Low-Light Image Enhancement Using Hybrid Deep-Learning and Mixed-Norm Loss Functions.

This study introduces a low-light image enhancement method using a hybrid deep-learning network and mixed-norm loss functions, in which the network consists of a decomposition-net, illuminance enhance-net, and chroma-net. To consider the correlation between R, G, and B channels, YCbCr channels converted from the RGB channels are used for training and restoration processes. With the luminance, the decomposition-net aims to decouple the reflectance and illuminance and to train the reflectance, leading to a more accurate feature map with noise reduction. The illumination enhance-net connected to the decomposition-net is used to enhance the illumination such that the illuminance is improved with reduced halo artifacts. In addition, the chroma-net is independently used to reduce color distortion. Moreover, a mixed-norm loss function used in the training process of each network is described to increase the stability and remove blurring in the reconstructed image by reflecting the properties of reflectance, illuminance, and chroma. The experimental results demonstrate that the proposed method leads to promising subjective and objective improvements over state-of-the-art deep-learning methods.

Clinical Evaluation of 68Ga-PSMA-II and 68Ga-RM2 PET Images Reconstructed With an Improved Scatter Correction Algorithm.

OBJECTIVE: Gallium-68-labeled radiopharmaceuticals pose a challenge for scatter estimation because their targeted nature can produce high contrast in these regions of the kidneys and bladder. Even small errors in the scatter estimate can result in washout artifacts. Administration of diuretics can reduce these artifacts, but they may result in adverse events. Here, we investigated the ability of algorithmic modifications to mitigate washout artifacts and eliminate the need for diuretics or other interventions. MATERIALS AND METHODS: The model-based scatter algorithm was modified to account for PET/MRI scanner geometry and challenges of non-FDG tracers. Fifty-three clinical 68Ga-RM2 and 68Ga-PSMA-11 whole-body images were reconstructed using the baseline scatter algorithm. For comparison, reconstruction was also processed with modified sampling in the single-scatter estimation and with an offset in the scatter tail-scaling process. None of the patients received furosemide to attempt to decrease the accumulation of radiopharmaceuticals in the bladder. The images were scored independently by three blinded reviewers using the 5-point Likert scale. RESULTS: The scatter algorithm improvements significantly decreased or completely eliminated the washout artifacts. When comparing the baseline and most improved algorithm, the image quality increased and image artifacts were reduced for both 68Ga-RM2 and for 68Ga-PSMA-11 in the kidneys and bladder regions. CONCLUSION: Image reconstruction with the improved scatter correction algorithm mitigated washout artifacts and recovered diagnostic image quality in 68Ga PET, indicating that the use of diuretics may be avoided.

Effects of arm truncation on the appearance of the halo artifact in 68Ga-PSMA-11 (HBED-CC) PET/MRI.

PURPOSE: PSMA ligand imaging with hybrid PET/MRI scanners could be an integral part of the clinical routine in the future. However, the first study about this novel method revealed a severe photopenic artifact ("halo artifact") around the urinary bladder causing significantly reduced tumor visibility. The aim of this evaluation was to analyze the role of arm truncation on the appearance of the halo artifact in 68Ga-PSMA-11 PET/MRI hypothesizing that this influences the appearance. METHODS: Twenty-seven consecutive patients were subjected to 68Ga-PSMA-11 PET/CT (1 h p.i.) followed by PET/MRI (3 h p.i.). PET/MRI was first started with scans of the abdomen to pelvis with arms positioned up above the head. Immediately thereafter, additional scans from the pelvis to abdomen were conducted with arms positioned down beside the trunk. All investigations were first analyzed separately and then compared with respect to tumor detection and tumor uptake (SUV) as well as the presence and intensity of the halo artifact. The Wilcoxon signed rank test was used to determine statistical differences including Bonferroni correction. RESULTS: The halo was significantly reduced if the arms were elevated. Lesions inside the halo artifact (n = 16) demonstrated significantly increased SUVmean (p = 0.0007) and SUVmax (p = 0.0024) with arms positioned up. The halo appearance and intensity was not dependent on the total activity and activity concentration of the urinary bladder. CONCLUSION: Positioning the arms down was shown to be significantly associated with the appearance of the halo artifact in PET/MRI. Positioning the arms up above the head can significantly reduce the halo artifact, thereby detecting more tumor lesions.

