinite, 8/1, and 4/1 target-to-background ratio (T/B). A mathematical fit describing the correlation between the lesion volume and the corresponding optimum threshold value was then deduced through analysis of the reconstructed images. An iterative method, based on this mathematical fit, was developed to determine the optimum threshold value. The effects of the lesion volume and T/B on the threshold value were investigated. This method was evaluated experimentally using the NEMA NU2-2001 IEC phantom, the ACNP cardiac phantom, a randomly deformed aluminum can, and a spheroidal shape phantom implemented artificially in the lung, liver, and brain of patient PET images. Clinically, the algorithm was evaluated in six lesions from five patients. Clinical results were compared to CT volumes. RESULTS: This mathematical fit predicts an existing relationship between the PET lesion size and the percent of maximum activity concentration within the target volume (or threshold). It also showed a dependence of the threshold value on the T/B, which could be eliminated by background subtraction. In the phantom studies, the volumes of the segmented PET targets in the PET images were within 10% of the nominal ones. Clinically, the PET target volumes were also within 10% of those measured from CT images. CONCLUSIONS: This iterative algorithm enabled accurately segment PET lesions, independently of their contrast value.

Role of repeated F-18 fluorodeoxyglucose imaging in management of patients with bone and soft tissue sarcoma.

AIM OF THE STUDY: To assess the impact of repeated F-18 FDG studies on the management of patients with bone and soft tissue (B&S) sarcomas. MATERIAL AND METHODS: Twenty patients with B&S tissue tumors (11 M and 9 F age 17-72 years) had 52 F-18 FDG Dual Head Coincidence Imaging (DHCI) studies. 7 patients were followed for 6 months to 2 years clinically after removal of the primary tumor. Thirteen patients were evaluated for suspected recurrences. Patient's preparation, F-18 FDG injection and imaging procedure were done according to department protocol. Attenuation corrected images were interpreted visually by 3 trained physicians. Tumor to background ratios were calculated for all lesions. RESULTS: In 13 patients having both studies, baseline FDG and CT/MRI were concordant in 8 patients, FDG detected more lesions in 3 patients but it did not detect 4 metastatic pulmonary nodules in 2 patients. Follow up studies showed stable disease in 10 patients while 6 patients who showed worsening disease needed to change their chemotherapy. Surgery was avoided in 2 patients and 2 patients showed improved response. CONCLUSION: Repeated F-18 FDG DHCI examinations proved to have an impact on the clinical management of patients with malignant bone and soft tissue sarcoma. It helps to differentiate postoperative changes from local recurrence.

SPECT brain perfusion findings in mild or moderate traumatic brain injury.

BACKGROUND: The purpose of this manuscript is to present the findings in the largest series of SPECT brain perfusion imaging reported to date for mild or moderate traumatic brain injury. PATIENTS AND METHODS: This is a retrospective evaluation of 228 SPECT brain perfusion-imaging studies of patients who suffered mild or moderate traumatic brain injury with or without loss of consciousness (LOC). All patients had no past medical history of previous brain trauma, neurological, or psychiatric diseases, HIV, alcohol or drug abuse. The patient population included 135 males and 93 females. The ages ranged from 11-88 years (mean 40.8). The most common complaints were characteristic of the postconcussion syndrome: headaches 139/228 (61%); dizziness 61/228 (27%); and memory problems 63/228 (28%). LOC status was reported to be positive in 121/228 (53%), negative in 41/228 (18%), and unknown for 63/228 (28%). RESULTS: Normal studies accounted for 52/228 (23%). For abnormal studies (176/228 or 77%) the findings were as follows: basal ganglia hypoperfusion 338 lesions (55.2%); frontal lobe hypoperfusion 146 (23.8%); temporal lobes hypoperfusion 80 (13%); parietal lobes hypoperfusion 20 (3.7%); insular and or occipital lobes hypoperfusion 28 (4.6%). Patients' symptoms correlated with the SPECT brain perfusion findings. The SPECT BPI studies in 122/228 (54%) were done early within 3 months of the date of the accident, and for the remainder, 106/228 (46%) over 3 months and less than 3 years from the date of the injury. In early imaging, 382 lesions were detected; in 92 patients (average 4.2 lesions per study) imaging after 3 months detected 230 lesions: in 84 patients (average 2.7 lesions per study). CONCLUSIONS: Basal ganglia hypoperfusion is the most common abnormality following mild or moderate traumatic brain injury (p = 0.006), and is more common in patients complaining of memory problem (p = 0.0005) and dizziness (p = 0.003). Early imaging can detect more lesions than delayed imaging (p = 0.0011). SPECT brain perfusion abnormalities can occur in the absence of LOC.

SPET brain perfusion imaging in mild traumatic brain injury without loss of consciousness and normal computed tomography.

