3D Slicer as an image computing platform for the Quantitative Imaging Network

The effect of aspirin on aneurysm wall enhancement: A study in rabbits and humans

ObjectiveAneurysm wall enhancement (AWE) is a potential biomarker of inflammation within the aneurysm wall that has been correlated with a higher risk of rupture. Aspirin (ASA) may decrease AWE due to its anti-inflammatory properties. We aimed to assess the effect of ASA on AWE in an animal model and a cohort of patients with unruptured intracranial aneurysms (UIAs).MethodsThree rabbits with elastase-induced aneurysms were exposed to ASA for 8 weeks and three rabbits were used as controls. 3 T high-resolution magnetic resonance imaging (HR-MRI) was performed at 7 days and 8 weeks to evaluate changes in AWE through histological and immunohistological analyses. Additionally, we evaluated AWE in patients who underwent imaging with a 3 T HR-MRI protocol. ASA exposure was defined as daily intake of 81 mg for at least six months prior to HR-MRI. AWE was quantified using three-dimensional AWE maps and histograms.ResultsAmong rabbits exposed to ASA, the mean AWE was lower at 8 weeks compared to the controls (2.11 vs 2.15, p = 0.13). Immunostaining of the aneurysm wall in rabbits that received ASA revealed a reduced expression of CD68 + or cyclooxygenase-2 + cells, compared to the controls. A total of 99 patients with 120 UIAs were included in the HR-MRI analysis of UIAs. UIAs exposed to ASA (22/120) had significantly lower median AWE than those that were not exposed (0.60 vs 0.72, p = 0.032).ConclusionASA therapy is associated with an objective reduction in AWE, suggesting a potential role in lowering the risk of aneurysm rupture.

Impact of elevated sclerostin levels on bone resorption : unravelling structural changes and mineral metabolism disruption

Aims

The physiological function of sclerostin remains unknown. Sclerostin is synthesized by osteocytes and operates by inhibiting the Wnt/β-catenin pathway. Similarly, it is well established that low levels of sclerostin lead to enhanced bone formation and reduced calciuria, and that high levels of sclerostin are associated with osteoporosis.

Methods

The impact of high levels of recombinant sclerostin on bone and mineral metabolism parameters was analyzed in this study. In male healthy rats, the effects of three elevated doses of sclerostin over a 14-day period were studied, involving bone histomorphometry, micro-CT (μCT), immunohistochemistry, and analysis of mineral metabolism parameters.

Results

Although there was increased bone formation, high doses of sclerostin led to a higher reduction in trabecular bone volume due to a significant increase in bone resorption through the direct activation of osteoclastogenesis. In vitro, sclerostin promoted the differentiation of bone marrow stem cells into osteoclasts. Bone resorption, as measured by tartrate-resistant acid phosphatase (TRAP) activity, was excessive in trabecular, cortical, and subchondral bone. Similarly, high doses of sclerostin increased the number of hypertrophic chondrocytes, consequently expanding the growth plate area. At the cortical level, positive TRAP staining could be observed, suggestive of osteocytic osteolysis and trabecularization of cortical bone. The increased bone resorption resulted in a substantial rise in the urinary excretion of phosphorus and calcium, accompanied by elevated levels of FGF23 and a significant decrease in parathyroid hormone (PTH).

Conclusion

These findings suggest that elevated levels of sclerostin promote bone resorption through the activation of osteoclasts and the generation of osteocytic osteolysis, resulting in increased calciuria, phosphaturia, and changes in mineral metabolism.

Optimization and evaluation of liver deformation modeling under microwave ablation based on ex vivo data

Objective. Microwave ablation (MWA) has emerged as a crucial therapeutic technique for treating hepatocellular carcinoma. Despite its effectiveness, temperature-dependent structural modifications in liver tissues can adversely affect outcomes. Numerical modeling and simulations are essential tools for predicting tissue temperature and deformation prediction. However, existing methods lack comprehensive consideration of deformation-causative factors and fail to validate accuracy throughout the ablation zone.Approach. To overcome these limitations, we analyzed the gap between theex vivoablation deformation results and numerical simulations, and combined them to optimize the physical fields of thermally induced deformation across the entire liver tissue ablation zone. Specifically, we employed a grid marker arrangement with delayed computed tomography (CT) imaging inex vivoexperiments to capture high-resolution global deformation data. The optimization of the simulation was based on updating the coefficient for protein denaturation shrinkage and incorporating vapor diffusion influence in the mechanical model. The effect of vapor diffusion was thoroughly investigated and modeled into the stress-strain equation.Main results. Evaluation results demonstrate that our method significantly improves simulation alignment with observed experimental data, enhancing prediction accuracy of tissue deformation by 30%-90%. Additionally, our model exhibits enhanced capability for expansion representation to describe localized region deformation, resulting in increases of 2.2%-10.0% in dice similarity coefficient (DICE) and 4.2%-19.0% in intersection over union (IoU) when quantifying morphological differences withex vivoexperimental results.Significance. The improved simulation modeling could benefit the planning and optimization of MWA procedures, potentially enhancing treatment efficacy.

Automatic CTA analysis for blood vessels and aneurysm features extraction in EVAR planning

Endovascular Aneurysm Repair (EVAR) is a minimally invasive procedure crucial for treating abdominal aortic aneurysms (AAA), where precise pre-operative planning is essential. Current clinical methods rely on manual measurements, which are time-consuming and prone to errors. Although AI solutions are increasingly being developed to automate aspects of these processes, most existing approaches primarily focus on computing volumes and diameters, falling short of delivering a fully automated pre-operative analysis. This work presents BRAVE (Blood Vessels Recognition and Aneurysms Visualization Enhancement), the first comprehensive AI-driven solution for vascular segmentation and AAA analysis using pre-operative CTA scans. BRAVE offers exhaustive segmentation, identifying both the primary abdominal aorta and secondary vessels, often overlooked by existing methods, providing a complete view of the vascular structure. The pipeline performs advanced volumetric analysis of the aneurysm sac, quantifying thrombotic tissue and calcifications, and automatically identifies the proximal and distal sealing zones, critical for successful EVAR procedures. BRAVE enables fully automated processing, reducing manual intervention and improving clinical workflow efficiency. Trained on a multi-center open-access dataset, it demonstrates generalizability across different CTA protocols and patient populations, ensuring robustness in diverse clinical settings. This solution saves time, ensures precision, and standardizes the process, enhancing vascular surgeons' decision-making.

Bi-VesTreeFormer: A bidirectional topology-aware transformer framework for coronary vFFR estimation

Fractional Flow Reserve (FFR) serves as the gold standard for evaluating the functional significance of coronary artery stenosis. However, traditional FFR involves the injection of vasodilator drugs and the utilization of additional guidewires, which consequently can lead to patient risks and increased costs. Computational fluid dynamics-based approaches can enable non-invasive virtual FFR (vFFR) estimation, but they are computationally intensive and time-consuming. Although deep learning can remarkably enhance computational efficiency, the existing vFFR methods rely heavily on manually crafted features and face difficulties in capturing long-distance dependencies within the vessel structure. In this study, we propose a novel framework for estimating coronary vFFR, which circumvents the laborious preprocessing procedures of previous methods. Specifically, a novel bidirectional topology-aware transformer network (Bi-VesTreeFormer) is proposed to conduct fully automated topological stenotic feature extraction of the vessel tree and capture the global dependencies among branches. Additionally, a contextual vFFR decoder is introduced to establish the correlation of FFR values between adjacent branches and achieve a stable mapping of FFR values to the latent vector space. To validate and train our method, we gathered FFR data from 43 patients with coronary artery stenosis and simulated 15,000 coronary artery centerline data with a reduced-order hemodynamic model. The results show that the proposed method attains root mean square errors of 0.038 and 0.048 for simulated and real data respectively, surpassing the state-of-the-art methods.

Real-Time Intraoperative Decision-Making in Head and Neck Tumor Surgery: A Histopathologically Grounded Hyperspectral Imaging and Deep Learning Approach

BACKGROUND

Accurate and rapid intraoperative tumor margin assessment remains a major challenge in surgical oncology. Current gold-standard methods, such as frozen section histology, are time-consuming, operator-dependent, and prone to misclassification, which limits their clinical utility.

OBJECTIVE

To develop and evaluate a novel hyperspectral imaging (HSI) workflow that integrates deep learning with three-dimensional (3D) tumor modeling for real-time, label-free tumor margin delineation in head and neck squamous cell carcinoma (HNSCC).

METHODS

Freshly resected HNSCC samples were snap-frozen and imaged ex vivo from multiple perspectives using a standardized HSI protocol, resulting in a 3D model derived from HSI. Each sample was serially sectioned, stained, and annotated by pathologists to create high-resolution 3D histological reconstructions. The volumetric histological models were co-registered with the HSI data (n = 712 Datacubes), enabling voxel-wise projection of tumor segmentation maps from the HSI-derived 3D model onto the corresponding histological ground truth. Three deep learning models were trained and validated on these datasets to differentiate tumor from non-tumor regions with high spatial precision.

RESULTS

This work demonstrates strong potential for the proposed HSI system, with an overall classification accuracy of 0.98 and a tumor sensitivity of 0.93, underscoring the system's ability to reliably detect tumor regions and showing high concordance with histopathological findings.

CONCLUSION

The integration of HSI with deep learning and 3D tumor modeling offers a promising approach for precise, real-time intraoperative tumor margin assessment in HNSCC. This novel workflow has the potential to improve surgical precision and patient outcomes by providing rapid, label-free tissue differentiation.

Optimization of the size and location of the FOVs for CBCT capture of the TMJ

BACKGROUND

Osseous pathologies of the temporomandibular joint (TMJ) such as degenerative joint disease, trauma, and deformity contribute to orofacial morbidity and are considered a major factor in temporomandibular dysfunction. Cone beam computed tomography (CBCT) is a recommended diagnostic tool in imaging of osseous tissue pathologies. However, CBCT contributes to patient radiation exposure, and limiting the CBCT field of view (FOV) may reduce it. This study aims to investigate the possibility and clinical applicability of optimizing the size and location of the FOVs for CBCT capture of the TMJ.

METHODS

Three-dimensional CBCT data sets in which the bilateral positions and dimensions of the TMJs were analyzed. A total of 201 data sets with 402 condyles were mapped in relation to the CBCT device. By transformation into a common coordinate space using the device's chin rest as a joint denominator, we were able to determine the optimal size and location for uni- and bilateral capture of the TMJ for both best-case and worst-case scenarios with regard to patient positioning.

