Comparison of noise equivalent count rate and image quality for two-dimensional and three-dimensional PET scans

The aim of this study was to investigate the correlation between noise equivalent count (NEC) rates and the signal-to-noise ratio (S/N) in reconstructed images. The NEC rates were determined using uniform 20 cm and 70 cm tall, 20 cm diameter cylinders filled with 11C. The phantoms were scanned in both two-dimensional and three-dimensional modes. The reconstructed image noise was evaluated using FBP and OSEM algorithms (4 iterations and 8 subsets). The images were filtered to a final image resolution of 6.5 mm. From the reconstructed image sets, averages and standard deviations of images were generated, from which the average image S/N (=average/standard deviation) was calculated within an 18 cm central ROI. The S/N of a central slice and an end slice was compared with the NEC. The NEC was found to have a linear relationship to the image S/N of all slices, depending on differences in noise properties specific to the reconstruction algorithm. In two-dimensional mode, although the image S/N of the central slice and the edge slice showed a linear relationship with the NEC, in three-dimensional mode, the S/N of the central slice did not show a relationship with the NEC. The linear relationship was also found in both two- and three-dimensional acquisition modes, as well as for the different activity distributions. These results indicate that the NEC is not only a measure for comparing the count rate performance of imaging systems. However, an absolute evaluation is impossible to depend on reconstruction algorithm, slice number, and phantom type.     

Evaluation of noise equivalent count parameters as indicators of adult whole-body FDG-PET image quality

Objectives

The aim of this study was to assess variation of qualitative and quantitative PET/CT image quality parameters with acquisition time, injection activity and body mass for a representative group of adults undergoing whole-body PET/CT imaging.

Methods

PET scan data from sixty patients were reconstructed with a scan time of 1, 2 and 3 min/bed position. These images were visually scored and three quantitative parameters were calculated: noise equivalent counts per axial length (NEC_patient), noise equivalent count density (NEC_density) and liver signal to noise ratio (liver SNR). The ability of the three quantitative parameters to discriminate qualitative image quality was assessed using ROC analysis.

Results

The quantitative parameters were shown to discriminate images of good/excellent quality from those of poorer image quality with a high degree of accuracy (ROC area >0.9); further, NEC_patient had significantly higher discrimination than either NEC_density or liver SNR (ROC area = 0.97).

Conclusions

NEC_patient, NEC_density and liver SNR all have high discrimination for qualitatively assessed PET image quality. NEC_patient in particular is an effective objective indicator of patient image quality, which will help to assess and standardise scan protocols for purposes such as multi-centre research trials.     

Optimization of injected dose based on noise equivalent count rates for 2- and 3-dimensional whole-body PET.

The noise equivalent count (NEC) rate index is used to derive guidelines on the optimal injected dose to the patient for 2-dimensional (2D) and 3-dimensional (3D) whole-body PET acquisitions. METHODS: We performed 2D and 3D whole-body acquisitions of an anthropomorphic phantom modeling the conditions for (18)F-FDG PET of the torso and measured the NEC rates for different activity levels for several organs of interest. The correlations between count rates measured from the phantom and those from a series of whole-body patient scans were then analyzed. This analysis allowed validation of our approach and estimation of the injected dose that maximizes NEC rate as a function of patient morphology for both acquisition modes. RESULTS: Variations of the phantom and patient prompt and random coincidence rates as a function of single-photon rates correlated well. On the basis of these correlations, we demonstrated that the patient NEC rate can be predicted for a given single-photon rate. Finally, we determined that patient single-photon rates correlated with the mean dose per weight at acquisition start when normalized by the body mass index. This correlation allows modifying the injected dose as a function of patient body mass index to reach the peak NEC rate in 3D mode. Conversely, we found that the peak NEC rates were never reached in 2D mode within an acceptable range of injected dose. CONCLUSION: The injected dose was adapted to patient morphology for 2D and 3D whole-body acquisitions using the NEC rate as a figure of merit of the statistical quality of the sinogram data. This study is a first step toward a more comprehensive comparison of the image quality obtained using both acquisition modes.     

Evaluation of PET count rate performance

A proposal is made for the test conditions to evaluate PET count rate performance. This performance depends in a complex manner on the spatial distribution of activity and scattering material. Therefore, a combined body phantom is proposed, which is as simple as possible but which adequately simulates the range of clinical application of a whole body tomograph. Taking into account the special properties of the new block detector design, a comprehensive test procedure is developed. This includes not only the common count rate characteristic, but also checks for the accuracy of randoms estimation and count loss correction schemes, and for the occurrence of pulse pile up. This is done for different source and scatter configurations, simulating brain, cardiac, and abdominal imaging, respectively. Examples are given, based on measurements of the latest generation PET scanners, namely the CT1 PT 931/08-12 and the Scanditronix PC 2048-07WB.     