We present SPET brain perfusion findings in 32 patients who suffered mild traumatic brain injury without loss of consciousness and normal computed tomography. None of the patients had previous traumatic brain injury, CVA, HIV, psychiatric disorders or a history of alcohol or drug abuse. Their ages ranged from 11 to 61 years (mean = 42). The study was performed in 20 patients (62%) within 3 months of the date of injury and in 12 (38%) patients more than 3 months post-injury. Nineteen patients (60%) were involved in a motor vehicle accident, 10 patients (31%) sustained a fall and three patients (9%) received a blow to the head. The most common complaints were headaches in 26 patients (81%), memory deficits in 15 (47%), dizziness in 13 (41%) and sleep disorders in eight (25%). The studies were acquired approximately 2 h after an intravenous injection of 740 MBq (20.0 mCi) of 99Tcm-HMPAO. All images were acquired on a triple-headed gamma camera. The data were displayed on a 10-grade colour scale, with 2-pixel thickness (7.4 mm), and were reviewed blind to the patient's history of symptoms. The cerebellum was used as the reference site (100% maximum value). Any decrease in cerebral perfusion in the cortex or basal ganglia less than 70%, or less than 50% in the medial temporal lobe, compared to the cerebellar reference was considered abnormal. The results show that 13 (41%) had normal studies and 19 (59%) were abnormal (13 studies performed within 3 months of the date of injury and six studies performed more than 3 months post-injury). Analysis of the abnormal studies revealed that 17 showed 48 focal lesions and two showed diffuse supratentorial hypoperfusion (one from each of the early and delayed imaging groups). The 12 abnormal studies performed early had 37 focal lesions and averaged 3.1 lesions per patient, whereas there was a reduction to--an average of 2.2 lesions per patient in the five studies (total 11 lesions) performed more than 3 months post-injury. In the 17 abnormal studies with focal lesions, the following regions were involved in descending frequency: frontal lobes 58%, basal ganglia and thalami 47%, temporal lobes 26% and parietal lobes 16%. We conclude that: (1) SPET brain perfusion imaging is valuable and sensitive for the evaluation of cerebral perfusion changes following mild traumatic brain injury; (2) these changes can occur without loss of consciousness; (3) SPET brain perfusion imaging is more sensitive than computed tomography in detecting brain lesions; and (4) the changes may explain a neurological component of the patient's symptoms in the absence of morphological abnormalities using other imaging modalities.

Fluorine-18 fluorodeoxyglucose splenic uptake from extramedullary hematopoiesis after granulocyte colony-stimulating factor stimulation.

Two patients with sarcoma, one with recurrent osteosarcoma of the spine and the other with metastatic synovial cell sarcoma, were treated with high-dose chemotherapy that produced severe leukopenia. The patients received granulocyte colony-stimulating factor (G-CSF) to stimulate the bone marrow (480 mg given subcutaneously twice daily for 5 to 7 days); their responses were seen as a marked increase in peripheral leukocyte count with no change in the erythrocyte or platelet counts. The patients had fluorine-18 fluorodeoxyglucose (F-18 FDG) imaging 24 hours after the end of G-CSF treatment. Diffusely increased uptake of F-18 FDG was seen in the bone marrow in both patients. In addition, markedly increased uptake in the spleen was noted in both, indicating that the spleen was the site of extramedullary hematopoiesis. The patients had no evidence of splenic metastases. The first patient had a history of irradiation to the dorsal spine, which was less responsive to G-CSF administration than was the nonirradiated lumbar spine.

Comparison of SPET brain perfusion and 18F-FDG brain metabolism in patients with chronic fatigue syndrome.

Chronic fatigue syndrome is a clinically defined condition of uncertain aetiology. We compared 99Tcm-HMPAO single photon emission tomography (SPET) brain perfusion with dual-head 18F-FDG brain metabolism in patients with chronic fatigue syndrome. Eighteen patients (14 females, 4 males), who fulfilled the diagnostic criteria of the Centers for Disease Control for chronic fatigue syndrome, were investigated. Thirteen patients had abnormal SPET brain perfusion scans and five had normal scans. Fifteen patients had normal glucose brain metabolism scans and three had abnormal scans. We conclude that, in chronic fatigue syndrome patients, there is discordance between SPET brain perfusion and 18F-FDG brain uptake. It is possible to have brain perfusion abnormalities without corresponding changes in glucose uptake.

SPECT brain perfusion abnormalities in mild or moderate traumatic brain injury.

