Search

Multi-modality imaging constitutes a foundation of precision medicine, especially in oncology where reliable and rapid imaging techniques are needed in order to insure adequate diagnosis and treatment. In cervical cancer, precision oncology requires the acquisition of 18F-labelled 2-fluoro-2-deoxy-D-glucose (FDG) positron emission tomography (PET), magnetic resonance (MR), and computed tomography (CT) images. Thereafter, images are co-registered to derive electron density attributes required for FDG-PET attenuation correction and radiation therapy planning. Nevertheless, this traditional approach is subject to MR-CT registration defects, expands treatment expenses, and increases the patient's radiation exposure. To overcome these disadvantages, we propose a new framework for cross-modality image synthesis which we apply on MR-CT image translation for cervical cancer diagnosis and treatment. The framework is based on a conditional generative adversarial network (cGAN) and illustrates a novel tactic that addresses, simplistically but efficiently, the paradigm of vanishing gradient vs. feature extraction in deep learning. Its contributions are summarized as follows: 1) The approach-termed sU-cGAN- uses, for the first time, a shallow U-Net (sU-Net) with an encoder/decoder depth of 2 as generator; 2) sU-cGAN's input is the same MR sequence that is used for radiological diagnosis, i.e. T2-weighted, Turbo Spin Echo Single Shot (TSE-SSH) MR images; 3) Despite limited training data and a single input channel approach, sU-cGAN outperforms other state of the art deep learning methods and enables accurate synthetic CT (sCT) generation. In conclusion, the suggested framework should be studied further in the clinical settings. Moreover, the sU-Net model is worth exploring in other computer vision tasks.

Purpose Generation of synthetic Computed Tomography (sCT) images from Magnetic Resonance (MR) is an imperative, yet not fully resolved task for attenuation correction in Positron Emission Tomography (PET)/MR and treatment planning in MR-only radiation therapy. Herein, we propose a Generative Adversarial Networks (GAN) model to generate quantitatively accurate sCT from Dixon thorax MR data. Materials/methods Paired image sets of Dixon MR and corresponding CT volumes from fourteen subjects were used. In-phase (IP) MR images were registered to the CT images using REGGUI. The resulting deformation field was applied to the three other [opposed-phase (OP), fat, water] MR images. Image processing was implemented in MATLAB 2016b (MathWorks) using COMKAT Image Tool. Manual contouring was performed on the CT images using MIM (version 6.6.10, MIM Software Inc., Cleveland, OH) software, to delineate the left lung, right lung, vertebral body, and spinal cord. The proposed RU-cGAN model incorporates R esNet and U-Net in the generator of a c onditional G AN. Results from Vgg16, Vgg19, and ResNet were used as references for comparison. The four networks were trained using two strategies. In the first, the four Dixon MR images were used as input features. In the second strategy, only fat and water images were used. Results Regardless of whether the input consists of two or four images types, sCT generated using RU- cGAN had the best agreement with the measured CT for all the metrics, having Mean Absolute Prediction Error (MAPE) ​ Conclusion RU-cGAN provides a rapid and accurate method for thorax sCT generation while obviating the bone-specific UTE pulse sequence and only requiring limited training data.

Purpose: MR-based pseudo-CT has an important role in MR-based radiation therapy planning and PET attenuation correction. The purpose of this study is to establish a clinically feasible approach, including image acquisition, correction, and CT formation, for pseudo-CT generation of the brain using a single-acquisition, undersampled ultrashort echo time (UTE)-mDixon pulse sequence. Methods: Nine patients were recruited for this study. For each patient, a 190-s, undersampled, single acquisition UTE-mDixon sequence of the brain was acquired (TE = 0.1, 1.5, and 2.8 ms). A novel method of retrospective trajectory correction of the free induction decay (FID) signal was performed based on point-spread functions of three external MR markers. Two-point Dixon images were reconstructed using the first and second echo data (TE = 1.5 and 2.8 ms). R2∗ images (1/T2∗) were then estimated and were used to provide bone information. Three image features, i.e., Dixon-fat, Dixon-water, and R2∗, were used for unsupervised clustering. Five tissue clusters, i.e., air, brain, fat, fluid, and bone, were estimated using the fuzzy c-means (FCM) algorithm. A two-step, automatic tissue-assignment approach was proposed and designed according to the prior information of the given feature space. Pseudo-CTs were generated by a voxelwise linear combination of the membership functions of the FCM. A low-dose CT was acquired for each patient and was used as the gold standard for comparison. Results: The contrast and sharpness of the FID images were improved after trajectory correction was applied. The mean of the estimated trajectory delay was 0.774 μs (max: 1.350 μs; min: 0.180 μs). The FCM-estimated centroids of different tissue types showed a distinguishable pattern for different tissues, and significant differences were found between the centroid locations of different tissue types. Pseudo-CT can provide additional skull detail and has low bias and absolute error of estimated CT numbers of voxels (−22 ± 29 HU and 130 ± 16 HU) when compared to low-dose CT. Conclusions: The MR features generated by the proposed acquisition, correction, and processing methods may provide representative clustering information and could thus be used for clinical pseudo-CT generation.

Purpose: To evaluate the information available from a single MRI acquisition for a complete MRI‐based simulation in brain.Method and Materials: A 3.0T MRI scanner (Achieva TX, Philips Healthcare) was used to acquire a complete brain simulation data set in a single acquisition on four consenting volunteers. The acquisition consisted of collecting the FID signal (TE1 = 100μs) followed by two additional gradient echoes (TE2/TE3= 1.4ms/2.5ms) using a 3D volumetric excitation and radial read‐out toachieve 1.3mm isotropic voxels. The data from this single acquisition was used to reconstruct five volumetric data sets: Bone‐enhanced, fat‐only, water‐only, in‐phase and out‐of‐phase. The resulting image sets were assessed for image quality sufficient for organ delineation and used to generate digitally‐reconstructed radiographs (DRRs) (Pinnacle Workstation, Philips Healthcare). Results: Each volunteer study took less than 10 minutes to complete, and the single acquisition required less than 3 minutes. Images from all four volunteers had excellent image quality sufficient for organ delineation and complete cortical bone segmentation. In addition, the DDRs for all four volunteers were sufficient for 2D patient matching. Conclusion: This study confirms the feasibility of using a single acquisition MRI as a sole imaging modality for treatment planning simulation in the brain. To validate this method, we plan to use this imaging protocol in a group of patients and compare the DRRs and dose plans with those acquired during CT simulation. Both authors are employees of Philips Healthcare.

The use of ionizing radiation for cancer treatment has undergone extraordinary development during the past hundred years. The advancement of medical imaging has been critical in helping to achieve this change. The invention of computed tomography (CT) was pivotal in the development of treatment planning. Despite some disadvantages, CT remains the only three-dimensional imaging modality used for dose calculation. Newer image modalities, such as magnetic resonance (MR) imaging and positron emission tomography (PET), are also used secondarily in the treatment-planning process. MR, with its better tissue contrast and resolution than those of CT, improves tumor definition compared with CT planning alone. PET also provides metabolic information to supplement the CT and MR anatomical information. With emerging molecular imaging techniques, the ability to visualize and characterize tumors with regard to their metabolic profile, active pathways, and genetic markers, both across different tumors and within individual, heterogeneous tumors, will inform clinicians regarding the treatment options most likely to benefit a patient and to detect at the earliest time possible if and where a chosen therapy is working. In the post-human-genome era, multimodality scanners such as PET/CT and PET/MR will provide optimal tumor targeting information.

4D flow MRI is a powerful technique for quantitative flow assessment and visualization of complex flow patterns and hemodynamics of cardiovascular flows. This technique results in more anatomical information and comprehensive assessment of blood flow. However, conventional 4D PC MRI suffers from a few obstacles for clinical applications. The total scan time is long, especially in large volumes with high spatial resolutions. Inaccuracy of conventional Cartesian PC MRI in the setting of atherosclerosis and in general, disturbed and turbulent blood flow is another important challenge. This inaccuracy is the consequence of signal loss, intravoxel dephasing and flow-related artifact in the presence of disturbed and turbulent flow. Spiral k-space trajectory has valuable attributes which can help overcome some of the problems with 4D flow Cartesian acquisitions. Spiral trajectory benefits from shorter TE and reduces the flow-related artifacts. In addition, short spiral readouts with spiral interleaves can significantly reduce the total scan time, reducing the chances of patient motion which may also corrupt the data in the form of motion artifacts. In this paper, the accuracy of flow assessment and flow visualization with reduced TE 4D Spiral PC was investigated and good agreement was observed between the spiral and conventional technique. The systolic mean velocity, peak flow and the average flow in CCA and ICA of normal volunteers using 4D spiral PC MRI showed errors less than 10% compared to conventional 4D PC MRI. In addition, the scan time using spiral sequence was 3∶31 min which is half of the time using conventional sequence.

Purpose: Accurate and robust photon attenuation derived from MR is essential for PET/MR and MR-based radiation treatment planning applications. Although the fuzzy C-means (FCM) algorithm has been applied for pseudo-CT generation, the input feature combination and the number of clusters have not been optimized. This study aims to optimize both for clinically practical pseudo-CT generation. Methods: Nine volunteers were recruited. A 190-second, single-acquisition UTE-mDixon with 25% (angular) sampling and 3D radial readout was performed to acquire three primitive MR features at TEs of 0.1, 1.5, and 2.8 ms: the free-induction-decay (FID), the first and the second echo images. Three derived images, Dixon-fat and Dixon-water generated by two-point Dixon water/fat separation, and R2* (1/T2*) map, were also created. To identify informative inputs for generating a pseudo-CT image volume, all 63 combinations, choosing one to six of the feature images, were used as inputs to FCM for pseudo-CT generation. Further, the number of clusters was varied from four to seven to find the optimal approach. Mean prediction deviation (MPD), mean absolute prediction deviation (MAPD), and correlation coefficient (R) of different combinations were compared for feature selection. Results: Among the 63 feature combinations, the four that resulted in the best MAPD and R were further compared along with the set containing all six features. The results suggested that R2* and Dixon-water are the most informative features. Further, including FID also improved the performance of pseudo-CT generation. Consequently, the set containing FID, Dixon-water, and R2* resulted in the most accurate, robust pseudo-CT when the number of cluster equals to five (5C). The clusters were interpreted as air, fat, bone, brain, and fluid. The six-cluster Result additionally included bone marrow. Conclusion: The results suggested that FID, Dixon-water, R2* are the most important features. The findings can be used to facilitate pseudo-CT generation for unsupervised clustering. Please note that the project was completed with partial funding from the Ohio Department of Development grant TECH 11-063 and a sponsored research agreement with Philips Healthcare that is managed by Case Western Reserve University. As noted in the affiliations, some of the authors are Philips employees

Purpose: Emerging technologies such as dedicated PET/MRI and MR-therapy systems require robust and clinically practical methods for determining photon attenuation. Herein, we propose using novel MR acquisition methods and processing for the generation of pseudo-CTs. Methods: A single acquisition, 190-second UTE-mDixon sequence with 25% (angular) sampling density and 3D radial readout was performed on nine volunteers. Three water-filled tubes were placed in the FOV for trajectory-delay correction. The MR data were reconstructed to generate three primitive images acquired at TEs of 0.1, 1.5 and 2.8 ms. In addition, three derived MR images were generated, i.e. two-point Dixon water/fat separation and R2* (1/T2*) map. Furthermore, two spatial features, i.e. local binary pattern (S-1) and relative spatial coordinates (S-2), were incorporated. A direct-mapping operator was generated using Artificial Neural Networks (ANNs) for transforming the MR features to a pseudo-CT. CT images served as the training data and, using a leave-one-out method, for performance evaluation using mean prediction deviation (MPD), mean absolute prediction deviation (MAPD), and correlation coefficient (R). Results: The errors between measured CT and pseudo-CT declined dramatically when the spatial features, i.e. S-1 and S-2, were included. The MPD, MAPD, and R were, respectively, 5±57 HU, 141±41 HU, and 0.815±0.066 for results generated by the ANN trained without the spatial features and were 32±26 HU, 115±18 HU, and 0.869±0.035 with the spatial features. The estimation errors of the pseudo-CT were smaller when both the S-1 and S-2 were used together than when either the S-1 or the S-2 was used. Pseudo-CT generation (256×256×256 voxels) by ANN took < 0.5 s using a computer having an Intel i7 3.4GHz CPU and 16 GB RAM. Conclusion: The proposed direct-mapping ANN approach is a technically accurate, clinically practical method for pseudo-CT generation and can potentially help improve the accuracy of MR-AC and MR-RTP applications. Please note that the project was completed with partial funding from the Ohio Department of Development grant TECH 11-063 and a sponsored research agreement with Philips Healthcare that is managed by Case Western Reserve University. As noted in the affiliations, some of the authors are Philips employees

Purpose: The ultrashort echo-time (UTE) sequence is a promising MR pulse sequence for imaging cortical bone which is otherwise difficult to image using conventional MR sequences and also poses strong attenuation for photons in radiation therapy and PET imaging. The authors report here a systematic characterization of cortical bone signal decay and a scanning time optimization strategy for the UTE sequence through k-space undersampling, which can result in up to a 75% reduction in acquisition time. Using the undersampled UTE imaging sequence, the authors also attempted to quantitatively investigate the MR properties of cortical bone in healthy volunteers, thus demonstrating the feasibility of using such a technique for generating bone-enhanced images which can be used for radiation therapy planning and attenuation correction with PET/MR. Methods: An angularly undersampled, radially encoded UTE sequence was used for scanning the brains of healthy volunteers. Quantitative MR characterization of tissue properties, including water fraction and R2∗ = 1/T2∗, was performed by analyzing the UTE images acquired at multiple echo times. The impact of different sampling rates was evaluated through systematic comparison of the MR image quality, bone-enhanced image quality, image noise, water fraction, and R2∗ of cortical bone. Results: A reduced angular sampling rate of the UTE trajectory achieves acquisition durations in proportion to the sampling rate and in as short as 25% of the time required for full sampling using a standard Cartesian acquisition, while preserving unique MR contrast within the skull at the cost of a minimal increase in noise level. The R2∗ of human skull was measured as 0.2–0.3 ms−1 depending on the specific region, which is more than ten times greater than the R2∗ of soft tissue. The water fraction in human skull was measured to be 60%–80%, which is significantly less than the >90% water fraction in brain. High-quality, bone-enhanced images can be generated using a reduced sampled UTE sequence with no visible compromise in image quality and they preserved bone-to-air contrast with as low as a 25% sampling rate. Conclusions: This UTE strategy with angular undersampling preserves the image quality and contrast of cortical bone, while reducing the total scanning time by as much as 75%. The quantitative results of R2∗ and the water fraction of skull based on Dixon analysis of UTE images acquired at multiple echo times provide guidance for the clinical adoption and further parameter optimization of the UTE sequence when used for radiation therapy and MR-based PET attenuation correction.

Purpose: To generate electron density (ED) maps and digitally reconstructed radiographs (DRRs) of the pelvis based only on magnetic resonance imaging (MRI). Methods: A conventional 3D fast‐field dual‐echo sequence was used to acquire image data of 8 healthy subjects with a Philips 3.0T Ingenia TX system in approximately 1:50 min scan time per volunteer. Dixon reconstruction of the nearly out‐of‐phase (echo 1) and in‐phase images (echo 2) allowed for water and fat classification. A bone‐enhanced image was generated by automatically thresholding the noise level of the in‐phase image with subsequent background removal. ED maps were then produced by assigning known bulk electron densities to the classified bone and tissue fractions. A bone probability atlas derived from CT data was registered to the ED map in order to filter out misclassified voxels. Finally, DRRs were reconstructed from bone‐enhanced images as well as from ED maps. Results: The proposed MRI sequence with subsequent Dixon reconstruction and probabilistic filtering makes it possible to classify cortical bone, soft tissue and adipose tissue in the pelvis and yields ED maps and corresponding DRRs. Bowel content was misclassified as cortical bone or air and compromised the segmentation in some slices as well as in the DRRs. Automatic probabilistic atlas filtering can significantly reduce artifacts induced by bowel content without affecting pelvic bone structures markedly. In total, the artifact/bone fraction dropped from 1.7 before filtering to 0.2 after filtering. The average reduction of artifact volume is 87%, and the average bone preservation is 99%. Remaining artifacts are spatially close to the true bone in areas of positive bone probability. