Background/Objectives: Magnetic resonance imaging (MRI) is a cornerstone of diagnostic imaging, providing exceptional soft tissue contrast and enabling detailed characterization of neurological, vascular, and musculoskeletal diseases. Despite its clinical utility, prolonged acquisition times remain a primary limitation of MRI. Extended scan durations increase patient discomfort, particularly in vulnerable populations, which elevate the risk of motion artifacts that can degrade image quality and necessitate repeat imaging. Recent advances in machine learning, especially deep learning (DL), promise to ameliorate these limitations by enabling high-quality MRI reconstruction from heavily undersampled data, increasing throughput capacity and improving healthcare delivery efficiency. This work describes an FDA-cleared DL algorithm known as DeepFoqus-Accelerate for the reconstruction of deliberately under-sampled magnetic resonance imaging data. Methods: Fully sampled k-space from 408 brain MRI volumes (108 from 18 healthy volunteers and 300 from the public fastMRI dataset) was retrospectively undersampled. A detailed comparison of reconstructed images using DeepFoqus-Accelerate with corresponding fully sampled clinical images is presented using expert evaluation of overall image quality and quantitative similarity metrics. The only conventional comparator included in the study was GE’s parallel imaging reconstruction (ARC) at matched retrospective undersampling; comparisons with compressed sensing and vendor-specific deep-learning reconstruction methods were outside the study design. Results: This algorithm can allow for a four-fold reduction in brain imaging times relative to a fully sampled acquisition while preserving image quality. No reader rated any reconstruction below the acceptability threshold, and DeepFoqus-Accelerate exceeded ARC on all metrics at 4×. Conclusions: The findings characterize image quality and structural fidelity relative to fully sampled references and support further evaluation of diagnostic performance, with the potential to improve imaging efficiency and patient experience.
Background/Objectives: Computed tomography pulmonary angiography (CTPA) is the first-line test for suspected acute pulmonary embolism (PE), yet respiratory and cardiac motion artifacts, non-diagnostic opacification, subsegmental interpretive uncertainty, contrast medium and radiation dose burdens, and historically variable reporting persist. Photon-counting detector CT (PCCT), together with the newly released Pulmonary Embolism Reporting and Data System (PE-RADS) v2026, creates the opportunity for a standardized one-stop examination. This technical note consolidates the PCCT-CTPA evidence base into a practical acquisition, reconstruction, and reporting protocol. Methods: The protocol was developed as a narrative, literature-informed synthesis of the PCCT-CTPA evidence base, identified by searching PubMed/MEDLINE, Scopus, and Web of Science from January 2021 (first clinical PCCT installation) to July 2026; sources included randomized and comparative clinical studies, phantom experiments, and a 2026 systematic review and meta-analysis pooling 12 studies and 950 patients. This technical note is not itself a systematic review, and pooled effect sizes are quoted from that existing meta-analysis rather than newly derived. Results: The protocol specifies a sub-second (<1 s), free-breathing, dual-source high-pitch acquisition in QuantumPlus multi-energy mode, 25–40 mL of contrast medium as evidence-based example regimens rather than fixed volumes, routine reading on 50–55 keV virtual monoenergetic images (VMIs) with 40 keV bolus salvage and 90–120 keV streak suppression, iodine distribution (perfused-blood-volume surrogate) maps from the same scan at no additional dose, an optional ultra-high-resolution (UHR) problem-solving tier that replaces, rather than adds to, the standard acquisition in selected patients, and reporting aligned with PE-RADS v2026 categories, designations, and accessory findings. Effective doses of 1.3–1.4 mSv have been reported for low-dose PCCT protocols (including the free-breathing ultra-high-pitch protocol) under specified conditions; 89% of routine PCCT examinations were entirely artifact-free versus 28.6% on energy-integrating detector CT (EID-CT); and iodine maps diagnosed chronic thromboembolism with sensitivity 0.94 and specificity 0.84 at 1.3 mSv. Reported dose and contrast savings are protocol-dependent and did not retain meta-analytic significance after conservative adjustment. Conclusions: PCCT enables an evidence-supported, PE-RADS-ready one-stop CTPA protocol delivering morphology and iodine-distribution (perfused-blood-volume surrogate) maps in every patient. It complements rather than replaces conventional CTPA, and local validation of dose, contrast use, and reporting outcomes is advised. Whether improved peripheral visibility translates into improved diagnostic accuracy for subsegmental PE remains unproven.
