Physics in Medicine & Biology

To evaluate whether eight-bin deep-silicon photon-counting CT can quantify iodine and gadolinium separately and distinguish radiopaque embolic beads from vascular contrast in tumor phantoms. In this phantom study, two-input decomposition of scanner-generated iodine and water images was compared with three- and five-material decomposition of five reconstructed energy-bin images in 56 common test vials, including nine simulated-blood preparations. One tantalum and one bismuth vial were also tested with five-material decomposition. A separately calibrated two-input method using scanner-generated 70-keV and iodine images was applied to four static 3D-printed tumor vascular models: iodine-only and gadolinium-only controls and two models with iodine-containing beads and gadolinium-filled channels. Concentration errors were summarized by bias and root mean square error (RMSE). For iodine, RMSE was 0.40, 0.73, and 1.12 mg I/mL with two-input, three-material, and five-material decomposition, respectively; corresponding gadolinium RMSEs were 0.57, 0.55, and 0.46 mg Gd/mL. Two-input biases were +0.01 mg I/mL and +0.29 mg Gd/mL. Five-material decomposition distinguished tantalum and bismuth in the separate test vials. Beads and vascular gadolinium were spatially separated in the tumor models. In the two bead-containing models, gadolinium measured 32.56 {+/-} 8.02 and 37.02 {+/-} 7.93 mg Gd/mL (mean {+/-} spatial SD), compared with the prepared 35.00 mg Gd/mL. Deep-silicon photon-counting CT separately visualized radiopaque embolic beads and quantified vascular gadolinium from a single acquisition. Recalibrated scanner-generated images supported quantitative dual-contrast imaging, while research energy-bin decomposition distinguished additional materials.

Background: Patient-specific dosimetry in targeted radionuclide therapy (TRT) is increasingly supported by quantitative imaging; however, voxel-scale dose estimates cannot resolve cellular and subcellular heterogeneity present within tumor tissue. Conventional cellular dosimetry models often represent cells as regularly packed spheres, which may not capture realistic tumor microarchitecture. This study investigated how histopathology-informed three-dimensional (3D) tumor microarchitecture influences cellular absorbed-dose distributions, predicted radiobiological response, and treatment-related conclusions in TRT. Methods: Realistic 3D cellular tumor models were constructed from thick-section 3D cyclic immunofluorescence (3D CyCIF) imaging data. Monte Carlo radiation-transport simulations were performed for Lu-177, Pd-103, Tb-161 and Y-90, with activity localized in cancer-cell subcellular compartments and in non-cancer targets. Absorbed dose per total cluster decay was calculated individually for each cell nucleus and compared with corresponding simplified spherical tumor models. Cell survival fractions were estimated using the linear-quadratic model, and tumor control probability (TCP) was calculated from the individual cancer-cell survival probabilities. Results: Simplified model overestimated mean cancer-cell nuclear absorbed dose by 9.54% to 14.51% for nuclear localization and by 33.23% to 44.88% for cytoplasmic and membrane localization. In larger tissue models, simplified and realistic geometries produced comparable mean absorbed doses in several cases; however, the realistic models consistently generated broader cellular dose distributions, including low-dose cancer-cell subpopulations not captured by the simplified models. These dose-distribution differences propagated into wider survival-fraction distributions and altered TCP predictions. In the spatially heterogeneous tumor region, the realistic model required 1.11- to 1.33-fold higher activity to reach TCP90 for nuclear localization, and 1.49- to 5.93-fold higher activity for cytoplasmic and membrane localization. For similar tumor control, the radionuclide and source-localization combination associated with the greatest preservation of surrounding non-cancer cells differed between the simplified and realistic models. The intra-voxel analysis further showed that homogeneous voxel-scale activity assumptions can obscure biologically relevant microscopic dose heterogeneity. Conclusion: Native 3D tumor microarchitecture substantially influences cellular absorbed-dose distributions, predicted biological response, and treatment-related conclusions. Histopathology-informed 3D cellular models can be used as a complementary framework for characterizing cellular dose heterogeneity and may improve the evaluation of radionuclides and subcellular targeting strategies in heterogeneous tumor tissue.

