Physics in medicine and biology最新文献

Objective:the recently developed V79-RBE₁₀biological weighting function (BWF) model is a simple and robust tool for a fast relative biological effectiveness (RBE) assessment for comparing different exposure conditions in particle therapy. In this study, the RBE₁₀derived by this model (through the particle and heavy ion transport code system (PHITS) simulatedd(y)spectra) is compared with values of RBE₁₀using experimentally derivedd(y)spectra from a silicon-on-insulator (SOI) microdosimeter.Approach:experimentally measuredd(y)spectra are used to calculate an RBE₁₀value utilizing the V79-RBE₁₀BWF model as well as the modified microdosimetric kinetic model (MKM) to produce an RBE₁₀-vs-y_Dtrend for a wide range of ions. In addition, a beamline specific PHITS simulation was conducted which replicated the exact experimental conditions that were used with the SOI microdosimeter at the heavy ion medical accelerator in Chiba biological beamline with¹²C ions.Main Results:the RBE₁₀-vs-y_Dtrend for¹H,⁴He,⁷Li,¹²C,¹⁴N,¹⁶O,²⁰Ne,²⁸Si,⁵⁶Fe, and¹²⁴Xe ions is examined with good agreement found between the SOI microdosimeter derived RBE₁₀values with the V79-RBE₁₀BWF model and MKM, as well as the PHITS simulations for¹H,⁴He,⁷Li,¹²C,¹⁶O, and⁵⁶Fe ions while some discrepancies were seen for¹⁴N,²⁰Ne, and²⁸Si ions. Deviations have been attributed to the difference in the derivation of thed(y) spectra based on the different methods utilized. Good agreement was found betweeny_Dvalues and an over estimation was observed for RBE₁₀values for the beamline specific simulation of the¹²Cion beam.Significance:overall, this study shows that the SOI microdosimeter is a valuable tool that can be utilized for quick and accurate experimental derivation of thed(y) spectra, which can then be convoluted with the weighting function of the V79-RBE₁₀BWF model to derive RBE₁₀. The SOI microdosimeter is able to derive experimental values ofy_Dand RBE₁₀for various ions in any irradiation condition utilizing other radiobiological models.

Objective.Magnetic particle imaging (MPI) was introduced in 2005 as a promising, tracer-based medical imaging modality with the potential for high sensitivity and spatial resolution. Since then, numerous preclinical devices have been built but only a few human-scale devices, none of which targeted functional neuroimaging. In this work, we probe the challenges of scaling the technology to meet the needs of human functional neuroimaging with sufficient sensitivity for detecting the hemodynamic changes following brain activation with a spatio-temporal resolution comparable to current functional magnetic resonance imaging approaches.Approach.We built a human brain-scale MPI system using a mechanically-rotated, permanent-magnet-based field-free line (FFL) (1.1Tm-1) with a water-cooled, 26 kHz drive coil producing a field of up to 7 mTpeak, and receive coil that can fit over a human head. Images are acquired continuously at a temporal resolution of 5 s/image, controlled by in-house LabView-based acquisition software with online reconstruction. We used a dilution series to quantify the detection limit, a series of parallel-line phantoms to assess the spatial resolution, and a large 'G' shaped phantom to demonstrate the human-scale field of view (FOV).Main results.The imager has a sensitivity of about 1 µgFeover a 2D imaging FOV of 181 mm diameter(132 pixels) in a 5 s image. Depending on the image reconstruction used, the spatial resolution defined by 50% contrast between adjacent lines was 5-7 mm.Significance.This proof-of-concept system demonstrates a pathway for human MPI functional neuroimaging with the potential for an order of magnitude increase of sensitivity compared to the other human hemodynamic imaging methods. It demonstrates the successful transition of the FFL based MPI architecture from the rodent to human scale and identifies areas which could benefit from further work.

Objective. The primary purpose of this work is to demonstrate the feasibility of a deep convolutional neural network (dCNN) based algorithm that uses two-dimensional (2D) electronic portal imaging device (EPID) images and CT images as input to reconstruct 3D dose distributions inside the patient.Approach. To generalize dCNN training and testing data, geometric and materials models of a VitalBeam accelerator treatment head and a corresponding EPID imager were constructed in detail in the GPU-accelerated Monte Carlo dose computing software, ARCHER. The EPID imager pixel spatial resolution ranging from 1.0 mm to 8.5 mm was studied to select optimal pixel size for simulation. For purposes of training the U-Net-based dCNN, a total of 101 clinical intensive modulated radiation treatment cases-81 for training, 10 for validation, and 10 for testing-were simulated to produce comparative data of 3D dose distribution versus 2D EPID image data. The model's accuracy was evaluated by comparing its predictions with Monte Carlo dose.Main Results. Using the optimal EPID pixel size of 1.5 mm, it took about 18 min to simulate the particle transport in patient-specific CT and EPID imager per a single field. In contrast, the trained dCNN can predict 3D dose distributions in about 0.35 s. The average 3D gamma passing rates between ARCHER and predicted doses are 99.02 ± 0.57% (3%/3 mm) and 96.85 ± 1.22% (2%/2 mm) for accumulated fields, respectively. Dose volume histogram data suggest that the proposed dCNN 3D dose prediction algorithm is accurate in evaluating treatment goals.Significance. This study has proposed a novel deep-learning model that is accurate and rapid in predicting 3D patient dose from 2D EPID images. The computational speed is expected to facilitate clinical practice for EPID-basedin-vivopatient-specific quality assurance towards adaptive radiation therapy.

This paper introduces a novel unsupervised inverse-consistent diffeomorphic registration network termed IConDiffNet, which incorporates an energy constraint that minimizes the total energy expended during the deformation process. The IConDiffNet architecture consists of two symmetric paths, each employing multiple recursive cascaded updating blocks (neural networks) to handle different virtual time steps parameterizing the path from the initial undeformed image to the final deformation. These blocks estimate velocities corresponding to specific time steps, generating a series of smooth time-dependent velocity vector fields. Simultaneously, the inverse transformations are estimated by corresponding blocks in the inverse path. By integrating these series of time-dependent velocity fields from both paths, optimal forward and inverse transformations are obtained, aligning the image pair in both directions. We evaluate our proposed method on a 3D image registration task with a large-scale brain MRI image dataset containing 375 subjects. The proposed IConDiffNet achieves fast and accurate DIR with better Dice scores, lower Hausdorff distance metric, and lower total energy spent during the deformation in the test dataset compared to competing state-of-the-art DL-based diffeomorphic DIR methods. Visualization shows that IConDiffNet produces more complicated transformations that better align structures than the VoxelMoprh-Diff, SYMNet, and ANTs-SyN methods. The proposed IConDiffNet represents an advancement in unsupervised deep-learning-based DIR approaches. By ensuring inverse consistency and diffeomorphic properties in the outcome transformations, IConDiffNet offers a pathway for improved registration accuracy, particularly in clinical settings where diffeomorphic properties are crucial. Furthermore, the generality of IConDiffNet's network structure supports direct extension to diverse 3D image registration challenges. This adaptability is facilitated by the flexibility of the objective function used in optimizing the network, which can be tailored to suit different registration tasks.

This study analyzed the spectral response of EBT3, EBT4, and EBT-XD radiochromic films using absorption spectroscopy. The primary focus was on characterizing the evolution of spectral signatures across a range of absorbed doses, thereby elucidating the unique dose-dependent response profiles of each film type. Ten samples of each film type were subjected to open field irradiation within their designated dose ranges (1-20 Gy for EBT3 and EBT4, 1-50 Gy for EBT-XD). The corresponding absorption spectra were recorded and studied via decomposition and parameterization of dose-dependent spectral features. Lorentzian profiles were employed for spectral decomposition. Each film type displayed unique spectral signatures with distinct absorption peaks: nine for EBT3, eleven for EBT4, and twelve constituent profiles for EBT-XD. Notably, the EBT4 film demonstrated a slight difference in the blue part of the absorption spectrum and a change in the response, relative to its EBT3 predecessor. Orientation dependence of the film spectra was most pronounced for the EBT3 film type, followed by a declining trend across EBT4 and EBT-XD films. Absorption spectroscopy portrayed distinct spectral fingerprints of the studied film types, aiding the selection of the most suitable film for specific applications.

Objective.A major limitation in cone beam CT (CBCT) application is the presence of metal artifacts when scanning metal-embedded objects or high attenuation materials. This study aims to develop a dual-energy based method for effective metal artifact reduction.Approach.The proposed method comprised three steps. Initially, the virtual monoenergetic (VM) projections were generated by combining high- and low-energy projections to mitigate metal artifacts caused by the beam hardening effect. Subsequently, the normalized metal artifact reduction (NMAR) projections were created using the VM projections through the NMAR method. Then, the NMAR CBCT was produced by reintegrating metal into the CBCT reconstructed from NMAR projections. Finally, the iterative reconstruction was employed to obtain the final CBCT, utilizing VM projections and the NMAR CBCT as the initial input. Validation of the proposed method was achieved through Monte Carlo (MC) simulations on digital dental and abdominal phantoms, and CBCT scanning on CIRS Model 062M head and body phantoms. The structural similarity index measurement (SSIM) and the root mean square error (RMSE) calculated within a metal-containing ROI were employed for image quality evaluation.Main Results.Both the MC simulation and phantom scanning demonstrated that the proposed method was superior to the frequency split metal artifact reduction (FSMAR) method in mitigating artifacts and preserving anatomic details around metal. Averaged over four phantoms, the SSIM was enhanced from 99.48% with FSMAR to 99.86% with our proposed method, and the RMSE was reduced from 93.62 HU to 70.75 HU. Furthermore, the proposed method could be implemented with less than two minutes after GPU acceleration.Significance.The proposed dual-energy based metal artifact correction method effectively corrects metal artifacts and preserves tissue details surrounding the metal region by leveraging the strengths of VM, projection interpolation and iterative reconstruction techniques. It has strong potential of clinical implementation due to the superior performance in image quality and process efficiency.

Objective.To investigate the impact of the positioning of plane-parallel ionization chambers in proton beams on the calculation of the chamber-specific factorf_Qand, hence, the beam quality correction factorkQ,Q0.Approach.Monte Carlo simulations were performed to calculate the chamber-specific factorf_Qin monoenergetic proton beams for six different plane-parallel ionization chambers while positioning the chambers with a) their reference point and b) their effective point of measurement accounting for the water equivalent thickness of the entrance window.Main results.For all ionization chamber models investigated in this study, the difference inf_Qbetween both positioning approaches was larger for steeper dose gradients and bigger differences between the geometrical thickness and water-equivalent thickness of the entrance window. The largest effect was 1.2% for the IBA PPC-05 ionization chamber at an energy of 60 MeV.Significance.The positioning of plane-parallel ionization chambers in proton beams has a systematic impact on thef_Qfactor. This is especially of relevance for thekQ,Q0factors presented in the recently updated TRS-398 code of practice (CoP) from IAEA. The background is that a positioning with the effective point of measurement is prescribed in TRS-398 CoP, however, all Monte Carlo derived data that have been employed for the update are based on a positioning of the ionization chambers with their reference point. Hence, the updatedkQ,Q0factors for plane-parallel ionization chambers in proton beams are subject to systematic errors that can be as large as 0.5%.

Objective.Diffusion-weighted MRI (DW-MRI) is used to quantitatively characterize the microscopic structure of muscle through anisotropic water diffusion in soft tissue. Applications such as tumor propagation modeling require precise detection of muscle fiber orientation. That is, the direction along the fibers that coincides with the direction of the principal eigenvector of the diffusion tensor reconstructed from DW-MRI data. For clinical applications, the quality of image data is determined by the signal-to-noise ratio (SNR) that must be achieved within the appropriate scan time. The acquisition protocol must therefore be optimized. This implies the need for SNR criteria that match the data quality of the application.Approach.Muscles with known structural heterogeneity, e.g. bipennate muscles such as the rectus femoris in the thigh, provide a natural quality benchmark to determine accuracy of inferred fiber orientation at different scan parameters. In this study, we analyze DW-MR images of the thigh of a healthy volunteer at different SNRs and use PCA to identify subsets of voxels with different directions of diffusion tensor eigenvectors corresponding to different pennate angles. We propose to use the separation index of spatial co-localization of the clustered eigenvectors as a quality metric for fiber orientation detection.Main results.The clustering in the PCA component coordinates can be translated to the separation of the two compartments of the bipennate muscle on either side of the central tendon according to the pennate angle. The separation index reflects the degree of the separation and is a function of SNR.Significance.Because the separation index allows joint estimation of spatial and directional noise in DW-MRI as a single parameter, it will allow future quantitative optimization of DW-MRI soft tissue protocols.

