Voxel Dimensions Research Articles

Automatic Adaptive Algorithm for Delineation of Cerebral-Spinal Fluid Regions for Non-contrast Magnetic Resonance Imaging Volumetry and Cisternography in Mice.

Magnetic resonance imaging (MRI) is an invaluable method of choice for anatomical and functional in vivo imaging of the brain. Still, accurate delineation of the brain structures remains a crucial task of MR image evaluation. This study presents a novel analytical algorithm developed in MATLAB for the automatic segmentation of cerebrospinal fluid (CSF) spaces in preclinical non-contrast MR images of the mouse brain. The algorithm employs adaptive thresholding and region growing to accurately and repeatably delineate CSF space regions in 3D constructive interference steady-state (3D-CISS) images acquired using a 9.4 Tesla MR system and a cryogenically cooled transmit/receive resonator. Key steps include computing a bounding box enclosing the brain parenchyma in three dimensions, applying an adaptive intensity threshold, and refining CSF regions independently in sagittal, axial, and coronal planes. In its original application, the algorithm provided objective and repeatable delineation of CSF regions in 3D-CISS images of sub-optimal signal-to-noise ratio, acquired with (33 μm)3 isometric voxel dimensions. It allowed revealing subtle differences in CSF volumes between aquaporin-4-null and wild-type littermate mice, showing robustness and reliability. Despite the increasing use of artificial neural networks in image analysis, this analytical approach provides robustness, especially when the dataset is insufficiently small and limited for training the network. By adjusting parameters, the algorithm is flexible for application in segmenting other types of anatomical structures or other types of 3D images. This automated method significantly reduces the time and effort compared to manual segmentation and offers higher repeatability, making it a valuable tool for preclinical and potentially clinical MRI applications. Key features • This protocol presents a fully automatic adaptive algorithm for the delineation of CSF space regions in 3D-CISS in vivo images of the mouse brain. • The algorithm represents an analytical method for adaptive CSF regions separation based on cumulative distribution of brain image intensities and contrast calculation-based slice-wise region growing. • Users can interactively alter the input parameters to modify the algorithm's output in a variety of 3D brain MR and μCT or CT images. • The algorithm is implemented in MATLAB 2021a and is compatible with all versions up to 2024a.

The conversion of electroanatomic maps for compatibility with treatment planning systems in cardiac radioablation target volume definition.

Cardiac radioablation (CRA) is a new and promising treatment modality for patients with ventricular tachycardia refractory to standard-of-care treatment. Electroanatomic maps are used to define radiation target volumes; however, there is currently no native method to import electroanatomic maps into the treatment planning system (TPS). To develop Edico, a semi-automated tool to enable electroanatomic map import into a TPS, by converting electroanatomic maps to a Digital Imaging and COmmunications in Medicine (DICOM) standard. The overall aim is to facilitate target volume delineation and improve workflow efficiency in treating patients. Edico imports voltage and spatial data from electroanatomic maps and sorts these into voxels to be exported in a DICOM format, with each voxel containing the average voltage value of the data that falls within it. Three different rectangular electroanatomic maps were created and processed using Edico to ensure that expected features are maintained through processing. A sensitivity analysis of voxel size was completed using 19 different electroanatomic maps processed at five different sets of voxel dimensions, for a total of 95 resulting voxelized datasets. The coefficient of variation in each populated voxel in the datasets was analyzed to determine which voxel sizes are necessary to ensure that data loss is kept to a minimum throughout processing, despite averaging. Five electroanatomic maps were used to re-contour clinical target volumes and planning target volumes for previously-treated patients with their electroanatomic maps now directly registered to their planning computed tomography (CT) scans. All three rectangular test electroanatomic maps were processed as expected. All tested voxel sizes resulted in low coefficients of variation overall, with the exception of the largest voxel size of 1.8×1.8×8mm. When using Edico, a user should choose voxel dimensions similar to or smaller than those of a planning CT. Of five pairs of clinical and planning target volumes from previously treated patients, adjustments were made to four (80%), retrospectively, using the electroanatomic maps generated using Edico, registered to the patients' planning CTs. Edico provides a reliable solution for electroanatomic map import into a TPS and facilitates clinical and planning target volume identification in CRA.

FEBio FINESSE: An Open-Source Finite Element Simulation Approach to Estimate In Vivo Heart Valve Strains Using Shape Enforcement.