Head and neck radiation oncology, with its complex locality and epidemiology, requires various imaging techniques to achieve sufficient visualization. Through collaboration with medical physicists, radiation therapists, radiation oncologists, and neuroradiologists, our group clinically implemented a standard set of magnetic resonance imaging (MRI) sequences on an MR-simulation scanner for three anatomical regions: skull base, oral cavity, and larynx/pharynx. A total of 41 patients with head and neck cancer and 3 volunteers were scanned on either a 1.5T or 3T MR-simulation scanner. Sequences described include Dixon-based approaches, SPAIR-based approaches, short T2-based approaches, and quantitative imaging methods. We describe and outline the steps to optimize these notable MRI sequences to aid in successful implementation in a radiation oncology clinic.

Trans-arterial radioembolization using yttrium-90 (Y-90) microspheres treats unresectable liver cancer with radiation. The tumor-to-parenchyma dose ratio is mainly governed by the patient-specific hepatic arterial flow distribution, which transports the microspheres. Even though computational fluid dynamics (CFD) simulation can predict this flow distribution, its high computational cost and time limit intraprocedural application. To overcome this computational limitation, a physics-informed neural network (PINN) surrogate was developed to provide rapid predictions of flow distribution in a patient-specific hepatic artery model. The surrogate maps spatial coordinates to the velocity vector U with Cartesian components u, v, w and the static pressure p. Reference velocity and pressure fields came from a converged finite-element CFD solve of the same geometry at about 1.05 x 105 nodes; a controlled fraction was used to supervise the training while the remainder was held out for validation. Using the NVIDIA PhysicsNeMo framework, we compared four coordinate networks (MLP, Modified Fourier, Multi-Scale Fourier and SIREN) under soft and hard inlet enforcement across a supervised data sweep (data fraction df = 0 to 0.20) and repeated df = 0.20 at a matched 300,000-step budget. Predicted flow splits were propagated through an established Y-90 dose-kernel pipeline. Accuracy is governed primarily by the supervised-data fraction, not the architecture: at df [≥] 0.10 the backbones agree to within one to two R2 points, and R2(u) rises from {approx} 0 under pure physics (df = 0) to {approx} 0.95 by df = 0.01 for the best backbones, saturating near 0.97 to 0.99 by df = 0.20. At a matched 300,000-step budget the Multi-Scale Fourier network leads (R2(u) = 0.978) ahead of SIREN (R2(u) = 0.976) and the MLP (R2(u) = 0.968); the best value seen (R2(u) = 0.985, Modified Fourier) needed 1.8 times that budget. On the df = 0.20 checkpoint the per-outlet flow split matches CFD (r = 0.999, L1 = 4.4 percentage points) and the derived dosimetry agrees to within 2.4% for perfused volume, mean dose and every isodose volume. At df = 0 training collapses toward the trivial near-zero state, which satisfies continuity, momentum and the zero-pressure outlets exactly and is ruled out only by the inlet and data terms; this, not architecture choice, is the limiting failure mode. Once trained, the network evaluates the full 105,284-node field in about 2 s on a four-core CPU. The supervised-data fraction, with a practical floor between df = 0.05 and 0.10, is the primary design lever for such hepatic-arterial PINN surrogates.

Accurate hippocampal electric field (E-field) prediction is important for individualized electroconvulsive therapy (ECT) dosing and emerging deep-brain stimulation approaches such as temporal interference. Because hippocampal fields are far weaker than those beneath the stimulation electrodes, they are highly sensitive to tissue interfaces, conductivity contrasts, and numerical discretization, so numerical error can rival the physical uncertainties of anatomy and conductivity. We investigate this numerical component of hippocampal dosimetry using individualized head models from eleven ECT patients. Identical head40 segmentations, comprising 40 tissue compartments including the bilateral hippocampi, were simulated with three independent numerical methods: a charge-based boundary-element fast multipole method (BEM-FMM), a voxel-based hexahedral finite-element solver, and a conformal tetrahedral finite-element solver. For each, resolution was increased systematically and convergence assessed from the average hippocampal E-field magnitude. At the highest attainable resolutions, the three methods agreed closely, differing by 6%-7% in the left hippocampus and 9%-12% in the right. At default or coarse resolutions, however, they disagreed by 28.8% on average and up to 50.3% in the right hippocampus. Numerical convergence should therefore be established before differences are attributed to anatomical or conductivity uncertainty, and independent-solver agreement provides a practical basis for reliable hippocampal dosimetry.