Runout error correction in tomographic reconstruction by intensity summation method.

An alignment method for correction of the axial and radial runout errors of the rotation stage in X-ray phase-contrast computed tomography has been developed. Only intensity information was used, without extra hardware or complicated calculation. Notably, the method, as demonstrated herein, can utilize the halo artifact to determine displacement.

Realistic Image Rendition Using a Variable Exponent Functional Model for Retinex.

The goal of realistic image rendition is to recover the acquired image under imperfect illuminant conditions, where non-uniform illumination may degrade image quality with high contrast and low SNR. In this paper, the assumption regarding illumination is modified and a variable exponent functional model for Retinex is proposed to remove non-uniform illumination and reduce halo artifacts. The theoretical derivation is provided and experimental results are presented to illustrate the effectiveness of the proposed model.

Quantitative and Qualitative Assessment of Urinary Activity of 18F-Flotufolastat-PET/CT in Patients with Prostate Cancer: a Post Hoc Analysis of the LIGHTHOUSE and SPOTLIGHT Studies.

PURPOSE: To evaluate the impact of urinary activity on interpretation of 18F-flotufolastat (18F-rhPSMA-7.3) PET/CT, we conducted a post hoc qualitative and quantitative analysis of scans acquired in two phase 3 studies of 18F-flotufolastat. PROCEDURES: Newly diagnosed or recurrent prostate cancer patients enrolled in LIGHTHOUSE (NCT04186819) or SPOTLIGHT (NCT04186845), respectively, underwent PET/CT 50-70 min after intravenous administration of 296 MBq 18F-flotufolastat. For the present analysis, 718 18F-flotufolastat scans (352 from LIGHTHOUSE and 366 from SPOTLIGHT) were re-evaluated by three board-certified nuclear medicine physicians. Reader 1 performed a quantitative assessment (SUVmax and SUVmean) of bladder activity in a circular region-of-interest over the maximum diameter of bladder activity in the transverse plane. All three readers qualitatively assessed the impact of any urinary activity in the bladder on image interpretation using a three-point scale (0 = no/minimal visible urinary activity, 1 = urinary activity visible but distinction between urine and disease possible and 2 = assessment inhibited by urinary activity) and the presence/absence of ureteric activity and halo artifacts. RESULTS: In total, 712/718 scans were evaluable. Reasons for exclusion were cystectomy, renal failure, or urinary catheter in situ (n = 2 each). The median bladder SUVmax and SUVmean were 17.1 and 12.5, respectively. By majority read, 682/712 (96%) patients had either no urinary activity (score = 0) or visible activity that could be distinguished from disease uptake (score = 1). In the minority of patients (24, 3.4%) where urinary activity did impact assessment (score = 2), the median bladder SUVmean was higher (20.5) than those scored 0 (3.8) or 1 (14.0). Ureteric activity was absent in 401 (56%) patients. Halo artifacts were observed in only two (0.3%) patients (majority read). CONCLUSIONS: 18F-Flotufolastat urinary activity did not influence disease assessment for the majority of patients. While this study was not designed as a head-to-head comparison, the median bladder SUVs are lower than previously reported values for other renally cleared PSMA-PET radiopharmaceuticals.

Halo artifacts of indwelling urinary catheter by inaccurate scatter correction in 18F-FDG PET/CT imaging: incidence, mechanism, and solutions.