Finite element simulations are an enticing tool to evaluate heart valve function; however, patient-specific simulations derived from 3D echocardiography are hampered by several technical challenges. The objective of this work is to develop an open-source method to enforce matching between finite element simulations and in vivo image-derived heart valve geometry in the absence of patient-specific material properties, leaflet thickness, and chordae tendineae structures. We evaluate FEBio Finite Element Simulations with Shape Enforcement (FINESSE) using three synthetic test cases considering a range of model complexity. FINESSE is then used to estimate the in vivo valve behavior and leaflet strains for three pediatric patients. Our results suggest that FINESSE can be used to enforce finite element simulations to match an image-derived surface and estimate the first principal leaflet strains within strain. Key considerations include: (i) defining the user-defined penalty, (ii) omitting the leaflet commissures to improve simulation convergence, and (iii) emulating the chordae tendineae behavior via prescribed leaflet free edge motion or a chordae emulating force. In all patient-specific cases, FINESSE matched the target surface with median errors of approximately the smallest voxel dimension. Further analysis revealed valve-specific findings, such as the tricuspid valve leaflet strains of a 2-day old patient with HLHS being larger than those of two 13-year old patients. FEBio FINESSE can be used to estimate patient-specific in vivo heart valve leaflet strains. The development of this open-source pipeline will enable future studies to begin linking in vivo leaflet mechanics with patient outcomes.

MISNeR: Medical Implicit Shape Neural Representation for Image Volume Visualisation

AbstractThree‐dimensional visualisation of mesh reconstruction of the medical images is commonly used for various clinical applications including pre / post‐surgical planning. Such meshes are conventionally generated by extracting the surface from volumetric segmentation masks. Therefore, they have inherent limitations of staircase artefacts due to their anisotropic voxel dimensions. The time‐consuming process for manual refinement to remove artefacts and/or the isolated regions further adds to these limitations. Methods for directly generating meshes from volumetric data by template deformation are often limited to simple topological structures, and methods that use implicit functions for continuous surfaces, do not achieve the level of mesh reconstruction accuracy when compared to segmentation‐based methods. In this study, we address these limitations by combining the implicit function representation with a multi‐level deep learning architecture. We introduce a novel multi‐level local feature sampling component which leverages the spatial features for the implicit function regression to enhance the segmentation result. We further introduce a shape boundary estimator that accelerates the explicit mesh reconstruction by minimising the number of the signed distance queries during model inference. The result is a multi‐level deep learning network that directly regresses the implicit function from medical image volumes to a continuous surface model, which can be used for mesh reconstruction from arbitrary high volume resolution to minimise staircase artefacts. We evaluated our method using pelvic computed tomography (CT) dataset from two public sources with varying z‐axis resolutions. We show that our method minimised the staircase artefacts while achieving comparable results in surface accuracy when compared to the state‐of‐the‐art segmentation algorithms. Furthermore, our method was 9 times faster in volume reconstruction than comparable implicit shape representation networks.

Effective and efficient coded aperture cone-beam computed tomography via generative adversarial U-Net

Coded aperture cone-beam computed tomography (CBCT) represents a crucial method for acquiring high-fidelity three-dimensional (3D) tomographic images while reducing radiation exposure. However, projections are non-uniformly and discontinuously sampled with the coded apertures placed in front of the x-ray source, leading to very small reconstruction scale and time-intensive iterations. In this study, an alternative approach to reconstruct coded aperture CBCT based on generative adversarial U-net is proposed to effectively and efficiently reconstruct large scale 3D CBCT images. Our method entails predicting complete and uniform projections from incomplete and non-uniform coded projections, enabling the requirement of continuity for the use of analytical algorithms in 3D image reconstruction. This novel technique effectively mitigates the traditional trade-off between image fidelity and computational complexity inherent in conventional coded aperture CBCT reconstruction methods. Our experimental results, conducted using clinical datasets comprising CBCT images from 102 patients at Nanjing Medical University, demonstrate that high-quality CBCT images with voxel dimensions of 400 × 400 × 400 can be reconstructed within 35 s, even when 95% of projections are blocked, yielding images with PSNR values exceeding 25dB and SSIM values surpassing 0.85.

A Pilot Clinical and Technical Validation of an Immersive Virtual Reality Platform for 3D Anatomical Modeling and Contouring in Support of Surgical and Radiation Oncology Treatment Planning.