Background: Reference morphometric data for the olfactory sulcus (OS) and olfactory bulb (OB) in the pediatric population are scarce. This study aimed to evaluate the age-related morphometric characteristics of these structures using magnetic resonance imaging (MRI). Methods: This retrospective cross-sectional study included pediatric patients with normal brain MRI findings. Olfactory bulb diameters and olfactory sulcus depth were measured. Patients were grouped by age, and correlation, group comparison, and regression analyses were performed to evaluate associations with age. Results: A total of 248 children were enrolled. The mean olfactory bulb transverse and craniocaudal diameters were 2.69 ± 0.67 mm and 2.15 ± 0.65 mm on the right and 2.85 ± 0.71 mm and 2.25 ± 0.63 mm on the left, respectively. The mean olfactory sulcus depth was 10.08 ± 1.76 mm on the right and 9.72 ± 1.65 mm on the left. Sulcus depth showed a weak positive correlation with age (right: r = 0.277, p < 0.001; left: r = 0.259, p < 0.001), whereas bulb measurements demonstrated weaker associations (e.g., right bulb craniocaudal diameter: r = 0.153, p = 0.016). Quadratic regression analyses indicated that age significantly predicted sulcus depth (explaining 27.1% and 25.5% of the variance on the right and left sides, respectively, p < 0.001), whereas bulb measurements remained relatively stable. Conclusion: OB dimensions are relatively stable across childhood, whereas OS depth increases with age in the pediatric population. These data provide reference morphometric parameters to support the radiological evaluation of olfactory system development in children.
Background/Objectives: Centerline Dice (clDice) emphasizes skeleton overlap, but a high score does not establish connectivity. This empirical study, which proposes no new metric, measured how often high-scoring pancreatic-duct segmentations on computed tomography (CT) differ from their references in connected-component count, and what changes that frequency. Methods: Dice, clDice and component counts were compared for 235 held-out PANORAMA CT predictions of a public model. The references were algorithm-generated, so a difference shows disagreement, not anatomical correctness. Thresholds, small-component cleanup and skeleton-based counting were varied. Further tests used a second model, known cuts, synthetic tubes and retinal labels. Synthetic-tube radii illustrate the mechanism, not clinical thresholds. Results: Among 81 predictions with both scores above 0.85, 51 (63.0%; 95% CI 52–73%) differed in component count; a second model gave 32 of 55 (58.2%). Among 86 predictions with Dice above 0.85, cleanup alone moved discordant cases between 25 and 60, and skeleton-based counting reclassified seven cases. After one known cut, 194 of 199 references kept both scores above 0.85 despite the break. Conclusions: High clDice does not ensure matching connectivity. Studies should report a direct connectivity measure alongside clDice and state their cleanup and skeletonization conventions, because these conventions alone changed the reported frequency.
Background: Although nivolumab plus platinum-based chemotherapy has been approved as a first-line systemic treatment for advanced gastric cancer (AGC), the pretreatment imaging and clinical factors associated with disease progression remain incompletely understood. Methods: This retrospective study included 172 patients with AGC who underwent baseline CT and were treated with nivolumab plus chemotherapy. Factors associated with progression-free survival (PFS) were analyzed using univariable and multivariable Cox proportional hazards regression. The performance of the selected variables was evaluated using Harrell’s concordance index (C-index) and the time-dependent area under the curve (AUC). Results: Progression or death occurred in 79.1% of patients (136/172). The number of metastatic sites (hazard ratio [HR], 1.60; 95% confidence interval [CI], 1.28–1.99; p < 0.001), Eastern Cooperative Oncology Group (ECOG) performance status (PS) (HR, 2.02; 95% CI, 1.29–3.16; p = 0.003), macroscopic tumor morphology (HR, 1.69; 95% CI, 1.00–2.84; p = 0.049), and seeding score (HR, 1.16; 95% CI, 1.00–1.36; p = 0.049) were significant prognostic factors for PFS. Together, these four selected variables yielded a C-index of 0.69 (0.64–0.74) and a 12-month time-dependent AUC of 0.73 (0.66–0.81). Median PFS differed significantly between the low- and high-risk groups (9.6 months, 95% CI, 8.0–17.3 vs. 5.2 months, 95% CI, 4.1–6.6; p < 0.001). Conclusions: Combined clinical and CT imaging features, including the number of metastatic sites and ECOG PS, were independent prognostic factors for PFS in patients with AGC receiving first-line nivolumab plus chemotherapy.