RESULTS

The minimal FOVs for unilateral capture were H 28.2 mm × R 22.9 mm in the best-case scenario assuming optimal patient positioning and H 47.0 mm × R 28.3 mm in the worst-case scenario with rotational deviation along the transversal axis. For bilateral capture, we determined the best-case FOV as H 24.9 mm × R 66.5 mm and the worst-case FOV as H 42.8 mm × R 66.7 mm.

DISCUSSION

This research yields indication-specific FOVs for both uni- and bilateral imaging of the TMJ. Considering the best clinical practices for CBCT imaging, clinically feasible FOV dimensions in consideration of the technical specifications of common CBCT devices can be suggested. The clinical application of the results may help reducing radiation exposure of patients receiving CBCT imaging of the TMJ. The transferability of the present results to other CBCT devices requires further research.

TRIAL REGISTRATION

The study is registered in the German Trial Register with the number DRKS00026149, 2024/02/21.

Dosimetric performance of cone beam CT-guided adaptive carbon-ion radiotherapy with daily replanning for pancreatic cancer

PURPOSE

We investigated whether an ultra-hypofractionated carbon-ion radiotherapy (CIRT) protocol for pancreatic cancer (PC) could produce satisfactory dosimetric results with or without cone-beam CT-guided adaptive replanning and explored the potential dosimetric advantages of the adapted protocol.

METHODS

Eleven PC patients who underwent CBCT-guided online adaptive photon radiotherapy were selected. Data were imported into a CIRT treatment planning software to develop new plans for an ultra-hypofractionated CIRT protocol. Prescriptions and constraints were recalculated for a five-fraction schedule using a linear quadratic model for organs-at-risk (OARs) and targets, respectively. The biologically effective dose-equivalent prescribed dose was set at 43.2 Gy (relative biological effectiveness [RBE]). Each day, a synthetic CT (SCT) was generated from the planning CT (PCT) with the daily CBCT. A reference plan based on the PCT was compared to an adapted plan based on the SCT. Deformable image registration was used to allow summation of the daily doses.

RESULTS

The adapted plans met the clinical goals, whereas the reference plans exceeded the constraints in 27 % (stomach), 53 % (duodenum), and 31 % (small bowel) of the fractions. The adapted plans notably decreased V35.5 Gy[RBE] for all gastrointestinal OARs, while significantly enhancing the gross tumor volume (GTV) D95% and planning target volume (PTV) D90%. The accumulated doses showed significant improvements in the duodenum V35.5 Gy[RBE], GTV D95%, and PTV D90%.

CONCLUSION

CBCT-guided adaptive CIRT for PC demonstrated favorable dosimetric results, notably enhancing the sparing of OARs and ensuring superior target coverage compared with non-adaptive CIRT protocols.

Dosimetric optimization and evaluation of hepatocellular carcinoma treatment effect prediction in Y-90 radioembolization

BACKGROUND

Liver transarterial radioembolization (TARE) with 90Y microspheres is a common treatment for hepatocellular carcinoma (HCC). A pre-treatment SPECT/CT dosimetric study is performed using 99mTc macroaggregated albumin, followed by PET-based dosimetry to assess dose distribution of 90Y. Recent studies show a significant correlation between absorbed doses and treatment outcomes in terms of radiological response, adverse events and overall survival. This study aims to present optimized TARE dosimetry protocols and assess outcome predictions according to voxel-based dosimetry.

METHODS

Dosimetry protocols were refined according to EANM guidelines using the Planet Dose software. Pre-treatment dosimetry was conducted for all patients while post-treatment dosimetry was performed for patients treated after the installation of a new PET/CT system. Statistical analysis was employed to evaluate predictors of complete radiological response (CR) and survival outcomes.

RESULTS

133 HCC patients treated with 90Y microspheres (95 resin, 38 glass) at single institution were analyzed. ROC curve analysis for resin microspheres indicated a dose threshold of 233.2 Gy (AUC = 0.62) as the best predictor for CR, with higher CR rates in patients receiving this dose. No lung toxicity was noted nor correlation was found between doses to normal liver tissue and adverse events.

CONCLUSIONS

Lesion absorbed dose is a significant predictor of CR in resin microspheres, with a mean dose of at least 233.2 Gy leading to better oncologic response rates. The absence of correlation between healthy liver tissue dose and adverse events suggests the potential for further increasing the dose to achieve event better outcomes in future protocols.

Quantitative Total-Body Imaging of Blood Flow with High-Temporal-Resolution Early Dynamic 18F-FDG PET Kinetic Modeling

Past efforts to measure blood flow with the widely available radiotracer 18F-FDG were limited to tissues with high 18F-FDG extraction fraction. In this study, we developed an early dynamic 18F-FDG PET method with high-temporal-resolution (HTR) kinetic modeling to assess total-body blood flow based on deriving the vascular phase of 18F-FDG transit and conducted a pilot comparison study against a 11C-butanol flow-tracer reference. Methods: The first 2 min of dynamic PET scans were reconstructed at HTR (60 × 1 s/frame, 30 × 2 s/frame) to resolve the rapid passage of the radiotracer through blood vessels. In contrast to existing methods that use blood-to-tissue transport rate as a surrogate of blood flow, our method directly estimated blood flow using a distributed kinetic model (adiabatic approximation to tissue homogeneity [AATH] model). To validate our 18F-FDG measurements of blood flow against a reference flow-specific radiotracer, we analyzed total-body dynamic PET images of 6 human participants scanned with both 18F-FDG and 11C-butanol. An additional 34 total-body dynamic 18F-FDG PET images of healthy participants were analyzed for comparison against published blood-flow ranges. Regional blood flow was estimated across the body, and total-body parametric imaging of blood flow was conducted for visual assessment. AATH and standard compartment model fitting was compared using the Akaike information criterion at different temporal resolutions. Results: 18F-FDG blood flow was in quantitative agreement with flow measured from 11C-butanol across same-subject regional measurements (Pearson correlation coefficient, 0.955; P < 0.001; linear regression slope and intercept, 0.973 and -0.012, respectively), which was visually corroborated by total-body blood-flow parametric imaging. Our method resolved a wide range of blood-flow values across the body in broad agreement with published ranges (e.g., healthy cohort values of 0.51 ± 0.12 mL/min/cm3 in the cerebral cortex and 2.03 ± 0.64 mL/min/cm3 in the lungs). HTR (1-2 s/frame) was required for AATH modeling. Conclusion: Total-body blood-flow imaging was feasible using early dynamic 18F-FDG PET with HTR kinetic modeling. This method may be combined with standard 18F-FDG PET methods to enable efficient single-tracer multiparametric flow-metabolism imaging, with numerous research and clinical applications in oncology, cardiovascular disease, pain medicine, and neuroscience.

Using baseline MRI radiomic features to predict the efficacy of repetitive transcranial magnetic stimulation in Alzheimer's patients

The efficacy of repetitive transcranial magnetic stimulation (rTMS) as a treatment for Alzheimer's disease (AD) is uncertain at baseline. Herein, we aimed to investigate whether radiomic features from the pre-treatment MRI data could predict rTMS efficacy for AD treatment. Out of 110 participants with AD in the active (n = 75) and sham (n = 35) rTMS treatment groups having T1-weighted brain MRI data, we had two groups of responders (active = 55 and sham = 24) and non-responders (active = 20 and sham = 11). We extracted histogram-based radiomic features from MRI data using 3D Slicer software; the most important features were selected utilizing a combination of a two-sample t-test, correlation test, least absolute shrinkage, and selection operator. The support vector machine classified rTMS responders and non-responders with a cross-validated mean accuracy/AUC of 81.9%/90.0% in the active group and 87.4%/95.8% in the sham group. Further, the radiomic features of the active group significantly correlated with participants' AD assessment scale-cognitive subscale (ADAS-Cog) change after treatment (false discovery rate corrected p < 0.05). Given that baseline radiomic features were able to accurately predict AD patients' responses to rTMS treatment, these radiomic features warrant further investigation for personalizing AD therapeutic strategies.

Brain and cerebrospinal fluid 3D center of mass shift after spaceflight

A subset of long-duration spaceflight astronauts at the International Space Station has been documented to develop spaceflight associated neuro-ocular syndrome (SANS). Researchers have sought to understand SANS by quantification of ocular and brain structural changes thought to be associated with weightlessness induced headward fluid shift. Brain tissue shift and cerebrospinal fluid (CSF) redistribution has been observed as measured by MRI on return to Earth, and not fully quantified. To improve the understanding of this phenomenon, we developed and applied automated methods to quantify 3D center of mass shift within the skull of the extra-axial cerebrospinal fluid (eaCSF) and brain after long-duration spaceflight in astronauts (N = 13) and controls not exposed to microgravity (N = 10). 3D center of mass shift of brain tissue and CSF was computed based on registration of an individual skull segmentation at a baseline timepoint versus follow-up. 3D center of mass shift was quantified in the Gx, Gy, and Gz axis defined as -posterior/+anterior, -left/+right, -inferior/+superior, respectively. For astronauts, average MRI follow-up time pre- to post-flight was 697 ± 137 days (average flight duration = 179 ± 59 days with post-flight MRIs collected an average of 2.23 ± 1.64 days after return to Earth). For controls, average MRI follow-up time was 307 ± 19 days. For astronauts, a superior Gz shift in whole brain was present (+ 0.74 ± 0.28 mm, p < 0.0001) with a concomitant inferior Gz shift in eaCSF (-2.45 ± 0.99 mm, p < 0.0001). In the control cohort, brain tissue Gz shift (-0.082 ± 0.048 mm) and eaCSF Gz shift (0.096 ± 0.26 mm) were not statistically significant. Gy shift lacked significance in both controls and astronauts. These findings support that sustained exposure to weightlessness impacts the overall position of fluids and tissues within the skull.

