Search Results (184)

This study assesses the clinical deployment of SubtlePET™, a commercial AI-based denoising algorithm, across three radiotracers—¹⁸F-FDG, ⁶⁸Ga-PSMA-11, and ¹⁸F-FDOPA—with the goal of improving image quality while reducing injected activity, technologist radiation exposure, and scan time. A retrospective analysis on a digital PET/CT system showed that SubtlePET™ enabled dose reductions exceeding 33% and time savings of over 25%. AI-enhanced images were rated interpretable in 100% of cases versus 65% for standard low-dose reconstructions. Notably, 85% of AI-enhanced scans received the maximum Likert quality score (5/5), indicating excellent diagnostic confidence and noise suppression, compared to only 50% with conventional reconstruction. The quantitative image quality improved significantly across all tracers, with SNR and CNR gains of 50–70%. Radiotracer dose reductions were particularly substantial in low-BMI patients (up to 41% for FDG), and the technologist exposure decreased for high-exposure roles. The daily patient throughput increased by an average of 4.84 cases. These findings support the robust integration of SubtlePET™ into routine clinical PET practice, offering improved efficiency, safety, and image quality without compromising lesion detectability. Full article

Background: Optimization of pulmonary nodule detection across varied imaging protocols remains challenging. We evaluated four DL-CAD systems and manual reading with volume rendering (VR) for performance under varying radiation doses and reconstruction methods. VR refers to a post-processing technique that generates 3D images by assigning opacity and color to CT voxels based on Hounsfield units. Methods: An anthropomorphic phantom with 169 artificial nodules was scanned at three dose levels using two kernels and three reconstruction algorithms (1080 image sets). Performance metrics included sensitivity, specificity, volume error (AVE), and Lung-RADS classification accuracy. Results: DL-CAD systems demonstrated high sensitivity across dose levels and reconstruction settings, with three fully automatic DL-CAD systems (0.92–0.95) outperforming manual CT readings (0.72), particularly for sub-centimeter nodules. However, DL-CAD systems exhibited limitations in volume measurement and Lung-RADS classification accuracy, especially for part-solid nodules. VR-enhanced manual reading outperformed original CT interpretation in nodule detection, particularly benefiting less-experienced radiologists under suboptimal imaging conditions. Conclusions: These findings underscore the potential of DL-CAD for lung cancer screening and the clinical value of VR in low-dose settings, but they highlight the need for improved classification algorithms. Full article

Background/Objectives: Iterative reconstruction (IR) techniques were developed to address the shortcomings of filtered back projection (FBP), yet research comparing different types of IR is still missing. This work investigates how reducing radiation dose influences both image quality and noise profiles when using two iterative reconstruction techniques—Sinogram-Affirmed Iterative Reconstruction (SAFIRE) and Advanced Modeled Iterative Reconstruction (ADMIRE)—in comparison to filtered back projection (FBP) in non-enhanced head CT (NECT). Methods: In this retrospective single-center study, 21 consecutive patients underwent standard NECT on a 128-slice CT scanner. Raw data simulated dose reductions to 90% and 70% of the original dose via ReconCT software. For each dose level, images were reconstructed with FBP, SAFIRE 3, and ADMIRE 3. Image noise power spectra quantified objective image noise. Two blinded neuroradiologists scored overall image quality, image noise, image contrast, detail, and artifacts on a 10-point Likert scale in a consensus reading. Quantitative Hounsfield unit (HU) measurements were obtained in white and gray matter regions. Statistical analyses included the Wilcoxon signed-rank test, mixed-effects modeling, ANOVA, and post hoc pairwise comparisons with Bonferroni correction. Results: Both iterative reconstructions significantly reduced image noise compared to FBP across all dose levels (p < 0.001). ADMIRE exhibited superior image noise suppression at low (<0.51 1/mm) and high (>1.31 1/mm) spatial frequencies, whereas SAFIRE performed better in the mid-frequency range (0.51–1.31 1/mm). Subjective scores for overall quality, image noise, image contrast, and detail were higher for ADMIRE and SAFIRE versus FBP at the original dose and simulated doses of 90% and 70% (all p < 0.001). ADMIRE outperformed SAFIRE in artifact reduction (p < 0.001), while SAFIRE achieved slightly higher image contrast scores (p < 0.001). Objective HU values remained stable across reconstruction methods, although SAFIRE yielded marginally higher gray and white matter (WM) attenuations (p < 0.01). Conclusions: Both IR techniques—ADMIRE and SAFIRE—achieved substantial noise reduction and improved image quality relative to FBP in non-enhanced head CT at standard and reduced dose levels on the specific CT system and reconstruction strength tested. ADMIRE showed enhanced suppression of low- and high-frequency image noise and fewer artifacts, while SAFIRE preserved image contrast and reduced mid-frequency noise. These findings support the potential of iterative reconstruction to optimize radiation dose in NECT protocols in line with the ALARA principle, although broader validation in multi-vendor, multi-center settings is warranted. Full article