The purpose of this atlas is to present a review of the literature showing the advantages of SPECT brain perfusion imaging (BPI) in mild or moderate traumatic brain injury (TBI) over other morphologic imaging modalities such as x-ray CT or MRI. The authors also present the technical recommendations for SPECT brain perfusion currently practiced at their center. For the radiopharmaceutical of choice, a comparison between early and delayed images using Tc-99m HMPAO and Tc-99m ECD showed that Tc-99m HMPAO is more stable in the brain with no washout over time. Therefore, the authors feel that Tc-99m HMPAO is preferable to Tc-99m ECD. Recommendations regarding standardizing intravenous injection, the acquisition, processing parameters, and interpretation of scans using a ten grade color scale, and use of the cerebellum as the reference organ are presented. SPECT images of 228 patients (age range, 11 to 88; mean, 40.8 years) with mild or moderate TBI and no significant medical history that interfered with the results of the SPECT BP were reviewed. The etiology of the trauma was in the following order of frequency: motor vehicle accidents (45%) followed by blow to the head (36%) and a fall (19%). Frequency of the symptoms was headache (60.9%), memory problems (27.6%), dizziness (26.7%), and sleep disorders (8.7%). Comparison between patients imaged early (<3 months) versus those imaged delayed (>3 months) from the time of the accident, showed that early imaging detected more lesions (4.2 abnormal lesions per study compared to 2.7 in those imaged more than 3 months after the accident). Of 41 patients who had mild traumatic injury without loss of consciousness and had normal CT, 28 studies were abnormal. Focal areas of hypoperfusion were seen in 77% (176 patients, 612 lesions) of the group of 228 patients. The sites of abnormalities were in the following order: basal ganglia and thalami, 55.2%, frontal lobes, 23.8%, temporal lobes, 13%, parietal, 3.7%, insular and occipital lobes together, 4.6%.

Acute cholecystitis in AIDS patients: correlation of Tc-99m hepatobiliary scintigraphy with histopathologic laboratory findings and CD4 counts.

BACKGROUND: AIDS patients are susceptible to opportunistic gastrointestinal infections including ascending cholangitis and cholecystitis, especially if CD4 count is < 200. Incidence of acalculous cholecystitis has not been reported previously. PURPOSE: We aim to evaluate the incidence of acalculous cholecystitis in AIDS patients and to identify causative organisms and mortality rate following cholecystectomy. MATERIALS AND METHODS: We reviewed the files of 46 patients in order to meet the objectives of this study. RESULTS: CD4 counts were < 200 in 31 patients and > 200 in 15 patients. HIDA imaging was performed in 31 patients; in 8, the CD4 count was > 200 and all had calculous cholecystitis. The gallbladder was visualized in 3 patients for a sensitivity of 63% and no organisms were found in the gallbladder specimens. In 23 patients, the CD4 count was < 200; the gallbladder was visualized in 5 patients for a HIDA sensitivity of 78%; 16 (52%) had acalculous cholecystitis; and 15 had calculous cholecystitis. In acalculous cholecystitis, Cryptosporidium was found in six cases, cytomegalovirus (CMV) in six cases, and fungus, yeast, tuberculosis, and mycobacterium avium intracellular each in one case. The thirty day mortality rate was 18%; 5 of 28 who underwent open cholecystectomy died within 30 days, 4 of them with a CD4 count < 200. There was no mortality in the 26 patients who underwent laparoscopic cholecystectomy. CONCLUSION AND RECOMMENDATIONS: (1) Because of the high incidence of 52% of acalculous cholecystitis in AIDS patients with a CD4 count < 200, we recommend using intravenous cholecystokinin if the gallbladder is visualized on hepatobiliary scintigraphy in order to determine gallbladder ejection fraction and exclude acalculous cholecystitis. (2) Laparoscopic rather than open cholecystectomy should be the surgical procedure of choice in AIDS patients especially if the CD4 count is < 200.

Cerebral perfusion imaging in asymptomatic carotid artery occlusion following gun shot.

Recognition of brain injury in asymptomatic carotid artery injuries with conventional methods can be difficult. We present a case of angiographically proven asymptomatic left internal carotid artery occlusion with normal CT after a gunshot wound. The SPECT brain perfusion imaging showed mild generalized supratentorial hypoperfusion of the bilateral cerebral cortices on the left side and severe left temporal lobe hypoperfusion.

Fluorine-18-fluorodeoxyglucose dual-head gamma camera coincidence imaging of recurrent colorectal carcinoma.

We report our experience with coincidence detection imaging of 18F-fluorodeoxyglucose (FDG) using a dual-head gamma camera. Scanning of the pelvis and abdomen of a patient with recurrent colorectal carcinoma showed recurrent disease in the pelvic floor and the base of the urinary bladder and metastatic disease in the retroperitoneal space of the pelvis. Although the tumor involving the bladder and pelvic floor was detected by CT and magnetic resonance imaging (MRI), metastatic spread to the retroperitoneal nodes on the left side was detected only by 18F-FDG imaging. Based on the ultrasound, CT, MRI and cystoscopy, a local recurrence of cancer was presumed in our patient. An exploratory laparotomy was performed to resect the tumor in its entirety. At the time of surgery, the retroperitoneal metastasis in the pelvis was confirmed. Had the findings of the coincidence detection imaging study been considered, the patient would have been spared the surgical procedure.