The role of the sacroiliac joint in sex estimation: Analysis of morphometry and variation types using machine learning techniques

This study aimed to evaluate the potential of machine learning algorithms in sex estimation by going beyond the traditional two-dimensional (2D) measurements of the pelvic bone, predominantly preferred in sex prediction, by including measurement data in three-dimensional (3D) images. Measurements were performed on abdominal multidetector computed tomography (MDCT) images of 152 individuals (77 females, 75 males) aged 18-85. 3D-Slicer software was used for measurements on 2D and 3D images. In 2D images, sacroiliac joint surface area, the angle of the joint surface, the distance between the right and left joints, joint space measurements, and joint variation typing were performed. The distances from the sacroiliac joint to the apex of the contralateral linea terminalis and from the joint to the superior and inferior pubic symphysis were measured on 3D images. It was confirmed that sacroiliac joint space measurements were significantly higher in males than in females. Among the sacroiliac joint variations, 46% in males and 28% in females were the most common standard joint. A strong positive correlation was found between sacroiliac joint distance and the distance of the sacroiliac joint to the contralateral linea terminalis apex and the sacroiliac joint to the superior and inferior pubic symphysis. In this study, the support vector machine algorithm gave the most successful result compared to other algorithms, reaching 88% accuracy in sex estimation. We hope our study will guide the use of artificial intelligence on 3D images for forensic identification, especially in sex estimation.

Catheter-to-tissue contact angle's effect on lesion formation and characterisation using multichannel bioimpedance method

Objective.Radiofrequency (RF) catheter ablation is a standard treatment for patients with cardiac arrhythmias, providing an efficient, minimally invasive solution. However, the ablation efficiency remains suboptimal due to numerous contributed factors that are overlooked in the literature and not monitored during the procedure. This paper explores the effect of catheter-to-tissue contact angles on lesion formations and the feasibility of the multichannel bioimpedance method in characterising the angles to inform cardiologists.Approach.Two silico simulations based on a realistic human model were built to: (1) simulate lesion formations with different catheter-to-tissue angles under varying conditions of powers and convection cooling, and (2) simulate multichannel bioimpedances measured at each catheter's location and angle. 13 locations were picked in all four chambers with 3 contact conditions (catheter lies along the muscle (0° and 180°), in perpendicular to the muscle (90°) and in middle angles (45° and 135°)). 64 electrodes divided into 4 bands were placed on the thorax for multichannel bioimpedances (3-terminal) measured between the catheter's second electrode E2 (I+,V+), and each pair of adjacent surface electrodes (I-,V-). ANOVA and Tukey's Honestly Significant Difference (HSD) tests were used to evaluate the contact angle's effect on the lesion formations and the bioimpedance's capability in distinguishing between angles.Main results.The results showed that 0° and 180° configurations generated significantly different lesions from other angles. The multichannel bioimpedances could recognise 0°/180° from other angles and correlated moderately to lesion sizes at low ablation power.Significance.This paper concludes that catheter-to-tissue angles can influence the lesion outcomes significantly and the multichannel bioimpedance is able to detect the angles that matter.

Semicircular canal morphology in Rodentia and its relationship to locomotion

Anatomical structures vary among mammals with different locomotor behaviours, including sensory structures such as the semicircular canals (SCCs) in the inner ear. Recent SCC research has examined various mammalian groups, but there has been a lack of research on rodents, the most speciose and diverse mammalian order. In this study, an extant sample of 98 rodent SCCs from 56 species across seven different locomotor behaviour categories (arboreal, fossorial, gliding, ricochetal, semiaquatic, semifossorial, terrestrial) was used to understand the correlations between SCC morphology and locomotion in rodents. Morphological correlates considered include the radius of curvature (R), overall 3-dimensional shape, and angles between pairs of canals (orthogonality). Our results show that agile arboreal taxa have larger R for their body size, and fossorial taxa have smaller R for their body size. Shape among specialized locomotor behaviours (arboreal, gliding vs. fossorial) can be differentiated, while other "generalist" categories overlap in morphospace. Specialized locomotor categories can be predicted with greater precision and sensitivity, while other generalist categories tend to be miscategorized as terrestrial. Angles between canals are not consistent across locomotor categories, and more agile groups do not have more orthogonal angles, contrary to our predictions. SCC R and overall shape are robust indicators of specialized locomotor behaviours and can be informative in reconstructing the behaviour of fossil rodents.

Artificial intelligence-based pulmonary vessel segmentation: an opportunity for automated three-dimensional planning of lung segmentectomy

OBJECTIVES

This study aimed to develop an automated method for pulmonary artery and vein segmentation in both left and right lungs from computed tomography (CT) images using artificial intelligence (AI). The segmentations were evaluated using PulmoSR software, which provides 3D visualizations of patient-specific anatomy, potentially enhancing a surgeon's understanding of the lung structure.

METHODS

A dataset of 125 CT scans from lung segmentectomy patients at Erasmus MC was used. Manual annotations for pulmonary arteries and veins were created with 3D Slicer. nnU-Net models were trained for both lungs, assessed using Dice score, sensitivity and specificity. Intraoperative recordings demonstrated clinical applicability. A paired t-test evaluated statistical significance of the differences between automatic and manual segmentations.

RESULTS

The nnU-Net model, trained at full 3D resolution, achieved a mean Dice score between 0.91 and 0.92. The mean sensitivity and specificity were: left artery: 0.86 and 0.99, right artery: 0.84 and 0.99, left vein: 0.85 and 0.99, right vein: 0.85 and 0.99. The automatic method reduced segmentation time from ∼1.5 hours to under 5 minutes. Five cases were evaluated to demonstrate how the segmentations support lung segmentectomy procedures. P-values for Dice scores were all below 0.01, indicating statistical significance.

CONCLUSIONS

The nnU-Net models successfully performed automatic segmentation of pulmonary arteries and veins in both lungs. When integrated with visualization tools, these automatic segmentations can enhance preoperative and intraoperative planning by providing detailed 3D views of patients anatomy.

Feasibility of individual dosimetry using RT-PHITS for patients with SPECT/CT imaging after 177Lu-DOTATATE peptide receptor radionuclide therapy

A radiotherapy package based on the Particle and Heavy Ion Transport code System (RT-PHITS) can calculate internal 3-dimensional dose distribution from SPECT/CT images of individual patients coupled with Monte Carlo radiation transport simulation. This study aims to determine the feasibility of individual dosimetry using RT-PHITS for patients after 177Lu-DOTATATE peptide receptor radionuclide therapy (PRRT). We acquired SPECT/CT images from two patients from the 177Lu SNMMI Dosimetry Challenge (patients A and B) and one from our institute (patient C) at various time points. The images were converted to source/geometry information in the PHITS input format using RT-PHITS. The 3D dose-rate distribution in each patient was calculated using Monte Carlo radiation transport simulation. The output data from the PHITS simulation were converted to DICOM RT-dose format and analyzed using 3D Slicer to identify dose rates in lesions and organs at risk. The time variations of the calculated dose rates were linearly interpolated, considering the physical decay constant. The absorbed dose was evaluated as the integration of the time variation of the dose rate. Agreements between the absorbed doses obtained from RT-PHITS and 177Lu Dosimetry Challenge Task 4 were generally satisfactory (< 20%), although discrepancies were noted in some normal organs of patient A. This was likely due to difficulties in estimating the tail of the dose rate curve after the last imaging time point. The EQD2 was slightly, but not significantly increased compared with the absorbed dose in patient C. Individual dosimetry using RT-PHITS is feasible for assessing the effects of 177Lu-DOTATATE PRRT.

Cell expansion for notochord mechanics and endochondral bone lengthening in zebrafish depends on the 5'-inositol phosphatase Inppl1a

Cell size is a key contributor to tissue morphogenesis. As a notable example, growth plate hypertrophic chondrocytes use cellular biogenesis and disproportionate fluid uptake to expand 10 to 20 times in size to drive lengthening of endochondral bone. Similarly, notochord vacuolated cells expand to one of the largest cell types in the developing embryo to drive axial extension. In zebrafish, the notochord vacuolated cells undergo vacuole fusion to form a single large, fluid-filled vacuole that fills the cytoplasmic space and contributes to vacuolated cell expansion. When this process goes awry, the notochord lacks sufficient hydrostatic pressure to support vertebral bone deposition, resulting in adult spines with misshapen vertebral bones and scoliosis. However, it remains unclear whether endochondral bone and the notochord share common genetic and cellular mechanisms for regulating cell and tissue expansion. Here, we demonstrate that the 5'-inositol phosphatase gene, inppl1a, regulates notochord expansion independent of vacuole fusion, thereby genetically decoupling these processes. We demonstrate that inppl1a-dependent vacuolated cell expansion is essential to establish normal mechanical properties of the notochord and to facilitate the development of a straight spine. Finally, we find that inppl1a is also important for hypertrophic chondrocyte differentiation and endochondral bone lengthening in fish, as has been shown in the human INPPL1-related endochondral bone disorder, opsismodysplasia. Overall, this work reveals a shared mechanism of cell size regulation that influences disparate tissues critical for skeletal development and short-stature disorders.

Accuracy of an nnUNet Neural Network for the Automatic Segmentation of Intracranial Aneurysms, Their Parent Vessels, and Major Cerebral Arteries from MRI-TOF

BACKGROUND AND PURPOSE

The automatic recognition of intracraial aneurysms by means of machine-learning algorithms represents a new frontier for diagnostic and therapeutic goals. Yet, the current algorithms focus solely on the aneurysms and not on the recognition of their parent vessels. The purpose of the present study is the development of a new machine-learning algorithm for fully automatic identification of cerebral arteries and intracranial aneurysms (IAs) based on a manually segmented MRA-TOF data set.

MATERIALS AND METHODS

In this retrospective single-center study, 62 MRA-TOF scans of a total of 73 untreated, unruptured IAs were manually color-labeled in 21 classes. A nnUNet architecture was trained on MRA-TOF images. The performance of the automatic segmentation was compared with the manual segmentation by using the Dice Similarity Coefficient (DSC), Centerline Dice (ClDice), and 95th percentile Hausdorff Distance (HD95). Sensitivity was computed for aneurysm detection.

RESULTS

Across all 21 classes, the median DSC was 0.86 [95% CI: 0.81-0.89], the median ClDice was 0.91 [0.85, 0.94], and the median HD95 was 2.9 [1.0, 14.9] mm. Sensitivity of the model for aneurysm detection was 0.8. For this class specifically, a median DSC of 0.88 [0.13, 0.92], median ClDice of 0.89 [0.06, 1.0], and median HD95 of 1.8 [0.58, 81] mm was achieved. The volume of the labeled anatomic structure was the most relevant determinant of accuracy in this model. Median time to predict was 130.6 [60.9, 284.1] seconds.