The aim of this study was to validate a novel medical virtual reality (VR) platform used for medical image segmentation and contouring in radiation oncology and 3D anatomical modeling and simulation for planning medical interventions, including surgery. The first step of the validation was to verify quantitatively and qualitatively that the VR platform can produce substantially equivalent 3D anatomical models, image contours, and measurements to those generated with existing commercial platforms. To achieve this, a total of eight image sets and 18 structures were segmented using both VR and reference commercial platforms. The image sets were chosen to cover a broad range of scanner manufacturers, modalities, and voxel dimensions. The second step consisted of evaluating whether the VR platform could provide efficiency improvements for target delineation in radiation oncology planning. To assess this, the image sets for five pediatric patients with resected standard-risk medulloblastoma were used to contour target volumes in support of treatment planning of craniospinal irradiation, requiring complete inclusion of the entire cerebral-spinal volume. Structures generated in the VR and the commercial platforms were found to have a high degree of similarity, with dicesimilarity coefficient ranging from 0.963 to 0.985 for high-resolution images and 0.920 to 0.990 for lower resolution images. Volume, cross-sectional area, and length measurements were also found to be in agreement with reference values derived from a commercial system, with length measurements having a maximum difference of 0.22mm, angle measurements having a maximum difference of 0.04°, and cross-sectional area measurements having a maximum difference of 0.16 mm2. The VR platform was also found to yield significant efficiency improvements, reducing the time required to delineate complex cranial and spinal target volumes by an average of 50% or 29min.

Impact of Deep Learning Denoising Algorithm on Diffusion Tensor Imaging of the Growth Plate on Different Spatial Resolutions.

To assess the impact of a deep learning (DL) denoising reconstruction algorithm applied to identical patient scans acquired with two different voxel dimensions, representing distinct spatial resolutions, this IRB-approved prospective study was conducted at a tertiary pediatric center in compliance with the Health Insurance Portability and Accountability Act. A General Electric Signa Premier unit (GE Medical Systems, Milwaukee, WI) was employed to acquire two DTI (diffusion tensor imaging) sequences of the left knee on each child at 3T: an in-plane 2.0 × 2.0 mm2 with section thickness of 3.0 mm and a 2 mm3 isovolumetric voxel; neither had an intersection gap. For image acquisition, a multi-band DTI with a fat-suppressed single-shot spin-echo echo-planar sequence (20 non-collinear directions; b-values of 0 and 600 s/mm2) was utilized. The MR vendor-provided a commercially available DL model which was applied with 75% noise reduction settings to the same subject DTI sequences at different spatial resolutions. We compared DTI tract metrics from both DL-reconstructed scans and non-denoised scans for the femur and tibia at each spatial resolution. Differences were evaluated using Wilcoxon-signed ranked test and Bland-Altman plots. When comparing DL versus non-denoised diffusion metrics in femur and tibia using the 2 mm × 2 mm × 3 mm voxel dimension, there were no significant differences between tract count (p = 0.1, p = 0.14) tract volume (p = 0.1, p = 0.29) or tibial tract length (p = 0.16); femur tract length exhibited a significant difference (p < 0.01). All diffusion metrics (tract count, volume, length, and fractional anisotropy (FA)) derived from the DL-reconstructed scans, were significantly different from the non-denoised scan DTI metrics in both the femur and tibial physes using the 2 mm3 voxel size (p < 0.001). DL reconstruction resulted in a significant decrease in femorotibial FA for both voxel dimensions (p < 0.01). Leveraging denoising algorithms could address the drawbacks of lower signal-to-noise ratios (SNRs) associated with smaller voxel volumes and capitalize on their better spatial resolutions, allowing for more accurate quantification of diffusion metrics.

Performance Evaluation of Deformable Image Registration Systems - SmartAdapt® and Velocity™.

To commission and validate commercial deformable image registration (DIR) systems (SmartAdapt® and Velocity™) using task group 132 (TG-132) digital phantom datasets. Additionally, the study compares and verifies the DIR algorithms of the two systems. TG-132 digital phantoms were obtained from the American Association of Physicists in Medicine website and imported into SmartAdapt® and Velocity™ systems for commissioning and validation. The registration results were compared with known shifts using rigid registrations and deformable registrations. Virtual head and neck phantoms obtained online (DIR Evaluation Project) and some selected clinical data sets from the department were imported into the two DIR systems. For both of these datasets, DIR was carried out between the source and target images, and the contours were then propagated from the source to the target image data set. The dice similarity coefficient (DSC), mean distance to agreement (MDA), and Jacobian determinant measures were utilised to evaluate the registration results. The recommended criteria for commissioning and validation of DIR system from TG-132 was error <0.5*voxel dimension (vd). Translation only registration: Both systems met TG-132 recommendations except computed tomography (CT)-positron emission tomography registration in both systems (Velocity ~1.1*vd, SmartAdapt ~1.6*vd). Translational and rotational registration: Both systems failed the criteria for all modalities (For velocity, error ranged from 0.6*vd [CT-CT registration] to 3.4*vd [CT-cone-beam CT (CBCT) registration]. For SmartAdapt® the range was 0.6*vd [CT-CBCT] to 3.6*vd [CT-CT]). Mean ± standard deviation for DSC, MDA and Jacobian metrics were used to compare the DIR results between SmartAdapt® and Velocity™. The DIR algorithms of SmartAdapt® and Velocity™ were commissioned and their deformation results were compared. Both systems can be used for clinical purpose. While there were only minimal differences between the two systems, Velocity™ provided lower values for parotids, bladder, rectum, and prostate (soft tissue) compared to SmartAdapt. However, for mandible, spinal cord, and femoral heads (rigid structures), both systems showed nearly identical results.

Accuracy of automated segmentation and volumetry of acute intracerebral hemorrhage following minimally invasive surgery using a patch-based convolutional neural network in a small dataset

PurposeIn cases of acute intracerebral hemorrhage (ICH) volume estimation is of prognostic and therapeutic value following minimally invasive surgery (MIS). The ABC/2 method is widely used, but suffers from inaccuracies and is time consuming. Supervised machine learning using convolutional neural networks (CNN), trained on large datasets, is suitable for segmentation tasks in medical imaging. Our objective was to develop a CNN based machine learning model for the segmentation of ICH and of the drain and volumetry of ICH following MIS of acute supratentorial ICH on a relatively small dataset.MethodsNinety two scans were assigned to training (n = 29 scans), validation (n = 4 scans) and testing (n = 59 scans) datasets. The mean age (SD) was 70 (± 13.56) years. Male patients were 36. A hierarchical, patch-based CNN for segmentation of ICH and drain was trained. Volume of ICH was calculated from the segmentation mask.ResultsThe best performing model achieved a Dice similarity coefficient of 0.86 and 0.91 for the ICH and drain respectively. Automated ICH volumetry yielded high agreement with ground truth (Intraclass correlation coefficient = 0.94 [95% CI: 0.91, 0.97]). Average difference in the ICH volume was 1.33 mL.ConclusionUsing a relatively small dataset, originating from different CT-scanners and with heterogeneous voxel dimensions, we applied a patch-based CNN framework and successfully developed a machine learning model, which accurately segments the intracerebral hemorrhage (ICH) and the drains. This provides automated and accurate volumetry of the bleeding in acute ICH treated with minimally invasive surgery.

Hybrid physics-guided data-driven modeling for generalizable geometric accuracy prediction and improvement in two-photon lithography

Two-photon lithography (TPL) is an additive manufacturing technique to produce three-dimensional (3D) micro- and nano-scale structures. Geometric accuracy is of vital importance to ensure the quality and functionality of additively manufactured 3D structures. However, there exists limited research on modeling and improving the geometric accuracy of TPL. Moreover, there is a lack of understanding of the influence of TPL process parameters (e.g., scanning rate, laser power) and structure design on the resulting geometric accuracy. This paper presents a new hybrid physics-guided data-driven modeling framework for generalizable cross-design and cross-parameter geometric accuracy prediction and improvement. The proposed model decomposes TPL-produced geometric features into two levels: a global trend and a spatial trend. Guided by physical laws that govern voxel dimensions, we construct the global trend as a function of TPL process parameters. Our new findings reveal strong linear relationships among the parameters of the global trend model across structure design dimensions, which enable generalizable predictive modeling. Moreover, the spatial trend captures the spatially varying patterns in geometric features. A large-scale experimental design consisting of six hemisphere sizes and six parameter combinations is carried out to thoroughly test the effectiveness of the proposed method. Our modeling approach demonstrates high accuracy in predicting the geometric error of a target design with knowledge derived from other designs and parameters. The average prediction errors for radius and height are 5.23% and 4.66%, respectively. Furthermore, the proposed compensation strategy is shown to reduce the geometric errors from 22.19% to 3.21% for radius, and from 12.18% to 4.96% for height. Meanwhile, the within-sample variations are greatly reduced, indicating improved process consistency. To the best of our knowledge, this study is among the first to develop a generalizable method for cross-design and cross-parameter geometric accuracy modeling and improvement in TPL.

Predicted effects of image acquisition and analysis conditions on DTMRI tractography-based muscle architecture estimates.