Biological effectiveness in ion therapy varies with radiation quality and biological context, yet cancer cell response data acquired under well-characterized physical conditions remain limited. We investigated clonogenic survival and relative biological effectiveness (RBE) in PANC-1 and BxPC-3 pancreatic ductal adenocarcinoma (PDAC) cells irradiated at four positions along pristine 180.1 MeV proton and 352.4 MeV/u carbon ion Bragg curves, with 4 cm spread-out Bragg peak (SOBP) irradiation included as a clinically relevant comparison. Cell plane positioning and dose were verified in a water-tank system using an Advanced Markus ionization chamber, and Monte Carlo simulations using TOPAS provided dose-averaged linear energy transfer (LETD) and microdosimetric quantities. RBE was calculated relative to 200 kVp X-rays using the linear-quadratic survival model. Across the sampled pristine beam positions, spanning LETD values of approximately 1-9 keV/m for protons and 12-163 keV/m for carbon ions, RBE increased with LETD. Proton RBE10 exceeded the conventional clinical value of 1.1 at distal positions, whereas carbon ion RBE10 increased from 2.3 to 5.2 in PANC-1 and from 1.3 to 3.7 in BxPC-3, with larger differences between cell lines observed near the Bragg peak. SOBP measurements showed different biological responses across field conditions despite comparable LETD. These physically characterized survival data provide PDAC-relevant benchmarks for evaluating whether RBE models reproduce biological responses across pristine and SOBP irradiation conditions, supporting the future integration of radiation-quality-dependent effects into ion planning.

Deep-silicon photon-counting CT uses a detector architecture distinct from cadmium-telluride systems and has not been evaluated for pediatric head imaging. To compare attenuation-dependent performance of deep-silicon photon-counting CT (dSi-PCCT) with energy-integrating detector CT (EID-CT) and assess spectral and ultra-high-resolution (UHR) reconstructions from one acquisition. Graded-thickness cranial attenuator and pediatric head phantoms were scanned at a matched target volume CT dose index of 18 mGy. dSi-PCCT used 120 kVp and generated 50-, 65-, and 100-keV images; EID-CT used 80 and 100 kVp. We measured CT-number bias, noise, contrast, and contrast-to-noise ratio (CNR) across 0--9 mm of nominal skull thickness. Mixed-effects models assessed imaging-condition effects. We compared matched standard-resolution and UHR temporal-bone images and generated calcium- and water-density maps from the same dSi-PCCT scan. Noise increased by 4.54%, 3.07%, and 1.35% per millimeter at 50, 65, and 100 keV, respectively, versus 8.38% and 7.84% at 80 and 100 kVp. Corresponding CNR decreases were 4.17%, 3.25%, and 1.75% versus 8.94% and 8.54% (interaction {rho} < 0.001). Changes in absolute CT-number bias and contrast did not differ among conditions ({rho} = 0.92 and {rho} = 0.59). In the pediatric head phantom, model-estimated 50-keV CNR ratios were 1.299 (95% CI, 1.163--1.452) versus 80 kVp and 1.424 (95% CI, 1.275--1.592) versus 100 kVp. Across three temporal-bone edges, UHR transition widths were 58.6%-61.9% narrower than standard-resolution widths. dSi-PCCT maintained CNR more consistently as nominal skull thickness increased and provided complementary spectral and UHR pediatric head images from one acquisition.