Transcranial direct current stimulation (tDCS) is a non-invasive brain stimulation technique promisingly used to treat neurological and psychological disorders. Nevertheless, the inter-subject heterogeneity in its after-effects frequently limits its efficacy. This can be attributed to fixed-dose methods, which do not consider inter-subject anatomical variations. This work attempts to overcome this constraint by examining the effects of age and anatomical features, including the volume of cerebrospinal fluid (CSF), the thickness of the skull, and the composition of brain tissue, on electric field distribution and cortical excitability. A computational approach was used to map the electric field distribution over the brain tissues of realistic head models reconstructed from MRI images of twenty-three subjects, including adults and children of both genders. Significant negative correlations (p < 0.05) were found in the data between the maximum electric field strength and anatomical variable parameters. Furthermore, this study showed that the percentage of brain tissue exposed to an electric field amplitude above a pre-defined threshold (i.e., 0.227 V/m) was the main factor influencing the responsiveness to tDCS. In the end, the research suggests multiple regression models as useful tool to predict subjects’ responsiveness and to support a personalized approach that tailors the injected current to the morphology of the patient. Full article

Objective: This study proposes a novel deep learning approach for enhancing low-dose bone scintigraphy images using an Enhanced Convolutional Autoencoder (ECAE), aiming to reduce patient radiation exposure while preserving diagnostic quality, as assessed by both expert-based quantitative image metrics and qualitative evaluation. Methods: A supervised learning framework was developed using real-world paired low- and full-dose images from 105 patients. Data were acquired using standard clinical gamma cameras at the Nuclear Medicine Department of the University General Hospital of Alexandroupolis. The ECAE architecture integrates multiscale feature extraction, channel attention mechanisms, and efficient residual blocks to reconstruct high-quality images from low-dose inputs. The model was trained and validated using quantitative metrics—Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM)—alongside qualitative assessments by nuclear medicine experts. Results: The model achieved significant improvements in both PSNR and SSIM across all tested dose levels, particularly between 30% and 70% of the full dose. Expert evaluation confirmed enhanced visibility of anatomical structures, noise reduction, and preservation of diagnostic detail in denoised images. In blinded evaluations, denoised images were preferred over the original full-dose scans in 66% of all cases, and in 61% of cases within the 30–70% dose range. Conclusion: The proposed ECAE model effectively reconstructs high-quality bone scintigraphy images from substantially reduced-dose acquisitions. This approach supports dose reduction in nuclear medicine imaging while maintaining—or even enhancing—diagnostic confidence, offering practical benefits in patient safety, workflow efficiency, and environmental impact. Full article