To quantify the effects of the intrinsic signal pattern, image acquisition conditions, and data analysis conditions on diffusion-tensor MRI (DTMRI) tractography-based muscle architecture estimates using a sampling-reconstruction assessment framework. Numerical models of muscles were constructed with realistic architectural properties. DTMRI signals were computed at signal-to-noise ratio (SNR) of 24-96 and common voxel sizes. Fiber tracking was performed, and the results were compared with the known architectural properties. SNR exerted the most significant impact on the outcome. The outcome variables approached asymptotes at SNR ≈ 54. Large in-plane voxel dimensions reduced the similarity between reconstructed fibers and the known architectural properties. Higher order polynomials helped reconstruct fibers with more complicated geometry but overfit noise for less complex geometries. The intrinsic fiber curvature also affected the robustness of polynomial smoothing to SNR. Other conditions, such as the fiber dimensionality, voxel aspect ratio, and slice thickness, did not affect the outcomes. SNR ≥ 54 is recommended for accurate muscle architecture characterization using DTMRI. Averaged across all simulated conditions, the greatest percent errors under SNR = 54 were -5.6% and -4.0% for the pennation angle and fiber-tract length estimates, respectively. For fiber tracts with intermediate intrinsic curvature, the greatest percent error for the curvature estimate was 9.8% for SNR = 54. Smaller in-plane voxel size (≤1.5 mm) is preferred to minimize the estimation error in architectural properties. If necessary, slice thickness may be adjusted within typical ranges to achieve sufficient SNR when slices are aligned near the fiber direction. Third-order polynomial fitting is appropriate for smoothing fiber tracts.

Volumetric MRI with sparse sampling for MR-guided 3D motion tracking via sparse prior-augmented implicit neural representation learning.

Volumetric reconstruction of magnetic resonance imaging (MRI) from sparse samples is desirable for 3D motion tracking and promises to improve magnetic resonance (MR)-guided radiation treatment precision. Data-driven sparse MRI reconstruction, however, requires large-scale training datasets for prior learning, which is time-consuming and challenging to acquire in clinical settings. To investigate volumetric reconstruction of MRI from sparse samples of two orthogonal slices aided by sparse priors of two static 3D MRI through implicit neural representation (NeRP) learning, in support of 3D motion tracking during MR-guided radiotherapy. A multi-layer perceptron network was trained to parameterize the NeRP model of a patient-specific MRI dataset, where the network takes 4D data coordinates of voxel locations and motion states as inputs and outputs corresponding voxel intensities. By first training the network to learn the NeRP of two static 3D MRI with different breathing motion states, prior information of patient breathing motion was embedded into network weights through optimization. The prior information was then augmented from two motion states to 31 motion states by querying the optimized network at interpolated and extrapolated motion state coordinates. Starting from the prior-augmented NeRP model as an initialization point, we further trained the network to fit sparse samples of two orthogonal MRI slices and the final volumetric reconstruction was obtained by querying the trained network at 3D spatial locations. We evaluated the proposed method using 5-min volumetric MRI time series with 340ms temporal resolution for seven abdominal patients with hepatocellular carcinoma, acquired using golden-angle radial MRI sequence and reconstructed through retrospective sorting. Two volumetric MRI with inhale and exhale states respectively were selected from the first 30s of the time series for prior embedding and augmentation. The remaining 4.5-min time series was used for volumetric reconstruction evaluation, where we retrospectively subsampled each MRI to two orthogonal slices and compared model-reconstructed images to ground truth images in terms of image quality and the capability of supporting 3D target motion tracking. Across the seven patients evaluated, the peak signal-to-noise-ratio between model-reconstructed and ground truth MR images was 38.02±2.60dB and the structure similarity index measure was 0.98±0.01. Throughout the 4.5-min time period, gross tumor volume (GTV) motion estimated by deforming a reference state MRI to model-reconstructed and ground truth MRI showed good consistency. The 95-percentile Hausdorff distance between GTV contours was 2.41±0.77mm, which is less than the voxel dimension. The mean GTV centroid position difference between ground truth and model estimation was less than 1mm in all three orthogonal directions. A prior-augmented NeRP model has been developed to reconstruct volumetric MRI from sparse samples of orthogonal cine slices. Only one exhale and one inhale 3D MRI were needed to train the model to learn prior information of patient breathing motion for sparse image reconstruction. The proposed model has the potential of supporting 3D motion tracking during MR-guided radiotherapy for improved treatment precision and promises a major simplification of the workflow by eliminating the need for large-scale training datasets.

Intra voxel analysis in magnetic resonance imaging via deep learning

AbstractMagnetic Resonance Imaging (MRI) is a useful diagnostic method for producing anatomical images of the body. It differentiates tissues thanks to the measure of their proton density and relaxation times and . Since pathological tissues often present altered , and with respect to the physiological ones, MRI is largely used for the detection of large number of conditions. In addition, MRI scan presents an effective resolution able to detect very small anatomical elements, making the imaging system well suitable in contexts like prevention and early diagnosis. Nevertheless, system resolution imposes pathological volume to be larger than the voxel dimension. Since in some pathologies and conditions it could be very helpful to find tissues even smaller than the voxel dimension, this manuscript proposes an algorithm able to analyze voxel content and detect which one presents more than one tissue in its volume (i.e., which one is heterogeneous). More in detail, a machine learning algorithm is proposed, able to highlight, in the MR image, which pixel corresponds to an heterogeneous voxel. Method shows to be promising in combining good results and near real‐time processing in both simulated and real scenario.