To compare virtual noncontrast (VNC) error between deep-silicon photon-counting CT (PCCT) and dual-energy energy-integrating detector CT (EID-CT) in a matched phantom and characterize tissue-specific error in humans. This prospective exploratory study included a matched phantom experiment and pilot human evaluation. Prototype projection-domain water-density maps served as the VNC surrogate. Twenty-four vials containing 0.1--20 mg I/mL in deionized water or simulated blood were scanned with PCCT and EID-CT. Three men underwent true noncontrast and contrast-enhanced PCCT; 80 matched measurements were analyzed. Paired t tests and mixed-effects models assessed system differences and iodine associations. Three participants (mean age, 67 {+/-} 10 years; all men) were included. Mean VNC error was 29.1 HU lower with PCCT than EID-CT (95% CI, 15.5--42.7 HU; {rho} < 0.001). Error was not associated with iodine concentration for PCCT (slope, 0.03 HU per mg I/mL; 95% CI, -0.92 to 0.98; {rho} = 0.95) but increased for EID-CT (slope, 3.51 HU per mg I/mL; 95% CI, 2.56--4.46 HU; {rho} < 0.001). The slopes differed by 3.48 HU per mg I/mL ({rho} < 0.001). In humans, errors in five non-blood-pool tissues were within 5 HU. Error was not associated with iodine concentration among 61 non-blood-pool measurements (adjusted slope, -0.17 HU per mg I/mL; 95% CI, -1.22 to 0.88; {rho} = 0.75). PCCT provided VNC measurements with errors closer to zero and less dependent on iodine than EID-CT in the phantom. In humans, error was generally small in non-blood-pool tissues and was not associated with iodine concentration.

Faster ultrahigh-resolution (UHR) photon-counting CT acquisitions facilitate breath-hold imaging, but identical reconstruction settings may conceal differences in spatial resolution. To assess spatial resolution and depiction of small airways and venous stents across helical UHR acquisition configurations. Stationary wire, airway, and venous-stent phantoms underwent one acquisition per configuration using five configurations on a dual-source scanner at 120 kVp. Reconstructions used 0.2- and 0.4-mm thicknesses, the same sharp kernel, and iterative reconstruction strength 1. We measured the full-width at half-maximum of the slice sensitivity profile (SSP) and spatial frequency at 20% of the modulation transfer function (MTF20). Rank-based F tests with Holm adjustment assessed configuration differences in resolution and variability across slices. Airway detection used a profile contrast-to-noise criterion; stent assessment was qualitative. Calculated acquisition time for 400-mm coverage decreased from 9.80 to 1.30 seconds between the slowest and fastest configurations. At 0.2-mm reconstruction thickness, median SSP increased from 0.316 to 0.496 mm, and median MTF20 decreased from 2.870 to 2.059 lp/mm (both P<.001). Interquartile ranges increased from 0.047 to 0.197 mm for SSP and from 0.256 to 0.654 lp/mm for MTF20 (both variability P<.001). Increasing reconstruction thickness to 0.4 mm increased SSP by a median 0.221 mm across 2459 matched slice-wire measurements (P<.001). At 0.2-mm reconstruction thickness, 0.4- and 0.6-mm lumens with 0.2-mm walls were detected only in the slowest configuration. Faster configurations showed less distinct stent struts and greater adjacent-wall distortion. Faster helical UHR photon-counting CT configurations had poorer spatial resolution despite identical nominal slice thickness. Consistent acquisition and reconstruction settings are important for longitudinal comparisons.

Purpose: To quantify the sampling error of conventional 2D regions of interest (ROIs) compared with whole-organ 3D volumetric CT attenuation for opportunistic hepatic steatosis screening, and to evaluate its dependency on intraparenchymal spatial heterogeneity. Materials and Methods: In this retrospective computational study (CLAIM-compliant), 971 abdominal CT scans were analyzed across a multicenter derivation cohort (TotalSegmentator v2, n = 841) and an external validation cohort (Medical Segmentation Decathlon Task03, n = 130). Automated 3D liver segmentation was performed using nnU-Net, followed by an 8-mm spherical 3D erosion to eliminate peripheral partial volume effects and major hilar vessels. Three clinical sampling paradigms (1, 3, and 5 axial 2D ROIs of 15 mm diameter) were modeled through 500 stochastic Monte Carlo iterations per scan (485,500 total simulations). Absolute sampling error (|Delta HU| = |HU_ROI - HU_3D|) and diagnostic misclassification rates at the standard <= 40 HU steatosis threshold were computed. Results: In the derivation cohort, the conventional 3-ROI approach yielded a mean absolute error of 6.61 +/- 7.83 HU (median: 2.92 HU), significantly underperforming compared to the 5-ROI strategy (5.21 +/- 6.00 HU; Wilcoxon paired test, p < 0.001) but improving on the single-ROI approach (11.32 +/- 14.16 HU; p < 0.001). Sampling error correlated strongly with parenchymal spatial heterogeneity (Pearson r = 0.850, 95% CI: [0.830-0.868], p < 0.001). In borderline attenuation zones (35-45 HU), diagnostic misclassification reached 45.2% in heterogeneous livers. Findings were corroborated on external validation, confirming the reproducible decay of sampling error (|Delta HU| = 7.38 HU for 1 ROI vs. 5.14 HU for 3 ROI, p < 0.001). Conclusion: Manual 2D ROI sampling exhibits an intrinsic metrological uncertainty exceeding clinically relevant thresholds (5 HU) in heterogeneous livers, misclassifying nearly half of borderline cases. Opportunistic screening requires fully automated 3D volumetric extraction embedded into clinical PACS workflows.