Evaluation of PET count rate performance

A proposal is made for the test conditions to evaluate PET count rate performance. This performance depends in a complex manner on the spatial distribution of activity and scattering material. Therefore, a combined body phantom is proposed, which is as simple as possible but which adequately simulates the range of clinical application of a whole body tomograph. Taking into account the special properties of the new block detector design, a comprehensive test procedure is developed. This includes not only the common count rate characteristic, but also checks for the accuracy of randoms estimation and count loss correction schemes, and for the occurence of pulse pile up. This is done for different source and scatter configurations, simulating brain, cardiac, and abdominal imaging, respectively. Examples are given, based on measurements of the latest generation PET scanners, namely the CTI PT 931/08-12 and the Scanditronix PC 2048-07WB.This article was presented at the 1st EEC workshop on accuracy determination in PET, January 19–20th. 1989 Pisa, Italy (COMAC-BME Concerted Project Characterization and Standardization of PET Instrumentation)     

Evaluation of N parameter in the staging of non-small cell lung cancer: role of CT and PET

PURPOSE: To determine diagnostic accuracy of CT and FDG-PET for the evaluation of N status in non-small cell lung cancer. MATERIALS AND METHODS: Thirty-eight CT scans and PET scans of patients with non small-cell lung cancer were retrospectively reviewed. The data of the noninvasive techniques about N status were compared with the pathology findings obtained by standard lymphadenectomy. RESULTS: The CT results were concordant with surgery in 24 out of 38 cases (63%); in discordant cases CT understaged 8 patients and overstaged 6. The PET images were concordant with surgery in 29 cases (76%); of the remaining 9, PET understaged 5 cases and overstaged 4. Concerning the N parameter, CT had a sensitivity of 42.8% and a specificity of 83.3%, while PET had a sensitivity of 71.4% and a specificity of 91.6%. CONCLUSIONS: In our experience the diagnostic accuracy of PET is superior to that of CT, in agreement with the most important studies in the literature. On only one occasion did PET fail to differentiate between hilar uptake (N1) and the central primary tumour, an area in which CT provided more precise anatomic details. Nonetheless, we believe that PET should be performed in all patients affected by lung cancer, with the only exception of patients shown to be not suitable for surgery after CT examination.     

Evaluation of neoadjuvant therapy response of osteogenic sarcoma using FDG PET.

According to the current treatment protocol of the Cooperative Osteosarcoma Study (COSS), monitoring preoperative chemotherapy response and estimating grade of tumor regression in patients with osteosarcoma is mandatory before surgical removal of the tumor, particularly if a limb salvage procedure is intended. In addition, response to neoadjuvant chemotherapy is considered as an important prognostic indicator. The aim of this prospective study was to assess the usefulness of 2-(18F) fluoro-2-deoxy-D-glucose (FDG) PET in the noninvasive evaluation of neoadjuvant chemotherapy response in osteosarcoma. METHODS: In 27 patients with osteosarcoma, we determined tumor-to-background ratios (TBRs) of FDG uptake with PET, before and after neoadjuvant chemotherapy according to COSS 86c or COSS 96 protocols, respectively. We compared changes in glucose metabolism of osteosarcomas with the histologic grade of regression in the resected specimen, according to Salzer-Kuntschik, discriminating responders (grades I-III; n = 17) and nonresponders (grades IV-VI; n = 10). RESULTS: The decrease of FDG uptake in osteosarcomas expressed as a ratio of posttherapeutic and pretherapeutic TBRs showed a close correlation to the amount of tumor necrosis induced by polychemotherapy (P < 0.001; Spearman). With a TBR ratio cutoff level of 0.6, all responders and 8 of 10 nonresponders could be identified by PET. In addition, lung metastases of osteosarcoma were detected with FDG PET in 4 patients. CONCLUSION: FDG PET provides a promising tool for noninvasive evaluation of neoadjuvant chemotherapy response in osteosarcoma. This could imply consequences for the choice of surgical strategy, because a limb salvage procedure cannot be recommended in patients nonresponsive to preoperative chemotherapy unless wide surgical margins can safely be achieved.     

Reproducibility assessment of uptake on dedicated breast PET for noise discrimination

Annals of Nuclear Medicine - Dedicated breast PET (dbPET) systems have improved the detection of small breast cancers but have increased false-positive diagnoses due to an increased chance of noise...     