BACKGROUND: Halo artifacts from urinary catheters can occur due to inaccurate scatter correction, and the artifacts affect the tumor visibility in 18F-FDG PET/CT images. We investigated the incidence rate and the mechanisms of halo-artifact generation and explored several scatter correction techniques to prevent artifacts. METHODS: We conducted patient and phantom studies. (1) We retrospectively reviewed the cases of patients who had undergone 18F-FDG PET/CT scans. To determine the frequency of halo-artifact generation, we used the patients' PET images with a standard scatter correction based on a tail-fitted single-scatter simulation (TF-SSS) using 4-mm voxel µ-maps (TFS 4-mm). (2) We performed phantom studies to evaluate the effects of a urine catheter and two scatter correction techniques, i.e., TF-SSS with 2-mm voxel µ-maps (TFS 2-mm) and a Monte Carlo-based single-scatter simulation (MC-SSS) using 4-mm voxel µ-maps (MCS 4-mm). The average standardized uptake values (SUVs) were measured for axial PET images. (3) Using the patients' data, we investigated whether TFS 2-mm and MCS 4-mm can eliminate the artifacts in the clinical images. RESULTS: (1) There were 61 patients with urinary catheters; in five (8.2%), halo artifacts were observed in the TFS 4-mm PET images. (2) The phantom study clearly reproduced the halo artifacts in the TFS 4-mm PET images. The halo artifacts were generated when urine moved in the interval between the CT and PET imaging, and when the urinary catheter was placed in a circular shape. The SUVs for the TFS 4-mm and TFS-2mm PET images were underestimated at the halo-artifact regions, whereas the SUVs for the MCS 4-mm PET images were close to the true values. (3) The halo artifacts disappeared in the TFS 2-mm PET images in 4/5 patients but not 1/5 patient, whereas the halo artifacts were completely absent in the MCS 4-mm PET images in 5/5 patients. CONCLUSIONS: These data suggest that halo artifacts are caused if the PET images do not correspond to the physical material in the µ-maps, which induces the scatter correction error. With the MC-SSS, it was possible to accurately estimate the scatter without generating halo artifacts.

Improving 68Ga-PSMA PET/MRI of the Prostate with Unrenormalized Absolute Scatter Correction.

A limitation of using 68Ga-labeled prostate-specific membrane antigen (68Ga-PSMA) for detection and staging of prostate cancer is a frequently observed halo artifact around the urinary bladder caused by inaccurate scatter correction (SC) of PET data. The aim of this study was to investigate the impact of unrenormalized absolute SC on 68Ga-PSMA PET quantification in PET/MRI of the prostate in 100 patients. Methods: The PET data of 100 patients were reconstructed twice using standard SC and improved unrenormalized SC. The visual presence of the halo artifact was rated in each PET data reconstruction using 5 grades (0, no halo artifact; 4, severe halo artifact). The number of visible lesions in the pelvis was recorded. SUVmean and SUVmax were measured in the lesions, in the bladder, in the gluteus maximus, and within the halo margin. Furthermore, the signal-to-noise-ratio and image noise were measured in all PET data. Relative differences between standard and unrenormalized SC were calculated. Results: With standard SC, the average grade in the presence of the halo artifact was 2 (moderate halo artifact), whereas for unrenormalized SC, the average grade was 0.9 (slight halo artifact). The same number of congruent lesions (n = 74) was detected for both PET data reconstructions. Relative changes in PET signal-to-noise-ratio and image noise were not statistically significant (P > 0.05). The mean (±SD) increase in SUVmean using unrenormalized SC was 23.0% ± 9.2% in the gluteus maximus, 7.1% ± 4.5% in the bladder, 325.4% ± 748.5% in the halo margin, and 12.4% ± 16.8% in all 74 detected lesions. The mean increase using unrenormalized SC in SUVmean was 17.5% for lesions inside the halo margin (38 lesions) and 6.9% for lesions outside the halo margin (36 lesions). Conclusion: For PET/MRI of prostate cancer using 68Ga-PSMA, a proper SC is important to ensure the best possible diagnostic quality and PET quantification. Unrenormalized absolute SC significantly reduces the halo artifact around the bladder and improves PET/MRI of the prostate.