The Microvascular Network Density and Morphology Affect the Radiotherapy Outcome at the Microscale Level: A Computational Analysis.

Half of the cancer patients are treated with radiotherapy to obtain tumor control while minimizing radiation toxicity. In this context, hypoxia is known to determine treatment resistance, driving tumor relapse after treatment. Therefore, we investigated the role of microvasculature density and morphology in shaping the tissue oxygen distribution and consequently affecting the treatment outcome. We developed an advanced computational model to describe oxygen delivery in the vascular network and the surrounding tissue. A peculiar aspect of the model is its mesoscale mixed-dimensional approach, which allows the explicit inclusion of vascular network geometry and the description of the red blood cells' effect. The oxygen delivery is modeled considering both diffusive and convective phenomena. A 30 × 2GyRBE treatment is delivered in silico simulating photons, protons, and carbon ions. First, we estimated the surviving fraction by the classical linear quadratic model modified to account for the oxygen effect. Then, leveraging the 3D description of the surviving fraction at the microenvironment scale, we compute the local tumor control probability (LTCP) in different oxygenation states (reference, acute hypoxia, high oxygen consumption). We report correlations between the LTCP and the hypoxic volume fractions (with pO2 lower than 1 mmHg) starting with photons. These hypoxic regions are present locally, even in highly vascularized tissue, if the network is not uniformly distributed, as it might be in cancer. They are also present in tissue with low microvascular density, even with regular morphology. Interestingly, the domain considered is comparable to or smaller than the clinical imaging standard voxel dimension, and the average oxygen partial pressure in the tissue region fails to spot treatments with low LTCP, questioning whether these hypoxia areas are visible clinically via imaging. Protons have a similar effect, highlighting a similar behavior across the oxygenation levels at the microscale level. Finally, carbon ions seem more effective than photons and protons in the presence of hypoxia due to the lower oxygen effect at high LET. For this reason, the treatment with carbon ions results in high LTCPs, whatever vascular network is considered (density and morphology differences). These results show the effect of microvascular density and regularity on the radiotherapy outcome and they help us understand how the microvascular network morphology affects tumor oxygenation and the radiotherapy outcome.

Real Time Volumetric MRI for MR-Guided 3D Motion Tracking via Sparse Prior-Augmented Neural Representation Learning

To reconstruct volumetric MRI from orthogonal cine acquisition aided by sparse priors of 2 static 3D MRI through implicit neural representation (NeRP) learning, with the goal of eliminating large-scale training datasets for data-driven sparse MRI reconstruction and supporting clinical workflow of real time 3D motion tracking during MR-guided radiotherapy. A multi-layer perceptron network was trained to learn the NeRP of a patient-specific MRI dataset, where the network takes 4D data coordinates of voxel locations and motion states as inputs and outputs corresponding voxel intensities. By first learning the NeRP of 2 static 3D MRI with different breathing motion states, prior knowledge of patient breathing motion was embedded into network weights through optimization. The prior knowledge was then augmented from 2 to 31 motion states by querying the optimized network at interpolated/extrapolated motion state coordinates. Starting from the prior-augmented network as an initialization point, the network was further trained using sparse samples of 2 orthogonal cine slices. The final volumetric reconstruction was obtained by querying the trained network at desired 3D spatial locations. We evaluated the proposed method using 5-minute volumetric MRI time series with 340 ms temporal resolution collected from 7 liver carcinoma patients. The time series was acquired using golden-angle radial MRI sequence and reconstructed through retrospective sorting. Two MRI with inhale and exhale states respectively were selected from the first 30 sec of the time series for prior embedding and augmentation. The remaining 4.5-min time series was used for volumetric reconstruction evaluation, where we retrospectively subsampled each MRI to 2 orthogonal slices and compared network-reconstructed images to ground truth images in terms of image quality and the capability of supporting 3D target motion tracking. Across the 7 patients evaluated, the peak signal to noise ratio between model reconstruction and ground truth was 54.66 ± 6.16 dB and the structural similarity index measure was 0.99 ± 0.01. Gross tumor volume (GTV) contours estimated by deforming a reference state MRI to model-reconstructed and ground truth MRI showed good consistency. The 95-percentile Hausdorff distance between GTV contours was 1.89 ± 1.13 mm, which is less than the voxel dimension. The mean GTV centroid position difference between ground truth and model estimation was less than 1 mm in all 3 orthogonal directions. Volumetric MRI from orthogonal cine acquisition with sparse priors is feasible by modeling prior knowledge through implicit neural representation learning. The model-reconstructed images showed sufficient accuracy in supporting 3D motion tracking of abdominal targets. By eliminating the need for large scale training datasets, the method promises to enable clinical implementation of 3D motion tracking for precision radiation therapy.