PET imaging for response assessment in lymphoma: potential and limitations

Fluorodeoxyglucose positron emission tomography (FDG-PET) is now considered the most accurate tool for the assessment of treatment response and prognosis in patients with Hodgkin lymphoma and aggressive non-Hodgkin lymphoma. This article discusses the potential and limitations of FDG-PET for response assessment in malignant lymphoma during chemotherapy (interim PET) and at the end of chemotherapy. Interim PET is used to predict the likelihood for a complete response at the end of such therapy. End-of-treatment PET aims to establish the completeness of response or the presence of residual viable tumor tissue. Until the results of ongoing clinical trials emerge over the next 5 years, interim PET should be considered investigational and should not be used for patient management outside of study protocols.     

Investigation of noise sources for digital radiography systems

The performance of digital radiography systems can be evaluated in terms of spatial resolution and noise. Noise plays an important role in the achievable image quality for detecting small and low-contrast structures in digital images created by these systems. Our aim in this study was to investigate the noise sources both in the spatial and frequency domain for three digital radiography systems, one digital fluoroscopy system, and one digital mammography system, and to obtain information about the effective operating dose range of these detectors. Noise evaluation in the spatial domain was done with the relative standard deviation–detector air kerma relationship evaluation method. The characterization of the noise in the spatial domain gives information about the types of noise, but does not give information about the noise power distribution in frequency space. Therefore, noise evaluation in the frequency domain was carried out by noise power spectrum measurement. The observed dominant noise component at lower detector doses was electronic noise for the digital mammography system, whereas structured noise was observed to make up nearly half of the total noise at higher detector doses for one of the digital radiography systems. The structured noise component was increased by use of a grid in these systems, independent of the grid ratio and grid frequency, but this increase was lower for higher grid frequencies. Furthermore, the structured noise coefficient was decreased with gain and offset calibrations. The five systems which we evaluated behaved as a quantum noise limited for clinically used detector doses.     

Evaluation of geometric sensitivity for hybrid PET.

Hybrid PET systems have spatially varying sensitivity profiles. These profiles are dependent on imaging parameters, namely, number of heads, head configuration, spacing between gantry stops, radius of rotation (RoR), and coincident head acceptance angle. METHODS: Sensitivity profiles were calculated across a 500-mm field of view (FoV) for a representative set of existing and theoretic 2-, 3-, and 4-head hybrid PET systems. The head configuration was defined by alpha(n), which describes the angular separation between head 1 and head n. Simulated configurations were 2 head ([alpha(2)] = [180 degrees ]), 3 head ([alpha(2), alpha(3)] = [120 degrees, 240 degrees ] and [90 degrees, 180 degrees ]), and 4 head ([alpha(2), alpha(3), alpha(4)] = [90 degrees, 180 degrees, 270 degrees ]). Four transverse acceptance angles, measured from the perpendicular of the crystal to the surface, were simulated: 90 degrees, 45 degrees, 23 degrees, and 11 degrees. Two RoRs were considered: 250 and 300 mm. Each head was rotated through 360 degrees in 128 steps, and no physical collimation was modeled. RESULTS: For a 250-mm RoR and 90 degrees acceptance angle, the sensitivities relative to [alpha(2)] = [180 degrees ] were [alpha(2), alpha(3)] = [120 degrees, 240 degrees ], 183%; [alpha(2), alpha(3)] = [90 degrees, 180 degrees ], 159%; and [alpha(2), alpha(3), alpha(4)] = [90 degrees, 180 degrees, 270 degrees ], 317%. Increasing RoR to 300 mm decreased [alpha(2)] = [180 degrees ] sensitivity by approximately 12%; all other configurations were decreased by approximately 75% of their 250-mm RoR sensitivities. Decreasing the acceptance angle to 45 degrees decreased sensitivities to [alpha(2), alpha(3)] = [120 degrees, 240 degrees ], 100%; [alpha(2), alpha(3)] = [90 degrees, 180 degrees ], 105%; and [alpha(2), alpha(3), alpha(4)] = [90 degrees, 180 degrees, 270 degrees ], 210%. The 2-head [alpha(2)] = [180 degrees ] system sensitivity was not affected. The configuration was the most important factor affecting the shape of the sensitivity profiles. For a 250-mm RoR and 90 degrees acceptance angle, [alpha(2)] = [180 degrees ] concentrated sensitivity in the FoV center, [alpha(2), alpha(3)] = [120 degrees, 240 degrees ] had a slightly increased peripheral sensitivity, and the profiles for both [alpha(2), alpha(3)] = [90 degrees, 180 degrees ] and [alpha(2), alpha(3), alpha(4)] = [90 degrees, 180 degrees, 270 degrees ] were completely flat. CONCLUSION: Sensitivity profiles are affected strongly by imaging parameters; however, profiles can be shaped to concentrate on an annulus or distribute sensitivity uniformly over the FoV. Also, the 4-head system showed a markedly higher sensitivity than either of the 3-head systems.