CONCLUSIONS

The nnUNet MRA-TOF-based algorithm provided a fast and adequate automatic extraction of unruptured IAs, their parent vessels, and the most relevant cerebral arteries. Future steps involve the expansion of the training set with the inclusion of more MRA-TOF studies with and without IAs and its incorporation in 3D imaging viewers and treatment prediction.

Clinical Evidence on the Influence of Implant Position onto Maximum Output with the Bonebridge Bone Conduction Implant

HYPOTHESIS

In bone conduction implantation, the position of the implant influences the audiological benefit of the patient.

BACKGROUND

One way of treating hearing loss is the implantation of bone conduction implants (BCIs), which effectively transmit vibrations through the skull bone to the cochlea given that the implant transducer is securely fixated. Laboratory research on the efficacy of bone conduction sound transmission found that a closer proximity of the transducer to the ipsilateral cochlea yields significantly higher cochlear promontory vibrations and hence, higher stimulation efficacy. Up to now, this finding has not been reproduced using clinical data such as the functional or effective gain.

METHODS

The present, retrospective study was conducted on a cohort of 28 BCI patients to correlate the implantation site of the BC transducer, derived from clinical postoperative imaging and defined in a standardized coordinate system, with maximum output values that are exclusively based on a novel calculation method only employing clinical audiological data.

RESULTS

It could be shown that the efficacy of BCI stimulation is in fact correlated with the transducer distance to the cochlea, and that this correlation is frequency dependent. Furthermore, the longitudinal distance of the transducer and the ipsilateral external auditory canal is negatively correlated with the maximal output while the sagittal distance is not.

CONCLUSION

The present study is hence the first one to clinically demonstrate the significance of BCI placement for maximizing patient benefit, which should be considered during the preoperative planning of bone conduction implantation.

Statistical Shape Modeling of an Experimental, Induced Cam-Femoroacetabular Impingement Deformity in a Rabbit Model: A Platform to Study Mechanism of Hip Disease

BACKGROUND

Femoroacetabular impingement (FAI) is a common determinant of hip pain in young adults and an established risk factor in the subsequent development of osteoarthritis (OA). The mechanism of hip OA secondary to FAI is unknown. Small-animal models are critical translational tools to understand mechanisms of disease and develop interventional therapies. Kamenaga and colleagues proposed a novel animal model to mimic cam-FAI; however, 3D morphology of the induced deformity has not been objectively investigated.

PURPOSE

To use statistical shape modeling to quantitatively describe the induced proximal femoral head-neck deformity in order to take the necessary step in validating this animal model as a translational model for human cam-FAI.

STUDY DESIGN

Controlled laboratory study.

METHODS

Six-week-old immature New Zealand White rabbits (n = 13) were subject to right femur physis injury, with left femurs serving as controls. Micro-computed tomography images of femurs were taken at minimum 4 weeks after injury. 3D reconstructions were aligned and underwent statistical shape modeling with 2048 particles placed on each femur. Differences between mean shapes were calculated and analyzed using the Hotelling T2 test. Principal component analysis was used to describe shape variation, and parallel analysis was used to determine the statistically significant modes.

RESULTS

Hotelling T2 test demonstrated significant differences between cam-FAI and control mean shapes (P < .01). The cam-FAI mean shape protruded above the control mean by a maximum of 0.8 mm in the anterolateral head-neck junction with sustained protrusions of ~0.6 to 0.8 mm over the anterosuperior aspect and anteroposterior midline of the femoral head-neck junction. Maximum deviations between individual cam-FAI femurs and the mean control femur ranged between 0.1 and 1.6 mm in the same region. The first 6 modes explained 92.1% of the cumulative variation, and the first 13 modes were statistically significant, confirming the deformity.

CONCLUSION

The proposed model resulted in a head-neck cam deformity similar to human cam-FAI.

CLINICAL RELEVANCE

This proposed animal model creates a cam-type deformity similar to that observed in human FAI, helping validate the model as a platform to study mechanisms of hip FAI OA and develop future interventional therapies for this disease.

Reirradiation for recurrent glioblastoma: the significance of the residual tumor volume

PURPOSE

Recurrent glioblastoma has a poor prognosis, and its optimal management remains unclear. Reirradiation (re-RT) is a promising treatment option, but long-term outcomes and optimal patient selection criteria are not well established.

METHODS

This study analyzed 71 patients with recurrent CNS WHO grade 4, IDHwt glioblastoma (GBM) who underwent re-RT at the University of Erlangen-Nuremberg between January 2009 and June 2019. Imaging follow-ups were conducted every 3 months. Progression-free survival (PFS) was defined using RANO criteria. Outcomes, feasibility, and toxicity of re-RT were evaluated. Contrast-enhancing tumor volume was measured using a deep learning auto-segmentation pipeline with expert validation and jointly evaluated with clinical and molecular-pathologic factors.

RESULTS

Most patients were prescribed conventionally fractionated re-RT (84.5%) with 45 Gy in 1.8 Gy fractions, combined with temozolomide (TMZ, 49.3%) or lomustine (CCNU, 12.7%). Re-RT was completed as planned in 94.4% of patients. After a median follow-up of 73.8 months, 88.7% of patients had died. The median overall survival was 9.6 months, and the median progression-free survival was 5.3 months. Multivariate analysis identified residual contrast-enhancing tumor volume at re-RT (HR 1.040 per cm3, p < 0.001) as the single dominant predictor of overall survival.

CONCLUSION

Conventional fractionated re-RT is a feasible and effective treatment for recurrent high-grade glioma. The significant prognostic impact of residual tumor volume highlights the importance of combining maximum-safe resection with re-RT for improved outcomes.

Spinal lumbar Urocortin 3-expressing neurons are associated with both scratching and Compound 48/80-induced sensations

ABSTRACT

Urocortin 3 is a neuropeptide that belongs to the corticotropin-releasing hormone family and is involved in mechanosensation and stress regulation. In this study, we show that Urocortin 3 marks a population of excitatory neurons in the mouse spinal cord, divided into 2 nonoverlapping subpopulations expressing protein kinase C gamma or calretinin/calbindin 2, populations previously associated with mechanosensation. Electrophysiological experiments demonstrated that lumbar spinal Urocortin 3 neurons receive both glycinergic and GABAergic local tonic inhibition, and monosynaptic inputs from both Aβ and C fibers, which could be confirmed by retrograde trans-synaptic rabies tracing. Furthermore, fos analyses showed that subpopulations of lumbar Urocortin 3 neurons are activated by artificial scratching or Compound 48/80-induced sensations. Chemogenetic activation of lumbar Urocortin 3-Cre neurons evoked a targeted biting/licking behavior towards the corresponding dermatome and chemogenetic inhibition decreased Compound 48/80-induced behavior. Hence, spinal lumbar Urocortin 3 neurons represent a mechanically associated population with roles in both scratching and Compound 48/80-induced sensations.

Deep Learning Model of Primary Tumor and Metastatic Cervical Lymph Nodes From CT for Outcome Predictions in Oropharyngeal Cancer

Importance

Primary tumor (PT) and metastatic cervical lymph node (LN) characteristics are highly associated with oropharyngeal squamous cell carcinoma (OPSCC) prognosis. Currently, there is a lack of studies to combine imaging characteristics of both regions for predictions of p16+ OPSCC outcomes.

Objectives

To develop and validate a computed tomography (CT)-based deep learning classifier that integrates PT and LN features to predict outcomes in p16+ OPSCC and to identify patients with stage I disease who may derive added benefit associated with chemotherapy.

Design, Setting, and Participants

In this retrospective prognostic study, radiographic CT scans were analyzed of 811 patients with p16+ OPSCC treated with definitive radiotherapy or chemoradiotherapy from 3 independent cohorts. One cohort from the Cancer Imaging Archive (1998-2013) was used for model development and validation and the 2 remaining cohorts (2002-2015) were used to externally test the model performance. The Swin Transformer architecture was applied to fuse the features from both PT and LN into a multiregion imaging risk score (SwinScore) to predict survival outcomes across and within subpopulations at various stages. Data analysis was performed between February and July 2024.

Exposures

Definitive radiotherapy or chemoradiotherapy treatment for patients with p16+ OPSCC.

Main Outcomes and Measures

Hazard ratios (HRs), log-rank tests, concordance index (C index), and net benefit were used to evaluate the associations between multiregion imaging risk score and disease-free survival (DFS), overall survival (OS), and locoregional failure (LRF). Interaction tests were conducted to assess whether the association of chemotherapy with outcome significantly differs across dichotomized multiregion imaging risk score subgroups.

Results

The total patient cohort comprised 811 patients with p16+ OPSCC (median age, 59.0 years [IQR, 47.4-70.6 years]; 683 men [84.2%]). In the external test set, the multiregion imaging risk score was found to be prognostic of DFS (HR, 3.76 [95% CI, 1.99-7.10]; P < .001), OS (HR, 4.80 [95% CI, 2.22-10.40]; P < .001), and LRF (HR, 4.47 [95% CI, 1.43-14.00]; P = .01) among all patients with p16+ OPSCC. The multiregion imaging risk score, integrating both PT and LN information, demonstrated a higher C index (0.63) compared with models focusing solely on PT (0.61) or LN (0.58). Chemotherapy was associated with improved DFS only among patients with high scores (HR, 0.09 [95% CI, 0.02-0.47]; P = .004) but not those with low scores (HR, 0.83 [95% CI, 0.32-2.10]; P = .69).

Conclusions and Relevance

This prognostic study of p16+ OPSCC describes the development of a CT-based imaging risk score integrating PT and metastatic cervical LN features to predict recurrence risk and identify suitable candidates for treatment tailoring. This tool could optimize treatment modulations of p16+ OPSCC at a highly granular level.

Ultrasound radiomics predict the success of US-guided percutaneous irrigation for shoulder calcific tendinopathy

OBJECTIVE

Calcific tendinopathy, predominantly affecting rotator cuff tendons, leads to significant pain and tendon degeneration. Although US-guided percutaneous irrigation (US-PICT) is an effective treatment for this condition, prediction of patient' s response and long-term outcomes remains a challenge. This study introduces a novel radiomics-based model to forecast patient outcomes, addressing a gap in the current predictive methodologies.

MATERIALS AND METHODS

The study involved 84 patients who underwent US-PICT, with data collected on clinical and demographic factors, alongside radiomic features extracted from ultrasound images. Key radiomic features predictive of the outcome were discerned through Least Absolute Shrinkage and Selection Operator (LASSO) method. Machine Learning models, including Random Forest, XGBoost, and Support Vector Machines, were employed to analyze the radiomics, the clinical and the combined dataset, focusing on calcium removal extent. An external testing was conducted using an independent cohort from a different institution to assess the model's generalizability. Metrics were calculated for the best-performing models, namely area under the curve (AUC) score, sensitivity, specificity, precision or positive predictive value, and negative predictive value.

RESULTS

The selected features were merged with clinical data, notably the calcification's maximum diameter. This enriched dataset was fed into classification models. The superior model achieved an AUC of 0.88 (95% CI 0.73-0.99), with a positive predictive value of 0.92 and a sensitivity of 0.90. In external testing, the combined model achieved an AUC of 0.78. SHAP analysis was employed to highlight the impact of the selected features on the optimal model's effectiveness.

CONCLUSION

The developed radiomics model offers a promising tool for predicting outcomes of US-PICT, potentially guiding clinical decision-making.

Clinically applicable semi-supervised learning framework for multiple organs at risk and tumor delineation in lung cancer brachytherapy

PURPOSE

The generalization ability of deep learning-based automatic segmentation techniques for lung cancer in practical clinical applications remains under-validated. We reported an investigation that validated a robust semi-supervised conditional nnU-Net (SSC-nnUNet) model in multiple organs at risk (OARs) and tumor segmentation in lung cancer brachytherapy, also explored its potential in robot-assisted puncture diagnosis and treatment.

MATERIALS AND METHODS

Six hundred seventy-four patients with CT data from four partially labeled datasets were divided into training and validation sets at a ratio of 4:1, 181 patients from multiple centers (private dataset) with fully annotated data provided by 3 experienced radiation experts were used for testing comparison. Six experienced experts from multiple centers were asked to correct model-generated contours, and 8 junior oncologists were assigned to delineate contours based on model supporting. To verify the feasibility of the contouring model in robot-assisted surgical operations, an equivalent human model experiment was designed specifically for lung cancer puncture treatment.

RESULTS

In model-based experienced expert assessment, the mean revision degree achieved a competitive score of 1.38 % by 6 multicenter experts. In model-based junior oncologist assessment, they acquired a mean revision degree and efficiency improvement of -1.82 % and 83.4 %, respectively. Guided by the segmentation results of OARs and tumors, an average puncture error of 0.78 mm was achieved across 10 puncture experiments.

CONCLUSION

The SSC-nnUNet model showed a significant improvement in the segmentation quality and efficiency especially in junior oncologist delineation. Specifically, robot-assisted experiments illustrated that the model has great application potential in clinical treatment.

Metastasis Detection Using True and Artificial T1-Weighted Postcontrast Images in Brain MRI

OBJECTIVES

Small lesions are the limiting factor for reducing gadolinium-based contrast agents in brain magnetic resonance imaging (MRI). The purpose of this study was to compare the sensitivity and precision in metastasis detection on true contrast-enhanced T1-weighted (T1w) images and artificial images synthesized by a deep learning method using low-dose images.

MATERIALS AND METHODS

In this prospective, multicenter study (5 centers, 12 scanners), 917 participants underwent brain MRI between October 2021 and March 2023 including T1w low-dose (0.033 mmol/kg) and full-dose (0.1 mmol/kg) images. Forty participants with metastases or unremarkable brain findings were evaluated in a reading (mean age ± SD, 54.3 ± 15.1 years; 24 men). True and artificial T1w images were assessed for metastases in random order with 4 weeks between readings by 2 neuroradiologists. A reference reader reviewed all data to confirm metastases. Performances were compared using mid- P McNemar tests for sensitivity and Wilcoxon signed rank tests for false-positive findings.

RESULTS

The reference reader identified 97 metastases. The sensitivity of reader 1 did not differ significantly between sequences (sensitivity [precision]: true, 66.0% [98.5%]; artificial, 61.9% [98.4%]; P = 0.38). With a lower precision than reader 1, reader 2 found significantly more metastases using true images (sensitivity [precision]: true, 78.4% [87.4%]; artificial, 60.8% [80.8%]; P < 0.001). There was no significant difference in sensitivity for metastases ≥5 mm. The number of false-positive findings did not differ significantly between sequences.

CONCLUSIONS

One reader showed a significantly higher overall sensitivity using true images. The similar detection performance for metastases ≥5 mm is promising for applying low-dose imaging in less challenging diagnostic tasks than metastasis detection.

Muscle Force Dynamics Across Increasing Squat Intensity Conditions in Elite Powerlifters

The growing popularity of powerlifting, which consists of the squat, bench press, and deadlift, calls for biomechanically comprehensible coaching strategies. Understanding the muscle forces at work can play a key part in this endeavor. Therefore, the aim of this study was to investigate the effects of increasing intensity in the squat on muscle forces in elite powerlifters. Twenty-nine top-ranked powerlifters from the Austrian team (age: 26.1 ± 5.4 years; 1-repetition-maximum (1-RM): 2.4 ± 0.4 × body mass) performed squats at 70%, 75%, 80%, 85%, and 90% of their 1-RM. Force plates and 3D motion capture data were used to estimate muscle forces utilizing musculoskeletal models in OpenSim. Muscle forces significantly changed with increased intensity, particularly in the gluteus maximus and semitendinosus, which showed the greatest relative increase in muscle force. The vastii muscles exhibited the highest absolute muscle forces. Notably, the hamstrings, calf, and vastii muscle forces barely increased during the deepest and most challenging region of the squat (the sticking region) with increasing intensity. Furthermore, no correlation was found between the athletes' performance level and the ratio of single-joint to multijoint hip extensor muscle forces. These findings highlight the importance of focusing on hip-dominant techniques when squatting with high intensities and supplementary training for knee extensors to optimize performance.

Leveraging descriptor learning and functional map-based shape matching for automated anatomical Landmarking in mouse mandibles

Geometric morphometrics is used in the biological sciences to quantify morphological traits. However, the need for manual landmark placement hampers scalability, which is both time-consuming, labor-intensive, and open to human error. The selected landmarks embody a specific hypothesis regarding the critical geometry relevant to the biological question. Any adjustment to this hypothesis necessitates acquiring a new set of landmarks or revising them significantly, which can be impractical for large datasets. There is a pressing need for more efficient and flexible methods for landmark placement that can adapt to different hypotheses without requiring extensive human effort. This study investigates the precision and accuracy of landmarks derived from functional correspondences obtained through the functional map framework of geometry processing. We utilize a deep functional map network to learn shape descriptors, which enable us to achieve functional map-based and point-to-point correspondences between specimens in our dataset. Our methodology involves automating the landmarking process by interrogating these maps to identify corresponding landmarks, using manually placed landmarks from the entire dataset as a reference. We apply our method to a dataset of rodent mandibles and compare its performance to MALPACA's, a standard tool for automatic landmark placement. Our model demonstrates a speed improvement compared to MALPACA while maintaining a competitive level of accuracy. Although MALPACA typically shows the lowest RMSE, our models perform comparably well, particularly with smaller training datasets, indicating strong generalizability. Visual assessments confirm the precision of our automated landmark placements, with deviations consistently falling within an acceptable range for MALPACA estimates. Our results underscore the potential of unsupervised learning models in anatomical landmark placement, presenting a practical and efficient alternative to traditional methods. Our approach saves significant time and effort and provides the flexibility to adapt to different hypotheses about critical geometrical features without the need for manual re-acquisition of landmarks. This advancement can significantly enhance the scalability and applicability of geometric morphometrics, making it more feasible for large datasets and diverse biological studies.

Peritumoral edema resolves infrequently in surgically treated patients with intracranial meningioma- a retrospective study of 279 meningioma patients

BACKGROUND

The resolution of peritumoral brain edema (PTBE) following surgery for intracranial meningioma (IM) is poorly understood. We hypothesized that PTBE represents a more permanent rather than resolving parenchymal change. Therefore, our aim was to assess the frequency of PTBE resolution following gross total resection (GTR) of IM.

METHODS

IM patients who were operated on in the study hospital between 2000 and 2020, who had preoperative magnetic resonance imaging (MRI) showing PTBE and a follow-up MRI performed at least one year after surgery, were retrospectively identified. To minimize confounding by PTBE related to a postoperative residual tumor, only patients who had undergone GTR were included. PTBE was defined as hyperintensity on either pre- or postoperative fluid-attenuated inversion recovery (FLAIR) MRI sequences.

RESULTS

A total of 279 adult meningioma patients were retrospectively identified. Of these, 208 (74.6%) were graded as World Health Organization grade 1 and 71 (25.4%) as grade 2. Of the 279 patients who had the first postoperative follow-up MRI at one year or later, PTBE changes persisted in 270 (96.8%) patients. However, over 90% resolution in PTBE volume was observed in 102 (35.8%) patients during the median MRI follow-up of 5.0 years (2.3-6.5). Higher edema index (p <.001) and temporal PTBE location (p =.018) were associated with higher resolution percentage of preoperative PTBE.

CONCLUSION

Persisting PTBE is a common finding following GTR of IMs. While complete resolution of PTBE is rare, considerable resolution is often seen. The nature and exact cause of these persisting parenchymal changes are unclear, but they likely represent gliosis.

Transmission-Based Monitoring of Dual-Applicator Microwave Ablation Discriminates Discontiguous from Contiguous Ablation Zones in an In Vivo Porcine Liver Model

PURPOSE

To determine the technical feasibility of discriminating discontiguous from contiguous ablation zones between a pair of microwave ablation (MWA) applicators using broadband microwave transmission signal measurements in an in vivo porcine liver model.

MATERIALS AND METHODS

Dual applicator 2.45 GHz MWA was performed using 1 directional and 1 omnidirectional applicator, spaced 3 cm apart, under imaging guidance. The study involved 15 hepatic MWAs across 4 swine, with ablation durations of 200 seconds (n = 8) for discontiguous ablation and 600 seconds (n = 7) for contiguous ablation, each at 60 W; these ablation durations and applied power combinations were selected with the intent of creating discontiguous (200 s) and contiguous (600 s) ablation zones. A custom software periodically measured transmission signals between the applicators at 46-second intervals. Contrast-enhanced computed tomography (CT), gross pathology, and histopathologic analyses were used to assess the processed transmission signal (PTS).

RESULTS

Statistical analyses revealed significant differences between contiguous and discontiguous ablation zones on contrast-enhanced CT imaging (volume, 16.9 cm3 [SD ± 5.2] vs 3.9 cm3 [SD ± 1.5]; P = .0002) and gross tissue sections and histology (area, 10 cm3 [SD ± 3.3] and 6.5 cm3 [SD ± 1.3]; P = .001), and PTS datasets showed values of 85.1% (SD ± 11) and 37.3% (SD ± 12.9; P = .02). PTS values functioned well as predictors of complete versus incomplete ablation (area under the receiver operating characteristic curve, 0.90), with a PTS threshold of 53% being optimal for indicating ablation zone contiguity. Ablation zone contiguity was strongly correlated with PTS (Spearman correlation coefficient, 0.86; P < .0001).

CONCLUSIONS

This study demonstrated that PTS between dual MWA applicators can distinguish between contiguous and discontiguous ablation zones.

Deep learning-based segmentation of the mandibular canals in cone-beam CT reaches human-level performance

OBJECTIVES

This study evaluated the accuracy and reliability of deep learning-based segmentation techniques for mandibular canal identification in cone-beam CT (CBCT) data to provide a reliable and efficient support tool for dental implant treatment planning.

METHODS

A dataset of 90 CBCT scans was annotated as ground truth for mandibular canal segmentation. The dataset was split into training (n = 69), validation (n = 1), and testing (n = 20) subsets. A deep learning model based on a hierarchical convolutional neural network architecture was developed and trained. The model's performance was evaluated using dice similarity coefficient (DSC), 95% Hausdorff distance (HD), and average symmetric surface distance (ASSD). Qualitative assessment was performed by 2 experienced dental imaging practitioners who evaluated the segmentation quality in terms of trust and safety on a 5-point Likert scale at 3 mandibular locations per side.

RESULTS

The trained model achieved a mean DSC of 0.77 ± 0.09, HD of 1.66 ± 0.86 mm, and ASSD of 0.31 ± 0.15 mm on the testing subset. Qualitative assessment showed no significant difference between the deep learning-based segmentation and ground truth in terms of trust and safety across all investigated locations (P > 0.05).

CONCLUSIONS

The proposed deep learning-based segmentation technique exhibits sufficient accuracy for the reliable identification of mandibular canals in CBCT scans. This automated approach could streamline the pre-operative planning process for dental implant placement, reducing the risk of neurovascular complications and enhancing patient safety.

Association of Exercise Tolerance with Respiratory Health Outcomes in Mild-to-Moderate Chronic Obstructive Pulmonary Disease

Rationale: Previous studies have identified exercise intolerance in patients with mild-to-moderate chronic obstructive pulmonary disease (COPD). The association of exercise tolerance with lung function decline and acute exacerbation risk in mild-to-moderate COPD is unclear, especially in the community population. Objectives: We evaluated exercise tolerance in patients with mild-to-moderate COPD and analyzed its associations with respiratory health outcomes. Methods: We analyzed data from the community-based ECOPD (Early Chronic Obstructive Pulmonary Disease) study of patients with mild-to-moderate COPD (postbronchodilator forced expiratory volume in 1 second (FEV1):forced vital capacity < 0.70 and FEV1 ≥ 50% predicted). Patients who completed questionnaires, spirometry, and cardiopulmonary exercise testing at baseline were included. Annual exacerbation assessment and spirometry were conducted for 2 years consecutively. Exercise tolerance was defined as the percentage of predicted peak oxygen uptake ([Formula: see text]o2peak% predicted). We analyzed the association between exercise tolerance, annual lung function decline, and acute exacerbation risk. Results: Overall, 338 patients were included in the baseline analysis, and 319 completed the 2-year follow up. The mean ± standard deviation of [Formula: see text]o2peak% predicted was 79.8 ± 13.7%. Low [Formula: see text]o2peak% predicted was associated with more chronic respiratory symptoms, worse lung function, severer emphysema, and air trapping at baseline. During the 2-year follow up, a decrease of 13.7% (1 standard deviation) in [Formula: see text]o2peak% predicted was associated with a decline in prebronchodilator FEV1:forced vital capacity (difference, 0.4% [95% confidence interval, 0.1-0.7%]; P = 0.003) and higher total exacerbation risk (relative risk, 1.25 [95% confidence interval, 1.08-1.46]; P = 0.004) after adjustment. Conclusions: Patients with mild-to-moderate COPD and exercise intolerance have worse respiratory health outcomes, for which low exercise tolerance is a prognostic marker. Clinical trial registered with www.chictr.org.cn (ChiCTR1900024643).

Vestibular electrode position stability over time

OBJECTIVES

In vestibular implants (VI), the electrode position is thought to be important for optimal neural activation. The objective of this study was to evaluate the stability of the vestibular electrode position over time.

METHODS

Seven patients implanted with a VI were followed for one year. When possible, the fenestrations of the semicircular canals were kept very small (approximately 0.8 mm) to stabilize the electrode lead. Additionally, the electrodes were fixed at their fenestration sites using bone cement. A temporal bone CT scan was performed intraoperatively, and one week and one year postoperatively. In one patient reliable analysis of the intraoperative CT scan was not possible due to a technical error. A displacement of the vestibular electrodes of more than 0.5 mm was considered significant.

RESULTS

Fourteen out of 18 electrodes did not show a significant displacement between the intraoperative scan and the first postoperative scan. In the remaining four electrodes, a displacement of ≥ 0.5 mm occurred (mean 0.54 mm, range 0.50-0.58 mm). These four electrodes were found in the two first implanted patients. In both cases, the intraoperative CT scan had a slice thickness of 0.5 mm and showed severe scattering. This might imply that the measured displacement was (partially) related to a higher measurement error. None of the vestibular electrodes migrated outside of the ampulla. No displacement was observed in any of the vestibular electrodes between the first postoperative scan and the one-year follow-up scan.

CONCLUSION

The current surgical technique seems to securely stabilize the vestibular VI electrodes over time.

Structure and variability in the female genital atrium of Uropodina (Acari: Parasitiformes)

Primary and secondary sexual characters of Mesostigmata are often used in species descriptions and phylogenetic analyses. The use of these characters has been focused almost exclusively on external structures. Digital 3D reconstruction based on synchrotron X-ray microtomography (SR-μCT) data allowed a comparative investigation of the structure of an internal system, the female genital atrium, in the mite lineage Uropodina (Parasitiformes: Mesostigmata). Despite substantial variability in observed structures, a general model for the endogynium, vagina, and muscle structure has been generated using a combination of SR-μCT and light microscopy. Most of the variations are hypothesized as related to species recognition and/or manipulation of the endospermatophore. The recorded variability may have substantial phylogenetic value, as a previously unreported modification of the vagina appears to diagnose a substantial lineage of "higher" Uropodina. This set of observations also support the hypothesis that the large family Urodinychidae is polyphyletic. Overall, SR-μCT and 3D reconstruction turned out to be very helpful for studies on internal organ systems in these very small organisms, lessening the need for laborious dissections or extensive Transmission electron microscopy-based investigations.

Thrombus perviousness in acute ischemic stroke: a scoping review of methodology, predictive value, and future perspectives

PURPOSE

Thrombus perviousness, depicting the interaction of occlusive clot with contrast media as measured with computed tomography (CT) is a relatively new imaging biomarker in acute ischemic stroke (AIS). This approach holds the potential to tailor revascularization strategies and post-interventional treatments, thereby enhancing functional outcomes. However, its predictive value is not yet conclusive despite its association with several clinical parameters.  METHODS: This scoping review provides a comprehensive overview of 51 articles that explore this issue. It focuses on the analysis of applied methodology in measuring perviousness, the predictive value of perviousness based on available data, and the future perspectives and potentials this biomarker may have in AIS imaging.  RESULTS: Although some data are contradictory, in the vast majority of published studies, pervious thrombi were easier to remove with mechanical thrombectomy, responded better to thrombolytic treatment and showed better functional outcome than impervious thrombi. The methodology of measuring perviousness is not yet standardized, which may lead to inconsistency in the findings. New data on time-resolved (dynamic) perviousness show more promising results and refined characterization of occlusive thrombi in AIS.

T2-weighted MRI radiomics for the prediction of pediatric and young adult rhabdomyosarcoma alveolar subtype and distant metastasis: a pilot study

INTRODUCTION

Rhabdomyosarcomas are the most common soft tissue sarcoma in children. While treatment outcomes have improved, risk-based therapy classification relies on staging and tumor subtypes for therapeutic planning.

OBJECTIVE

This study investigated the utility of T2-weighted MR radiomics features and machine learning models in identifying the presence of distant metastasis and alveolar histological subtypes at baseline imaging in children diagnosed with rhabdomyosarcoma.

MATERIALS AND METHODS

This retrospective cross-sectional study utilized MRIs from 86 patients, 49 (median age (IQR) 59 months (37-161), alveolar subtype=15, distant metastasis=9) of whom had been imaged at outside imaging centers (training set); and 37 (median age 52 months (24-164), alveolar subtype=14, distant metastasis=8) of whom were imaged at our institution (holdout validation set). Radiomic features were extracted from T2-weighted images. We selected features that demonstrated intra-scan repeatability and used maximum relevance and minimum redundancy supervised feature selection to identify the 50 most important features. Lasso logistic regression and support vector machine (SVM) classifiers were trained to predict binary outcomes. The median of all predictions for a given patient was used as patient-level predictions. DeLong's test compared the area under the receiver operating characteristic curves (AUC). Cut-offs obtained by maximizing the Youden index were evaluated on an external validation set, and accuracy metrics were reported.

RESULTS

On the validation set, the Lasso and SVM classifiers obtained patient level AUCs of 0.76 (95% CI 0.59-0.94) and 0.73 (0.54-0.92), respectively, in predicting alveolar subtype, with the Lasso regressor obtaining 71.4% (41.9-91.6) sensitivity and 60.9% (38.5-80.3) specificity. When predicting the presence of distant metastasis, the Lasso and SVM classifier had AUCs of 0.81 (0.67-0.95) and 0.77 (0.58-0.97), respectively. There were no differences between model performance (P>0.05). A total of 12 and 18 features had nonzero coefficients in the Lasso regressors for predicting alveolar subtype and tumor metastasis, respectively.

CONCLUSION

MRI radiomics from baseline T2-weighted MRI demonstrated potential in predicting alveolar subtype and distant metastatic disease at presentation. Larger studies are needed to explore multinomial multiclass models for better prognostication of pediatric rhabdomyosarcomas.

Guiding patient-specific cardiac simulations through data-assimilation of soft tissue kinematics from dynamic CT scan

Fluid-structure interaction (FSI) can be key in the generation of accurate digital replica of cardiovascular systems. To personalize these models, however, several patient-specific parameters need to be measured, which can be challenging to accomplish in a non-invasive manner. Alternatively, the cardiac kinematics of the patient can be extracted from imaging data and then directly imposed as a dynamic boundary condition in the computational model, also incorporating temporal and spatial measurement errors. A more advanced method combines FSI with kinematic driven simulations using data-assimilation. Despite its potential, the application of this technique to complex multi-physics cardiovascular simulations remains limited. In this study, we develop an FSI model of a patient's left ventricle (LV) and aorta, personalized with dynamic imaging data using a Nudging algorithm-a data assimilation technique-which is tailored to each cardiac chamber. In particular, for the LV, which embeds small-scale and irregular endocardial structures (higher measurement errors), the active contraction of the patient is replicated primarily using integral measurements (ventricular volume and surface area). On the other hand, the passive motion of the aorta is guided in the simulation relying directly on the local tissue positions from CT scan. The algorithm's simplicity and zero additional computational cost make it particularly suitable for multi-physics problems. Our results show that the assimilation procedure must be tuned to guide the system toward the measurements within the uncertainty range of the in-vivo data.

Preservation of the lateral femoral circumflex artery in total hip arthroplasty using the bikini-type direct anterior approach : effect on muscle status and clinical outcomes

Aims

The direct anterior approach (DAA) is increasing in popularity as a minimally invasive technique for total hip arthroplasty (THA). DAA-THA involves ligation of the ascending branch of the lateral femoral circumflex artery (a-LFCA), considered to contribute to the blood supply of the tensor fasciae latae (TFL) muscle. To determine the morbidity of these surgical steps, periarticular muscle status and clinical outcomes were assessed after bikini-type DAA-THA with a-LFCA preservation versus ligation.

Methods

We evaluated the surgical records of 140 patients undergoing DAA-THA with continuous attempt of a-LFCA preservation from May to October 2021. A total of 92 patients were eligible and 46 consented to study participation (n = 20 preservation, n = 26 ligation). Preoperative and six-week clinical and radiological data were retrospectively extracted from patient files, and patient-reported outcome measures (PROMs) from the institutional registry. Clinical and MRI examinations were performed two years postoperatively to analyze volume and fatty infiltration of the TFL, gluteus medius, and gluteus minimus relative to the contralateral hip. A total of 13 patients underwent contralateral THA and were excluded from the analysis of muscle status.

Results

Coxa valga morphology and less muscular habitus were more frequent in a-LFCA preservation. After a-LFCA preservation, less anterolateral soft-tissue swelling was described at six weeks (p < 0.001) and TFL local pain at two years (p = 0.034) postoperatively. PROMs did not differ between groups. Mean TFL volume side-difference was not significantly different after a-LFCA preservation (p = 0.276), but it was significantly different after ligation (11.6% smaller (SD 15.5); p = 0.022). TFL fatty infiltration side-difference was larger after a-LFCA ligation (p = 0.010). Muscle status of the gluteus medius and minimus did not differ between sides and groups.

Conclusion

a-LFCA preservation had a minor effect on TFL muscle status. Since preservation was primarily feasible in hips with simpler morphology, it remains uncertain whether differences were due to preserved vascularity or reduced TFL injury. Hence, a-LFCA preservation does not appear essential. However, until further evidence becomes available, attempting a-LFCA preservation may be advisable.

Methodology for comparing absorbed dose rate calculation algorithms in molecular radiotherapy dosimetry

INTRODUCTION

This study compared the results obtained using three of the most frequently proposed algorithms, Local Energy Deposition (LED), Convolution of Voxel S Values (CONV) and Monte Carlo modeling (MC).

METHODS

OpenDose3D, a free, open-source clinical dosimetry software was used to perform this comparison. The assessment focused on absorbed dose rate (ADR), thereby increasing the dataset size and avoiding potential bias associated with the time integration step. From patients treated with 90Y-microspheres, [177Lu]Lu-DOTATATE or Na[131I]I, 52 datasets were processed. Voxel-based ADR maps were computed in homogeneous (water) and heterogeneous (CT-derived) media, for a total of 312 datasets. In a heterogeneous medium, density correction was applied at the voxel level. ADR values were averaged over VOIs, and relative differences (rd) were calculated using the MC results as the reference.

RESULTS

In the homogeneous medium, LED underestimated the ADR by up to 10 % in soft tissues, particularly when cross-irradiation cannot be neglected. Conversely, the ADR obtained with CONV presentedexcellent agreement with the MC simulations (rd ≤ 1 %). In the heterogeneous medium, density correction was crucial. For example in 131I, LED underestimated the ADR values by up to 50 % in lungs and bone marrow. CONV showed excellent agreement with MC in soft tissues (rd ∼ 1 %) and good agreement in organs/tissues where self-irradiation was predominant. The implemented density correction was much less efficient for organs/tissues that experience mostly cross-irradiation.

CONCLUSIONS

Our work proposes a procedure for evaluating ADR algorithms. It also underscores the impact of the medium heterogeneity and cross-irradiation on dosimetry calculations.

Image-Derived Modeling to Assess Coronary Proximity in Patients Undergoing Transcatheter Pulmonary Valve Replacement With Self-Expanding Valves

BACKGROUND

The potential for coronary artery compression (CC) during transcatheter pulmonary valve replacement (TPVR) using self-expanding valves (SEV) is not fully understood, yet anecdotal reports suggest that this risk exists.

AIMS AND METHODS

We performed a retrospective cohort study of patients evaluated for SEV-TPVR to evaluate the relationship between the right ventricular outflow tract (RVOT) and coronary arteries (CA). CT-derived segmentations of the RVOT and CA were created using machine learning. A 2D map of the distance between the RVOT surface and CA, in systole and diastole, was created. In the subset of patients with post-procedural CTA, the distance before and after TPVR was measured.

RESULTS

Forty-two individuals underwent screening for SEV-TPVR, of which 83% (n = 35) had SEV implanted (Harmony = 24; Alterra = 11). Median age was 22.9 years (range 12-60) and 76% had tetralogy of Fallot (TOF). There was no significant change in the distance between the RVOT and LCA between diastole and systole (p = 0.31), yet the RVOT area nearest to the LCA displaced proximally by 11 mm (IQR: 5.6-19.9) in systole. In 8 patients with pre- and post-TPVR CTA, no statistically significant differences were observed in the RVOT-to-LCA relation after intervention. The distance to the LCA was smaller in pulmonary stenosis/atresia patients than those with TOF (median distance 1.2 and 2.1 mm, respectively; p = 0.185).

CONCLUSION

The RVOT area in closest proximity to LCA is dynamic and should be considered when planning TPVR. Special attention should be given to patients with a diagnosis of pulmonary stenosis/atresia.

A magnetic resonance image-based deep learning radiomics nomogram for hepatocyte cytokeratin 7 expression: application to predict cholestasis progression in children with pancreaticobiliary maljunction

BACKGROUND

Hepatocyte cytokeratin 7 (CK7) is a reliable marker for evaluating the severity of cholestasis in chronic cholestatic cholangiopathies. However, there is currently no noninvasive test available to assess the status of hepatocyte CK7 in pancreaticobiliary maljunction patients.

OBJECTIVE

We aimed to develop a deep learning radiomics nomogram using magnetic resonance images (MRIs) to preoperatively identify the hepatocyte CK7 status and assess cholestasis progression in patients with pancreaticobiliary maljunction.

MATERIALS AND METHODS

In total, 180 pancreaticobiliary maljunction patients were retrospectively enrolled and were randomly divided into a training cohort (n = 144) and a validation cohort (n = 36). CK7 status was determined through immunohistochemical analysis. Pyradiomics and pretrained ResNet50 were used to extract radiomics and deep learning features, respectively. To construct the radiomics and deep learning signature, feature selection methods including the minimum redundancy-maximum relevance and least absolute shrinkage and selection operator were employed. The integrated deep learning radiomics nomogram model was constructed by combining the imaging signatures and valuable clinical feature.

RESULTS

The deep learning signature exhibited superior predictive performance compared with the radiomics signature, as evidenced by the higher area under the curve (AUC) values in validation cohort (0.92 vs. 0.81). Further, the deep learning radiomics nomogram, which incorporated the radiomics signature, deep learning signature, and Komi classification, demonstrated excellent predictive ability for CK7 expression, with AUC value of 0.95 in the validation cohort.

CONCLUSION

The proposed deep learning radiomics nomogram exhibits promising performance in accurately identifying hepatic CK7 expression, thus facilitating prediction of cholestasis progression and perhaps earlier initiation of treatment in pancreaticobiliary maljunction children.

Alpelisib in pediatric PIK3CA- and TIE-2-mutant vascular anomalies: a case series on safety, efficacy, and drug exposure

Alpelisib was recently approved by the FDA for the management of pediatric patients with PIK3CA-related overgrowth spectrum. However, this medication was approved in the absence of pediatric pharmacokinetic data, as a fixed 50 mg dose, with no consideration of weight, the primary pharmacokinetically relevant covariate. This raises concerns regarding potential under and over-exposure. Given this gap in information, we aimed to assess the effect of alpelisib in relation to drug exposure (clinical response and drug safety). Alpelisib plasma concentrations were obtained from eight patients under treatment for vascular malformations. Drug exposure determined with area under the curve (AUC) was correlated to drug effect determined by a decrease in the size of lesions and grade of adverse events. Analysis was performed retrospectively. Eight patients received oral alpelisib through the compassionate use program of Novartis. AUC revealed substantial variability (3036 to 16620 ug*h/L) and inversely correlated to weight. Alpelisib resulted in marked clinical improvement, reducing pain, resolving coagulopathy, and improving mobility. Volumetric MRI indicated a 17.4% decrease in targeted vascular anomaly volume after 6 months of alpelisib therapy (p < 0.05), although volume decrease did not correlate with AUC. Adverse events including insulin resistance (n = 8/8) and growth restriction (n = 1/8) were documented, with severity directly correlating to drug exposure. We observed significant weight-related variability in alpelisib plasma concentrations, suggesting that the FDA-approved fixed-dose regimen of alpelisib is not optimal for pediatric patients. Weight-based dosing and therapeutic drug monitoring should be considered to enhance alpelisib safety.

Identification and Protective Strategies for "Cortical Dangerous Veins" in Neurosurgical Craniotomies

OBJECTIVE

To investigate the identification and management techniques and strategies for cortical dangerous veins during neurosurgical craniotomies.

METHODS

We retrospectively analyzed 37 patients who underwent craniotomies involving the intraoperative protection of cortical dangerous veins between July 2022 and June 2024. Preoperatively, high-resolution magnetic resonance imaging data were used for three-dimensional visualization to delineate the resection margins and plan the management of related cortical dangerous veins, including bridging veins, superficial Sylvian veins, anastomotic veins, and veins in the eloquent cortex. Intraoperatively, attention was paid to preserving the cortical dangerous veins. Postoperatively, venous flow disturbances were assessed on the basis of symptoms, imaging findings, physical examinations, and other clinical findings.

RESULTS

Preoperative three-dimensional visualization of the cerebral structures and veins corresponded to the intraoperative observations in all 37 patients. Forty-seven cortical dangerous veins were dissected and protected intraoperatively, including 14 bridging veins, five superficial Sylvian veins, eight anastomotic veins, and 20 veins involving the central lobe. Six patients experienced intraoperative venous injury, which was managed by suturing or pressure hemostasis. No patient exhibited definite postoperative symptoms or radiological changes attributable to venous infarction.

CONCLUSIONS

This study highlights the importance of integrating cortical dangerous vein protection into surgical planning and lesion resection. Employing three-dimensional visualization to advance identification of cortical dangerous veins and preserve vein integrity during resection may mitigate postoperative venous infarction-related deficits without adversely affecting the treatment efficacy.

A modular system of flexible receive-only coil arrays for 3 T Magnetic Resonance Imaging

Flexible form-fitting radiofrequency coils provide high signal-to-noise ratio (SNR) for magnetic resonance imaging (MRI), and in array configuration large anatomical areas of interest can be covered. We propose a modular system - "ModFlex"- of flexible lightweight 4-channel coaxial coil arrays for 3 T MRI. We investigated the performance difference between commercial reference coils and 8- and 16-channel ModFlex receive-only array systems. In vivo, six anatomical targets in four regions of interest - the neck, the ankle, the spine and the hip - were imaged with the novel coil array system. The versatility of ModFlex and the robustness of the coil characteristics for different use cases is demonstrated. We measured an SNR gain for 4 out of 6 and similar SNR for 2 out of 6 anatomical target regions as compared to commercial reference coils. Parallel imaging capabilities are comparable to standard coils in hip and neck imaging, but ModFlex outperforms standard coils in ankle and spine imaging. High SNR combined with high acceleration possibilities enables faster imaging workflows and/or high-resolution MR acquisitions. The coil's versatility is beneficial for use cases with varying subject sizes and could improve patient comfort.

The Influence of Anthropometric Factors on Renal mpMRI: Insights From Regional Analysis

BACKGROUND

Multiparametric MRI (mpMRI) provides detailed insights into renal function, but the impact of anthropometric factors on renal imaging is not fully understood.

PURPOSE

To investigate regional correlations between mpMRI parameters and age, body mass index (BMI), and body surface area (BSA).

STUDY TYPE

Prospective, cross-sectional observational study.

POPULATION

Twenty-nine healthy volunteers (44.5 ± 18.3 years, 18 females) without a history of renal disease.

FIELD STRENGTH/SEQUENCE

3-T, pseudo-continuous arterial spin labeling, multi-echo gradient-recalled echo, diffusion-weighted imaging, T1 and T2 mapping.

ASSESSMENT

Bilateral kidneys were segmented into nine concentric layers (outer cortex to inner regions) and nine equiangular sections (lower to upper pole). Key parameters (renal blood flow [RBF],R 2 * , apparent diffusion coefficient [ADC], T1 and T2 maps) were correlated with age, BMI, and BSA. Differences in parameters between age and BMI groups were also evaluated.

STATISTICAL TESTS

Spearman correlation, Mann-Whitney U test, and rank-biserial correlation coefficient for effect size. A P-value <0.05 was considered statistically significant.

RESULTS

RBF correlated negatively with age in all regions and BMI in inner layers and lower pole. ADC negatively correlated with BMI (significance was not reached in layers 2, 7, 8; P-value = 0.06-0.12) and BSA in layers 1-7. T1 negatively correlated with age in inner regions and lower medial pole. Significant positive correlations were found between age andR 2 * (outermost layer, upper pole), age and T2 (inner and cranial-caudal regions), as well as BMI and T2 (except upper pole; P-value = 0.06). Significant differences between age groups were observed for RBF (all regions),R 2 * (outermost and second innermost layers, central lateral region), T1 (innermost layer), and T2 (upper medial pole). Between BMI groups, ADC (middle layers, upper medial pole) and T2 (outermost and inner layers, lower pole to lateral region) differed significantly.

DATA CONCLUSION

Intrarenal variance of mpMRI parameters correlated with age, BMI, and BSA.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 1.

Diagnosis of Salivary Gland Tumors Using Ultrasound Radiomics

OBJECTIVE

The current diagnosis of salivary gland tumors (SGTs) is dependent on subjective ultrasound features. Here we aimed to develop an objective method using ultrasound radiomics.

METHODS

We collected 248 benign and 46 malignant images and divided them into training (80%) and testing (20%) groups, with 105 radiomic features extracted from each image. Data re-sampling, feature selection and classification were conducted. The diagnostic accuracy of different combinations was evaluated.

RESULTS

After data re-sampling using the Synthetic Minority Over Sampling Technique (SMOTE) and feature selection with LASSO+ANOVA, 10 radiomic features were selected. Using the Random Forest classifier, the testing set achieved an area under the receiver operating characteristic curve of 0.85, accuracy of 90%, sensitivity of 78% and specificity of 92% for diagnosing SGTs. It maintained an accuracy of 85% in a separate internal validation set.

CONCLUSION

This study offers significant insights into the use of radiomics for the diagnosis of SGTs. When selected properly and paired with a suitable classification model, radiomics can be used to differentiate between benign and malignant SGTs.

Effect of Paravertebral Muscle Degeneration on Lumbar and Pelvic Sagittal Alignment in Patients With Degenerative Scoliosis

Study DesignRetrospective study.ObjectivesTo explore the relationship between lumbar spine muscle mass and lumbar pelvic sagittal parameters in patients with degenerative scoliosis.MethodsThis study included ADS patients who were treated in our hospital from 2019 to 2023. The spinal parameters were evaluated through X-rays, and the relative muscle volume (RMV) and fat infiltration (FI) were measured through three-dimensional reconstruction. Patients were categorized into 3 groups based on SRS-Schwab sagittal balance correction (0, +, ++), and into 3 groups based on GAP score (proportioned, moderately dis-proportioned, severely dis-proportioned). Finally, patients were classified into low-quality and high-quality groups based on the FI of Paraspinal muscles (PSM).ResultsThe study included a total of 63 patients. Significant statistical differences were observed in the FI and RMV of MF, ES and PS among patients classified by SRS-Schwab PT classification. Additionally, significant statistical differences were found in the RMV of MF and PS among patients classified by SRS-Schwab PI-LL classification and GAP score. Furthermore, a significant correlation was found between the FI and RMV of PSM and lumbopelvic sagittal parameters. The ordinal regression model analysis revealed that FI of ES significantly impacted PT imbalance, while RMV of MF significantly impacted PI-LL imbalance. Moreover, significant differences were noted in PT and PI between the low-quality and high-quality multifidus groups.ConclusionsAs sagittal imbalance worsens, PSM degeneration also intensifies, primarily characterized by an increase in FI and a decrease in RMV. Notably, PT and PI-LL are positively correlated with RMV and negatively correlated with FI.

Development and validation of a voxel-based computational phantom of Halobatrachus didactylus (Toadfish) to study radiation protection of the marine environment

This paper describes the creation of a voxel-based dosimetric model of a toadfish (Halobatrachus didactylus) from available CT images in which 29 organs are identified and segmented using 3D Slicer software. To validate the present dosimetric model Specific Absorbed Fractions (SAFs) were calculated for ten vital organs as sources and all targets, for discrete energies of electrons and photons. Then, the obtained dosimetric coefficients were compared to those calculated in similar organs in a voxel-based phantom of a trout fish (Oncorhynchus mykiss). The results showed a general agreement between the comparative dosimetric data, highlighting the validity of our dosimetric model for Halobatrachus didactylus. We conclude that the current dosimetric fish model can be applied in future research to evaluate dosimetry calculations related to radiation exposure in regions surrounding the Fukushima accident site.

Comparison of Measurement Methods for Stone Volume Estimation: An In Vitro Study

Introduction: Urolithiasis guidelines still rely on the maximum stone diameter to propose treatment strategy, although this measure is known to have many pitfalls. Stone volume (SV) could represent a more accurate measurement, helping to plan the treatment or follow-up. Various methods to measure SV have been proposed. We aimed to compare different methods to estimate SV. Methods: Fifteen stones (human and artificial) were assessed. Real SV was measured using the water displacement method. Volume estimation included three diameter-based formulas (Ackerman, 4/3 Pi r3 and r3/2) and two 3D segmentation methods (Horos and Kidney Stone Calculator [KSC]). All measurements were done by a single operator. Spearman correlation test and comparative analyses were conducted between the real and the estimated SV. Results: Compared with real SVs, Ackerman and r3/2 formulas estimated volume accurately in 2/15 (13%) of stones each. No accurate measurement was reported using the sphere formula. KSC did estimate volume accurately in 4/15 (27%) stones compared with the reference SV; Horos did it in 7/15 (47%) stones. Both segmentation methods presented strong correlation coefficients (r = 0.9642 and 0.9659, p < 0.0001), while formula correlation was moderate (r = 0.7531, p < 0.0001). Conclusion: Formulas and segmentation methods for SV estimation resulted in divergent outcomes. Segmentation methods (Horos and KSC) presented higher accuracies in SV estimation, compared with real SV. Formulas were the least accurate.