Updating 90Y Voxel S-Values including internal Bremsstrahlung: Monte Carlo study and development of an analytical model

PurposeInternal Bremsstrahlung (IB) is a process accompanying β-decay but neglected in Voxel S-Values (VSVs) calculation. Aims of this work were to calculate, through Monte Carlo (MC) simulation, updated 90Y-VSVs including IB, and to develop an analytical model to evaluate 90Y-VSVs for any voxel size of practical interest. MethodsGATE (Geant4 Application for Tomographic Emission) was employed for simulating voxelized geometries of soft tissue, with voxels sides l ranging from 2 to 6 mm, in steps of 0.5 mm. The central voxel was set as a homogeneous source of 90Y when IB photons are not modelled. For each l, the VSVs were computed for 90Y decays alone and for 90Y + IB. The analytical model was then built through fitting procedures of the VSVs including IB contribution. ResultsComparing GATE-VSVs with and without IB, differences between + 25% and + 30% were found for distances from the central voxel larger than the maximum β-range. The analytical model showed an agreement with MC simulations within ± 5% in the central voxel and in the Bremsstrahlung tails, for any l value examined, and relative differences lower than ± 40%, for other distances from the source. ConclusionsThe presented 90Y-VSVs include for the first time the contribution due to IB, thus providing a more accurate set of dosimetric factors for three-dimensional internal dosimetry of 90Y-labelled radiopharmaceuticals and medical devices. Furthermore, the analytical model constitutes an easy and fast alternative approach for 90Y-VSVs estimation for non-standard voxel dimensions.

3-D joint space mapping at the ankle from weight-bearing CT: reproducibility, repeatability, and challenges for standardisation

ObjectivesWe present a 3-D approach to joint space width (JSW) measurement across the ankle from weight-bearing CT (WBCT) to demonstrate inter-operator reproducibility, test-retest repeatability, and how differences in angulation affect ankle JSW distribution.MethodsOne side from repeat WBCT imaging of both feet and ankles was analysed from 23 individuals as part of their routine clinical care pathway. Joint space mapping was performed at four facets across the talus: talonavicular, talar dome and medial gutter (dome-medial), lateral gutter, and posterior subtalar. Inter-operator reproducibility was calculated for two users, while test-retest repeatability was calculated by comparing the two visits, both presented as Bland-Altman statistics. Statistical parametric mapping determined any significant relationships between talocrural joint space angulation and 3-D JSW distribution.ResultsThe average ± standard deviation interval between imaging was 74.0 ± 29.6 days. Surface averaged bias ± limits of agreement were similar for reproducibility and repeatability, the latter being: talonavicular 0.01 ± 0.26 mm, dome-medial 0.00 ± 0.28 mm, lateral gutter − 0.02 ± 0.40 mm, and posterior subtalar 0.02 ± 0.34 mm. Results are presented as 3-D distribution maps, with optimum test–retest repeatability reaching a smallest detectable difference of ± 0.15 mm.ConclusionsJoint space mapping is a robust approach to 3-D quantification of JSW measurement, inter-operator reproducibility, and test–retest repeatability at the ankle, with sensitivity reaching a best value of ± 0.15 mm. Standardised imaging protocols and optimised metal artefact reduction will be needed to further understand the clinical value of these 3-D measures derived from WBCT.Clinical relevance statementWeight-bearing computed tomography is an increasingly important tool in the clinical assessment of orthopaedic ankle disorders. This paper establishes the performance of measuring 3-D joint space width using this technology, which is an important surrogate marker for severity of osteoarthritis.Key Points• Joint space width values and error metrics from across the ankle measured from weight-bearing CT can be presented as 3-D maps that show topographic variation.• The best sensitivity for detecting meaningful change in 3-D joint space width at the ankle was ± 0.15 mm, a value less than the isotropic imaging voxel dimensions.• Standardised imaging protocols and optimised metal artefact reduction will be needed to understand the clinical value of 3-D measures from weight-bearing CT.

Single‐Step 3D Printing of Micro‐Optics with Adjustable Refractive Index by Ultrafast Laser Nanolithography

AbstractLaser exposure defines voxel's dimensions as essential building blocks in direct write 3D nanolithography. However, the exposure conditions not only influence the size of the produced features but also their optical properties. This empowers the realization of an adjustable refractive index out of single material by varying the writing strategy while preserving laser 3D nanolithography's flexibility in geometry and high resolution. Here, the refractive index for the 450–1600 nm spectral range of the micro‐optics out of SZ2080 hybrid polymer is systematically studied by applying ray and wave optics approaches followed by optical resolution analysis. It reveals the exact value of the laser‐printed components instead of the determination assessed by other techniques measuring thin films or bulky volumes of the investigated substance. The studied micro‐lenses are of below 100 µm in dimensions and a clear distinction in their performance on low and high exposure doses is found by analyzing it in all different approaches and validating using different lithography setups. Findings reveal the complexity of the refractive index of the 3D micro‐optics which is influenced by the material density and morphology. A route for freedom in 3D printing shape and refractive index can be realized by the technological optimization of delicate exposure control in ultrafast laser nanolithography.

Technical Adequacy of Fully Automated Artificial Intelligence Body Composition Tools: Assessment in a Heterogeneous Sample of External CT Examinations.

BACKGROUND. Clinically usable artificial intelligence (AI) tools analyzing imaging studies should be robust to expected variations in study parameters. OBJECTIVE. The purposes of this study were to assess the technical adequacy of a set of automated AI abdominal CT body composition tools in a heterogeneous sample of external CT examinations performed outside of the authors' hospital system and to explore possible causes of tool failure. METHODS. This retrospective study included 8949 patients (4256 men, 4693 women; mean age, 55.5 ± 15.9 years) who underwent 11,699 abdominal CT examinations performed at 777 unique external institutions with 83 unique scanner models from six manufacturers with images subsequently transferred to the local PACS for clinical purposes. Three independent automated AI tools were deployed to assess body composition (bone attenuation, amount and attenuation of muscle, amount of visceral and sub-cutaneous fat). One axial series per examination was evaluated. Technical adequacy was defined as tool output values within empirically derived reference ranges. Failures (i.e., tool output outside of reference range) were reviewed to identify possible causes. RESULTS. All three tools were technically adequate in 11,431 of 11,699 (97.7%) examinations. At least one tool failed in 268 (2.3%) of the examinations. Individual adequacy rates were 97.8% for the bone tool, 99.1% for the muscle tool, and 98.9% for the fat tool. A single type of image processing error (anisometry error, due to incorrect DICOM header voxel dimension information) accounted for 81 of 92 (88.0%) examinations in which all three tools failed, and all three tools failed whenever this error occurred. Anisometry error was the most common specific cause of failure of all tools (bone, 31.6%; muscle, 81.0%; fat, 62.8%). A total of 79 of 81 (97.5%) anisometry errors occurred on scanners from a single manufacturer; 80 of 81 (98.8%) occurred on the same scanner model. No cause of failure was identified for 59.4% of failures of the bone tool, 16.0% of failures of the muscle tool, or 34.9% of failures of the fat tool. CONCLUSION. The automated AI body composition tools had high technical adequacy rates in a heterogeneous sample of external CT examinations, supporting the generalizability of the tools and their potential for broad use. CLINICAL IMPACT. Certain causes of AI tool failure related to technical factors may be largely preventable through use of proper acquisition and reconstruction protocols.

Acquisition and Analysis of Excised Neocortex from Pediatric Patients with Focal Cortical Dysplasia Using Mesoscale Diffusion MRI.

Non-invasive classification of focal cortical dysplasia (FCD) subtypes remains challenging from a radiology perspective. Quantitative imaging biomarkers (QIBs) have the potential to distinguish subtypes that lack pathognomonic features and might help in defining the extent of abnormal connectivity associated with each FCD subtype. A key motivation of diagnostic imaging is to improve the localization of a "lesion" that can guide the surgical resection of affected tissue, which is thought to cause seizures. Conversely, surgical resections to eliminate or reduce seizures provided unique opportunities to develop magnetic resonance imaging (MRI)-based QIBs by affording long scan times to evaluate multiple contrast mechanisms at the mesoscale (0.5 mm isotropic voxel dimensions). Using ex vivo hybrid diffusion tensor imaging on a 9.4 T MRI scanner, the grey to white matter ratio of scalar indices was lower in the resected middle temporal gyrus (MTG) of two neuropathologically confirmed cases of FCD compared to non-diseased control postmortem fixed temporal lobes. In contrast, fractional anisotropy was increased within FCD and also adjacent white matter tracts. Connectivity (streamlines/mm3) in the MTG was higher in FCD, suggesting that an altered connectivity at the lesion locus can potentially provide a tangible QIB to distinguish and characterize FCD abnormalities. However, as illustrated here, a major challenge for a robust tractographical comparison lies in the considerable differences in the ex vivo processing of bioptic and postmortem samples. Mesoscale diffusion MRI has the potential to better define and characterize epileptic tissues obtained from surgical resection to advance our understanding of disease etiology and treatment.