This study presents a systematic investigation of the brain's own microwave emissions in the 0.85-5.0 GHz range, aiming to validate microwave encephalography (MWEG) as a novel tool for cognitive state assessment. Early experiments (2016-2019) using comparisons with background noise, a control phantom, and different antenna types confirmed the endogenous origin of the recorded signals. In a 2021 experiment (118 measurements on one subject, right and left occipital areas), we observed a significant functional hemispheric asymmetry: the number of "substantial spikes" was significantly higher in darkness for the right hemisphere (p<0.001) and higher in light for the left hemisphere (p<0.001). A 2025 group study (three subjects) confirmed the asymmetry (p<0.05 for right, p<0.01 for left). We describe the signal processing pipeline, including spike detection and statistical analysis. Two complementary mathematical models are presented: an energy-based model linking optical input to microwave emission via neural activity modes, and an information-based model interpreting spectral changes in terms of structural pattern overlap with memory engrams. An estimation of the physical dimensions of possible brain resonators (2-22 mm) is provided. The results are discussed in the context of hemispheric specialization and the information commutation theory. This work opens new perspectives for developing low cost, contactless brain monitoring systems.

Background: Chimeric antigen receptor (CAR) T cell therapy has had clinical success in hematologic malignancies, but limited efficacy is seen in solid tumors. In this study, we investigated whether systemic CAR T cell therapy could be enhanced in metastatic models of neuroblastoma when combined with radiopharmaceutical therapy (RPT). Methods: Non-irradiated or irradiated tumor cells were co-cultured with CAR T cells (1:1) in vitro and supernatant media was subsequently collected for cytokines analyses. CAR T cell phenotypes were characterized by flow cytometry including checkpoint marker expression. Xenograft models of metastatic neuroblastoma were generated in NOD-Rag1nullIL2rgnull (NRG) mice. Tumor-bearing mice received 1.8 Gy of radiation delivered by 177Lu-NM600 RPT five days after tumor implantation. Nine days after RPT, CAR T cells were administered intravenously. To evaluate tumor burden, mice were imaged weekly for 4 weeks. Results: In models of metastatic neuroblastoma, 177Lu-NM600 RPT significantly increased overall survival when combined with CAR T cell therapy in vivo. Pre-treatment of tumor cells with 177Lu also significantly increased CAR T cell cytotoxicity while decreasing production of IL-4 and IL-10 in vitro. Co-culture of CAR T cells with irradiated tumors led to increases in PD-1+TIM3+LAG3+ T cells, suggesting that further combination with immune checkpoint inhibitors may enhance clinical efficacy. Conclusions: Our findings demonstrate that low-dose RPT can potentiate the anti-tumor efficacy of CAR T cells in metastatic neuroblastoma. To our knowledge, this is the first report of dosimetry-based RPT being combined with CAR T cells in a metastatic solid tumor setting. These findings underscore the potential of combining RPT and CAR T cells to overcome the unique challenges of solid tumors, particularly when treating metastatic disease.

In magnetic particle imaging (MPI), signal is generated from the non-linear magnetic response of superparamagnetic iron oxide (SPIO) nanoparticles. The optimization of SPIOs for MPI is an active area of investigation. Tracer performance can be evaluated by magnetic particle relaxometry (MPR) and image-based metrics. This study characterized and compared the performance of ten commercially available tracers, spanning a broad range of magnetic core sizes, hydrodynamic diameters and surface coatings, using the peak signal intensity measured from MPR, and the total signal and maximum signal intensity measured from 2D MPI images. Synomag-D tracers exhibited the highest MPR peak signal intensities and highest maximum signal per {micro}g iron. ProMag, a micron-sized iron oxide particle (MPIO), yielded a low MPR peak signal intensity but had the highest total MPI signal per {micro}g of iron. A strong correlation was observed between MPR peak signal intensity and MPI maximum signal intensity (R2=0.94). In contrast, there was a weak correlation between MPR peak signal intensity and MPI total signal intensity (R2=0.38), though this correlation improved when MPIOs were excluded from the analysis (R2=0.91). While MPR generally predicts the tracer performance, it does not completely replicate the complex imaging conditions. As such, comprehensive tracer evaluation requires combining both MPR and image-based metrics.

Introduction: Resection margins in oral squamous cell carcinoma (OSCC) are frequently inadequate. This study investigated the accuracy and diagnostic performance of ex-vivo 7T MRI against histology. Methods: Fifty OSCC specimens underwent ex-vivo 7T MRI. Tumours were segmented on corresponding histology and twice on MRI: blinded to histology (radiologist segmentation; clinical practice) and using histology (histology-MR segmentation; MR-based ground truth). Resection margins were measured in five regions and compared with histology. Results: Anterior and posterior margins were generally smaller on MRI than on histology (maximum mean difference -2.5 mm), whereas cranial, caudal and deep margins were generally larger on MRI (mean differences <1 mm). For close margins (<5 mm), sensitivity was 95% for histology-MR segmentations and 91% for radiologist segmentations, with 75% specificity for both. Conclusion: Ex-vivo 7T MRI accurately measures resection margins and detects close margins with high sensitivity. A future clinical trial will determine whether intraoperative implementation reduces inadequate margins and adjuvant treatment.

The lack of non-invasive, quantitative methods to track the location and proliferation of transplanted cells hampers the clinical translation of cardiac regenerative therapies. Here, we establish a PET reporter for tracking human induced pluripotent stem cell (hiPSC)-derived cardiac cells across complementary two-dimensional (2D) and three-dimensional (3D) cardiac tissue models. Methods hiPSCs were genetically engineered using CRISPR-Cas9 to express a human-derived, anticalin-based, bioorthogonal PET reporter (DTPA-R) that specifically binds radiolabeled DTPA-metal complexes. The engineered hiPSCs were differentiated into ventricular progenitors (HVPs) and cardiomyocytes (CMs). Stability of reporter gene expression and its effects on cellular differentiation and function were evaluated. Uptake of [18F]F-DTPA*Tb by DTPA-R-positive and -negative HVPs was analyzed in vitro, and DTPA-R HVPs were evaluated in ex vivo cultured porcine myocardial slices with and without radiofrequency ablation (RFA) injury. Phantom studies were performed to evaluate PET detectability under clinically relevant large-animal and human imaging conditions and to assess how activity distribution affects signal detection in relation to DTPA-R cardiac cell-equivalent numbers. Results DTPA-R expression was maintained during the differentiation stages of hiPSC-derived cardiac cells and had no measurable impact on migration, differentiation, and functional integration. PET imaging enabled the specific detection of DTPA-R HVPs in intact and injured myocardial slices. At the initial phantom scan, the measured activity corresponded to cell-equivalent numbers of approximately 3.2 x 10^6 HVPs or 5.5 x 10^6 CMs. The compact 200 uL source remained clearly detectable at 6 h, whereas the more dispersed 1,000 uL source progressively approached background activity, indicating that PET detectability depended strongly on local activity concentration and distribution. Conclusion This study establishes a bioorthogonal PET reporter gene platform for quantitative imaging of hiPSC-derived cardiac cells, with no measurable adverse effects on the biological parameters examined, and provides a methodological foundation for future translational and in vivo imaging studies. The approach may also be applicable to other hiPSC-derived cell products beyond cardiac cell lineages.

Background: Comprehensive evaluations of lung nodule detection methods remain limited, particularly regarding data from clinical routine, novel photon-counting CT (PCCT) technology, and the comparison of academic and commercial models. Materials and Methods: This retrospective study collected 1860 routine PCCT scans acquired at Hannover Medical School from 2021 to 2024. We compared four deep-learning-based lung nodule detection models, which were developed for conventional CT, namely the publicly available TotalSegmentator, nnDetection, and RadYOLO methods, as well as a commercial lung computer-aided detection (CAD) system. We evaluated the inter-model agreement in an unannotated cohort (n=1699), sensitivity in a smaller cohort of 25 manually annotated scans, and false positives as detections in 136 report-negative scans. Results: In the unannotated cohort, the commercial CAD system detected 7071, TotalSegmentator 4264, nnDetection 5254, and RadYOLO 3122 nodules. Overall, the inter-model agreement was low, and the models produced substantially different candidate nodules. Approximately 1300 nodules were detected by all models, representing 18.7% - 42.8% of each model's detections. 15.4% - 37.4% of a model's detections did not match any other model. The models showed moderate sensitivity (between 0.71, 95%-CI [0.64, 0.77], and 0.83, 95%-CI [0.77, 0.88]), and TotalSegmentator and RadYOLO produced few false positives per scan (both 0.85, 95%-CI [0.70, 1.02]). Conclusion: The lung nodules detected by four deep learning models showed high variation and a lack of consensus on clinical PCCT data. These findings highlight the importance of model benchmarking and local validation when deploying models in real-world clinical workflows.

Computed tomography (CT) interpretation requires recognition of the anatomical level of each slice. Classification may facilitate preprocessing and AI-assisted fracture diagnosis. We developed deep learning models to classify craniofacial CT slices into six anatomical categories. A total of 11,769 images were collected from a multicenter database involving 10 Japanese university hospitals and the public Head-Neck-Radiomics-HN1 dataset. Patients were divided into training (70%) and evaluation (30%) sets. Five CNN architectures were trained from scratch, and hard- and soft-voting ensembles were evaluated. Individual-model accuracy ranged from 94.72% to 96.09%; the hard-voting ensemble achieved 96.46% accuracy (Cohen's k = 0.954). Although performance differed among individual models (Cochran's Q = 99.29, p < 0.001), the ensemble did not significantly outperform EfficientNetV2B1 (McNemar's p = 0.163). Errors occurred mainly between adjacent anatomical regions. Deep learning can classify craniofacial CT slices accurately and may support automated dataset organization and AI-assisted fracture diagnosis. External validation is needed.

Objectives: Intraoperative ultrasound (IOUS) is an increasingly valuable real-time imaging modality in neurosurgery, enabling visualization of brain parenchyma, tumor margins, ventricular anatomy, and vascular structures during cranial procedures. Despite its clinical relevance, the development of artificial intelligence (AI)-assisted interpretation tools for neurosurgical IOUS has been limited by the scarcity of large, annotated datasets. Manual pixel-level annotation of ultrasound video frames is labor-intensive, requires specialized neuroanatomical expertise, and scales poorly. To address this bottleneck, we developed a proof-of-concept interactive semi-automated annotation pipeline designed to explore the feasibility of scalable, expert-guided labeling for intraoperative neurosurgical IOUS. Methods: We implemented the system as a napari-based graphical interface coupled with a scikit-learn Random Forest (RF) pixel classifier. Real intraoperative neurosurgical ultrasound videos from five craniotomy cases were used as training material. A neurosurgical expert annotated a minimal set of representative frames using brush-based labeling to define anatomical classes of interest. For each labeled pixel, a feature vector was computed comprising raw intensity, multi-scale Gaussian filtering ({sigma} = 1, 2, 3), Laplacian of Gaussian, and Sobel edge magnitude. The RF classifier was trained on accumulated annotations using class-balanced weighting and out-of-bag (OOB) validation, then applied across full video sequences. An incremental accumulator module enabled multi-case training. Segmentation outputs were exported as TIFF overlays for expert quality review. Results: Across five intraoperative cases, sparse expert annotation of a small number of representative frames per case was sufficient to train case-specific Random Forest classifiers that propagated labels across the full video sequences. Out-of-bag estimation served as an internal coherence check during training, and the classifier additionally produced per-class probability maps that exposed regions of uncertainty at acoustic boundaries. On expert qualitative review, the auto-segmented frames were anatomically coherent, with the classifier reliably distinguishing anechoic fluid compartments from echogenic tissue interfaces; residual misclassification was concentrated in low-contrast boundary regions. Conclusions: This proof-of-concept study demonstrates that semi-automated, expert-guided annotation of intraoperative neurosurgical ultrasound is technically feasible using classical machine learning. By combining neurosurgical domain expertise with an interactive, incremental annotation workflow applied to real intraoperative acquisitions, this framework represents a foundational step toward scalable generation of labeled datasets for neurosurgical IOUS. Future directions include formal multi-rater validation, expanded multi-class anatomical labeling, and eventual migration to deep learning architectures as annotated datasets grow.

Purpose: to introduce a novel magnetic resonance imaging (MRI) technique, MUSLIS MRI (muscle and lipid suppression), that generates two co-registered 3D sets of images in a single acquisition: one set with lipid signal suppression (MUSLIS1) and a second set with muscle signal suppression (MUSLIS2). Materials and Methods: Numerical simulations based on Bloch equations were performed to optimize the sequence parameters; the MP2RAGE sequence was employed for the implementation of MUSLIS imaging. MRI experiments were performed in vivo at 3 T. Quantitative measurements of signal contrast were carried out on MUSLIS1 and MUSLIS2. Various combination images and T1 parametric maps of the thighs were generated from the two co-registered 3D sets. Results and Discussion: Optimal sequence parameters to achieve lipid suppression in one dataset (MUSLIS1) and muscle suppression in the other dataset (MUSLIS2) were determined by numerical simulations. MUSLIS imaging was validated in vivo at 3 T: MUSLIS1 images and MUSLIS2 images displayed excellent lipid and muscle suppression, respectively. The values of the signal contrast between subcutaneous lipids and muscle, as well as between bone marrow lipids and muscle reached values close to 0.8. Thus, MUSLIS would allow for tissue characterization of lesion/masses and direct visualization of the adipose tissue. Combination images, generated from these two intrinsically co-registered sets of images (MUSLIS1 and MUSLIS2), displayed different contrasts and improved signal characteristics, such as mitigation of signal inhomogeneities. The development of MUSLIS imaging for musculoskeletal imaging followed the concept introduced with the FLAWS, an MRI sequence designed for brain MR imaging to generate two 3D sets of images, one with white matter suppression and the other with cerebrospinal fluid suppression. Similarly to FLAWS, which provides T1 maps of the brain, MUSLIS imaging yields T1 parametric maps of the musculoskeletal system. Furthermore, the unique characteristics of the combination images make MUSLIS imaging a potentially attractive technique at field strengths higher than 3 T. Conclusion: The proof-of-concept of a novel approach to musculoskeletal imaging, MUSLIS MRI, is proposed here and demonstrated by numerical simulations and MRI experiments performed in vivo at 3 T. Keywords: MRI, muscle, lipids, musculoskeletal imaging, high field MRI, fat suppression, FLAWS MRI, MP2RAGE, MUSLIS, inversion recovery, adipose tissue, bone marrow, subcutaneous fat, T1 mapping, 3 T, 7 T, neuromuscular disorders, metabolic diseases, muscular dystrophy.

Targeted alpha therapy with 225Ac is a promising modality for cancer treatment, as shown in recent preclinical and clinical studies. Preclinical imaging of 225Ac is essential to fully understand the biokinetics of developmental 225Ac radiopharmaceuticals. Although single photon emission computed tomography (SPECT) is typically used to achieve this in vivo, conventional preclinical scanners struggle to provide high contrast and accurate quantification for 225Ac imaging due to the very low injected activities (below 1 kBq/g), the unfavorable gamma-ray branching ratios (<26%), and its complex decay chain with several gamma-ray emissions. In this work, we employ a compact CZT-based Compton camera to achieve quantitative tomography of 225Ac daughters in live mice at 18.5 kBq (0.5 Ci) injected activities. Three mice bearing 22Rv1 prostate cancer subcutaneous xenografts were scanned in a single bed position for 30 minutes following intravenous administration of 18.5 kBq of the Anti-CD46 225Ac-Macropa-PEG4-YS5 conjugate. We successfully visualized the in vivo biodistributions of the 225Ac daughters 221Fr and 213Bi, distinguishing between the tumor xenograft and the central organs. Quantification of the tumor uptake from the images revealed an activity as low as 1.1 kBq. In vivo activity quantification computed from the images was compared to ex vivo biodistribution measurements using a gamma counter after harvesting the tissue, showing strong agreement.