This study investigates the generation of synthetic CT (sCT) images from zero echo time (ZTE) MRI to support MR-only radiotherapy, which can reduce image registration errors and lower treatment planning costs. Since MRI lacks the electron density data required for accurate dose calculations, generating reliable sCTs is essential. ZTE MRI, offering high bone contrast, was used with two deep learning models: attention deep residual U-Net (ADR-Unet) and derived conditional generative adversarial network (cGAN). Data from 17 head and neck cancer patients were used to train and evaluate the models. ADR-Unet was enhanced with deep residual blocks and attention mechanisms to improve learning and reconstruction quality. Both models were implemented in-house and compared to standard U-Net and Unet++ architectures using image quality metrics, visual inspection, and dosimetric analysis. Volumetric modulated arc therapy (VMAT) planning was performed on both planning CT and generated sCTs. ADR-Unet achieved a mean absolute error of 55.49 HU and a Dice score of 0.86 for bone structures. All the models demonstrated Gamma pass rates above 99.4% and dose deviations within 2–3%, confirming clinical acceptability. These results highlight ADR-Unet and cGAN as promising solutions for accurate sCT generation, enabling effective MR-only radiotherapy. Full article

Coronary CT angiography (CCTA) has seen steady progress since its inception, becoming a key player in the non-invasive assessment of coronary artery disease (CAD). Advancements in CT technology, including iterative and deep-learning-based reconstruction, wide-area detectors, and dual-source systems, have helped mitigate early limitations, such as high radiation doses, motion artifacts, high iodine load, and non-diagnostic image quality. However, the adjustments between ionizing radiation and iodinated contrast material (CM) volumes remain a critical concern, especially due to the increasing use of CCTA in various indications. This review explores the balance between radiation and CM volumes, emphasizing patient-specific protocol optimization to improve diagnostic accuracy while minimizing risks. Radiation dose reduction strategies, such as low tube voltage protocols, prospective ECG-gating, and modern reconstruction algorithms, have significantly decreased radiation exposure, with some studies achieving sub-millisievert doses. Similarly, CM volume optimization, including adjustments in strategies for calculating CM volume, iodine concentration, and flow protocols, plays a role in managing risks such as contrast-associated acute kidney injury, particularly in patients with renal impairment. Emerging technologies, such as photon-counting CT and deep-learning reconstruction, promise further improvements in dose efficiency and image quality. This review summarizes current evidence, highlights the benefits and limitations of dose control approaches, and provides practical recommendations for practitioners. By tailoring protocols to patient characteristics, such as age, renal function, and body habitus, clinicians can achieve an optimal trade-off between diagnostic accuracy and patient safety, ensuring optimal operation of CT systems in clinical practice. Full article

Lung computed tomography (CT) images are widely used to diagnose chronic obstructive pulmonary disease (COPD) by evaluating signs of lung tissue destruction. Accurate diagnosis requires standardizing the CT numbers in lung CT images to distinguish between normal and damaged tissue. The CT number standardization method proposed by Chen-Mayer et al., which uses the linearity of Martinez’s formula, showed promising results in phantom studies. However, our findings reveal that the CT number of water varies significantly, depending on scanning conditions and the characteristics of its container, making it an unreliable reference for lung CT number standardization. To enhance the standardization method, we modified the approach to exclude water and used only solid foams from the COPD gene2 phantom as references. To evaluate the proposed method, we collected 234 CT images of the COPD gene2 phantom from 8 different CT scanners and assessed performance by analyzing CT number standard deviations and variations. The modification resulted in improved reliability and consistency in CT number standardization. Additionally, for a detailed analysis, we segmented the dataset based on CT dose index (CTDI), X-ray tube potential, and reconstruction algorithms to examine the impact of different scanning protocols on standardization performance. Full article

Postmastectomy radiation therapy (PMRT) is an adjuvant treatment for breast cancer. Some mastectomized women undergoing PMRT can have breast reconstruction with expander implant reconstruction. However, the expander implant contains a magnetic metal port for its inflation, and in patients with a high risk of recurrence, the PMRT is performed before the expander replacement. The difficulties in radiation treatment near high-Z metals are mainly due to dose alterations around them. Therefore, this study proposes using a realistic breast phantom and gel dosimetry to investigate the effects of the metallic parts of the expandable prosthesis on the 3D delivery of the treatment. A conformal radiation treatment was planned and delivered to the gel phantom with the metal port. MAGIC-f gel was used with magnetic resonance imaging for dose assessment. The treatment plan dose distribution was compared to the measured dose distribution by gamma analysis (3%/3 mm/15% threshold). A significant gamma fail region was found near the metal port, corresponding to a dose reduction of approximately 5%. This underdose is within the tolerance threshold for dose heterogeneity established by the International Commission on Radiation Units (ICRU), but should be considered when treating these patients. Full article

Objectives: To evaluate the medical image segmentation performance of monochromatic images in various energy levels. Methods: The low-density module (25 mm in diameter, 6 Hounsfield Unit (HU) in density difference from background) from the ACR464 phantom was scanned at both 10 mGy and 5 mGy dose levels. Virtual monoenergetic images (VMIs) at different energy levels of 40, 50, 60, 68, 74, and 100 keV were generated. The images at 10 mGy reconstructed with 50% adaptive statistical iterative reconstruction veo (ASIR-V50%) were used to train an image segmentation model based on U-Net. The evaluation set used 5 mGy VMIs reconstructed with various reconstruction algorithms: FBP, ASIR-V50%, ASIR-V100%, deep learning image reconstruction (DLIR) with low (DLIR-L), medium (DLIR-M), and high (DLIR-H) strength levels. U-Net was employed as a tool to compare algorithm performance. Image noise and segmentation metrics, such as the DICE coefficient, intersection over union (IOU), sensitivity, and Hausdorff distance, were calculated to assess both image quality and segmentation performance. Results: DLIR-M and DLIR-H consistently achieved lower image noise and better segmentation performance, with the highest results observed at 60 keV, and DLIR-H had the lowest image noise across all energy levels. The performance metrics, including IOU, DICE, and sensitivity, were ranked in descending order with energy levels of 60 keV, 68 keV, 50 keV, 74 keV, 40 keV, and 100 keV. Specifically, at 60 keV, the average IOU values for each reconstruction method were 0.60 for FBP, 0.67 for ASIR-V50%, 0.68 for ASIR-V100%, 0.72 for DLIR-L, 0.75 for DLIR-M, and 0.75 for DLIR-H. The average DICE values were 0.75, 0.80, 0.82, 0.83, 0.85, and 0.86. The sensitivity values were 0.93, 0.91, 0.96, 0.95, 0.98, and 0.98. Conclusions: For low-density, non-enhancing objects under a low dose, the 60 keV VMIs performed better in automatic segmentation. DLIR-M and DLIR-H algorithms delivered the best results, whereas DLIR-H provided the lowest image noise and highest sensitivity. Full article

A novel reusable silicon foil dosimeter based on the new emerging optically stimulated luminescence (OSL) material MgB₄O₇:Ce,Li (MBO) is developed and characterized for dosimetric verification of spatially resolved radiotherapy doses. Direct comparison of the spatial (two-2D towards three-3D) proton dose mapping can be achieved with an appropriately designed optical detection setup equipped with a light source (e.g., LEDs) that illuminates the dosimeter and a highly sensitive CCD camera that simultaneously acquires the 2D OSL light from the foil. The newly designed (2nd generation) optical setup allows the registration of high-resolution 2D proton doses (below 0.1 mm resolution) and reconstruction of the 2D proton dose distribution with an accuracy comparable to that of the Gafchromic^TM foils, the current standard of passive 2D dosimetry in radiotherapy. This article outlines the technology’s potential application with respect to the commercially available Gafchromic^TM EBT3 films in measurements of the clinically relevant, spatial proton dose mapping. The obtained comparison of the proton radial dose profiles (for EBT3 films vs MBO foils) agrees within 5%. The resulting image resolution (0.074 mm/px for MBO foil) corresponded well with the tested EBT3 films (0.085 mm/px), indicating excellent properties for future 3D proton dose verifications of modern radiotherapy techniques (e.g., proton radiotherapy). Full article

Background: Spectral detector computed tomography (SDCT) has demonstrated superior diagnostic performance and image quality in liver disease assessment compared with traditional CT. Selecting the right reconstruction algorithm and tube voltage is essential to avoid increased noise and diagnostic errors. Objectives: This study evaluated improvements in image quality achieved using various virtual tube voltages and reconstruction algorithms for diagnosing common liver diseases with spectral CT. Methods: This retrospective study involved forty-seven patients who underwent spectral CT scans for liver conditions, including fatty liver, hemangiomas, and metastatic lesions. The assessment utilized signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), with images reconstructed using various algorithms (IMR, iDose) at different levels and virtual tube voltages. Three experienced radiologists analyzed the reconstructed images to identify the best reconstruction methods and tube voltage combinations for diagnosing these liver pathologies. Results: The signal-to-noise ratio (SNR) was highest for spectral CT images using the IMR3 algorithm in metastatic, hemangioma, and fatty liver cases. A strong positive correlation was found between IMR3 at 120 keV and 70 keV (p-value = 0.000). In contrast, iDOSE2 at 120 keV and 70 keV showed a low correlation of 0.291 (p-value = 0.045). Evaluators noted that IMR1 at 70 keV provided the best visibility for liver lesions (mean = 3.58), while IMR3 at 120 keV had the lowest image quality (mean = 2.65). Conclusions: Improvements in image quality were noted with SDCT, especially in SNR values for liver tissues at low radiation doses and a specific IMR level. The IMR1 algorithm reduced noise, enhancing the visibility of liver lesions for better diagnosis. Full article

Background/Objectives: Accurate patient-specific dosimetry is essential for optimizing radiopharmaceutical therapy (RPT), but current tools lack validation in clinically realistic conditions. This work aimed to develop a workflow for designing and fabricating patient-derived, organ-realistic RPT phantoms and evaluate their feasibility for commissioning patient-specific RPT radioactivity quantification. Methods: We used computed tomographic (CT) and magnetic resonance (MR) imaging of representative patients, computer-aided design, and in-house 3D printing technology to design and fabricate anthropomorphic kidney and parotid phantoms with realistic organ spacing, anatomically correct orientation, and surrounding tissue heterogeneities. We evaluated the fabrication process via geometric verification (i.e., volume comparisons) and leak testing (i.e., dye penetration tests). Clinical feasibility testing involved injecting known radioactivities of ¹⁷⁷Lu-PSMA-617 into the parotid and kidney cortex phantom chambers and acquiring SPECT/CT images. MIM SurePlan MRT SPECTRA Quant software (v7.1.2) reconstructed the acquired SPECT projections into a quantitative SPECT image and we evaluated the accuracy by region-based comparison to the known injected radioactivities and determined recovery coefficients for each organ phantom. Results: Phantom fabrication costs totaled < USD 250 and required <84 h. Geometric verification showed a slight systematic expansion (<10%) from the representative patient anatomy and leak testing confirmed watertightness of fillable chambers. Quantitative SPECT imaging systematically underestimated the injected radioactivity (mean error: −17.0 MBq; −13.2%) with recovery coefficients ranging from 0.82 to 0.93 that were negatively correlated with the surface-area-to-volume ratio. Conclusions: Patient-derived, 3D-printed fillable phantoms are a feasible, cost-effective tool to support commissioning and quality assurance for patient-specific RPT dosimetry. The results of this work will support other centers and clinics implementing patient-specific RPT dosimetry by providing the tools needed to comprehensively evaluate accuracy in clinically relevant geometries. Looking forward, widespread accurate patient-specific RPT dosimetry will improve our understanding of RPT dose response and enable personalized RPT dosing to optimize patient outcomes. Full article

Background/Objectives: To evaluate the detectability of iodine in mediastinal lesions with photon counting CT (PCCT) compared to conventional CT (CCT) in a phantom study. Methods: Mediastinal lesions were simulated by five cylindrical inserts with diameters from 1 to 12 mm within a 10 cm solid water phantom that was placed in the mediastinal area of an anthropomorphic chest phantom with fat ring (QRM-thorax, QRM L-ring, 30 cm × 40 cm cross-section). Inserts were filled with iodine contrast at concentrations of 0.238 to 27.5 mg/mL. A clinical chest protocol at 120 kV on a high-end CCT (Somatom Force, Siemens Healthineers) was compared to the same protocol on a PCCT (Naeotom Alpha, Siemens Healthineers). Images reconstructed with a soft tissue kernel at 1 mm thickness and a 512 matrix served as a reference. For PCCT, we studied the result of reconstructing virtual mono-energetic images (VMIs) at 40, 50, 60 and 70 keV, reducing exposure dose up by 66%, reducing slice thickness to 0.4 and 0.2 mm, and increasing matrix size from 512 to 768 and 1024. Two observers with similar experience independently determined the smallest insert size for which iodine enhancement could still be detected. Consensus was reached when detectability thresholds differed between observers. Results: CTDIvol on PCCT and CCT was 3.80 ± 0.12 and 3.60 ± 0.01 mGy, respectively. PCCT was substantially more sensitive than CCT for detection of iodine in small mediastinal lesions: to detect a 3 mm lesion, 11.2 mg/mL iodine was needed with CCT, while only 1.43 mg/mL was required at 40 keV and 50 keV with PCCT. Moreover, dose reduced by 66% resulted in a comparable detection of iodine between PCCT and CCT for all lesions, except 3 mm. Detection increased from 11.2 mg/mL on CCT to 4.54 mg/mL on PCCT. A matrix size of 1024 reduced this detection threshold further, to 0.238 mg/mL at 40 and 50 keV. For 5 mm lesions, this detection threshold of 0.238 mg/mL was already achieved with a 512 matrix. Very small, 1 mm lesions did not profit from PCCT except if reconstructed with a 1024 matrix, which reduced the detection threshold from 27.5 mg/mL to 11.2 mg/mL. Reduced slice thickness decreased iodine detection of 3–12 mm lesions but not for 1 mm lesions. Conclusions: Iodine detectability with PCCT is at least equal to CCT for simulated mediastinal lesions of 1–12 mm, even at a dose reduction of 66%. Iodine detectability further profits from virtual monoenergetic images of 40 and 50 keV and increased reconstruction matrix. Full article

X-ray computed tomography (CT) is vital for medical diagnostics, but frequent radiation exposure raises concerns, driving the adoption of low-dose CT (LDCT) to mitigate risks. However, LDCT often introduces noise, compromising diagnostic accuracy. This paper proposes a pure vision transformer (PViT) for LDCT denoising, enhanced with a gradient–Laplacian attention module (GLAM) to improve edge preservation and fine structural detail reconstruction. The model’s robustness was validated across five diverse datasets (piglet, head, abdomen, chest, thoracic), demonstrating consistent performance in preserving anatomical structures. Extensive ablation studies on attention configurations and loss functions further substantiated the contributions of each module. Quantitative evaluation using PSNR and SSIM, alongside radiologist assessment, confirmed significant noise suppression and sharper anatomical boundaries, particularly in regions with fine details such as organ interfaces and bone structures. Additionally, in benchmark comparisons against state-of-the-art LDCT models (RED-CNN, TED-Net, DSC-GAN, DRL-EMP) and traditional methods (BM3D), the model exhibited lower parameter and stable training performance. These findings highlight the model’s robustness, efficiency, and clinical applicability, making it a promising solution for improving LDCT image quality while maintaining computational efficiency. Full article