Objective. This study aimed to evaluate two real-time target tracking approaches for magnetic resonance imaging (MRI) guided radiotherapy (MRgRT) based on foundation artificial intelligence models.Approach. The first approach used a point-tracking model that propagates points from a reference contour. The second approach used a video-object-segmentation model, based on segment anything model 2 (SAM2). Both approaches were evaluated and compared against each other, inter-observer variability, and a transformer-based image registration model, TransMorph, with and without patient-specific (PS) fine-tuning. The evaluation was carried out on 2D cine MRI datasets from two institutions, containing scans from 33 patients with 8060 labeled frames, with annotations from 2 to 5 observers per frame, totaling 29179 ground truth segmentations. The segmentations produced were assessed using the Dice similarity coefficient (DSC), 50% and 95% Hausdorff distances (HD50 / HD95), and the Euclidean center distance (ECD).Main results. The results showed that the contour tracking (median DSC0.92±0.04and ECD1.9±1.0 mm) and SAM2-based (median DSC0.93±0.03and ECD1.6±1.1 mm) approaches produced target segmentations comparable or superior to TransMorph w/o PS fine-tuning (median DSC0.91±0.07and ECD2.6±1.4 mm) and slightly inferior to TransMorph w/ PS fine-tuning (median DSC0.94±0.03and ECD1.4±0.8 mm). Between the two novel approaches, the one based on SAM2 performed marginally better at a higher computational cost (inference times 92 ms for contour tracking and 109 ms for SAM2). Both approaches and TransMorph w/ PS fine-tuning exceeded inter-observer variability (median DSC0.90±0.06and ECD1.7±0.7 mm).Significance. This study demonstrates the potential of foundation models to achieve high-quality real-time target tracking in MRgRT, offering performance that matches state-of-the-art methods without requiring PS fine-tuning.

Objective.Conventionally, if two metabolic processes are of interest for image analysis, two separate, sequential positron emission tomography (PET) scans are performed. However, sequential PET scans cannot simultaneously display the metabolic targets. The concurrent study of two simultaneous PET scans could provide new insights into the causes of diseases.Approach.In this work, we propose a reconstruction algorithm for the simultaneous injection of aβ+-emitter emitting only annihilation photons and aβ+-γ-emitter emitting annihilation photons and an additional promptγ-photon. As in previous works, theγ-photon is used to identify events originating from theβ+-γ-emitter. However, due to e.g. attenuation and down-scatter, theγ-photon is often not detected and not all events can correctly be associated with theβ+-γ-emitter as they are detected as double coincidences. In contrast to previous works, we estimate this number of double coincidences with origin in theβ+-γ, emitter including the attenuation of the promptγ, and incorporate this estimation in the forward-projection of the maximum likelihood expectation maximization algorithm. For evaluation, we simulate different scenarios with varying objects and attenuation maps. The nuclide¹⁸F serves asβ+-emitter, while⁴⁴Sc functions asβ+-γemitter. The performance of the algorithm is assessed by calculating the residual error of theβ+-γ-emitter in the reconstructedβ+-emitter image. Additionally, the intensity values in the simulated cylinders of the ground truth (GT) and the reconstructed images are compared.Main results.The remaining activity in theβ+-emitter image varied from 0.4% to 3.7%. The absolute percentage difference between GT and reconstructed intensity for the pureβ+emitter images was found to be between 3.0% and 7.4% for all cases. The absolute percentage difference between the GT and the reconstructed intensity for theβ+-γemitter images ranged from 8.7% to 10.4% for all simulated cases.Significance.These results demonstrate that our approach can reconstruct two separate images with a good quantitation accuracy.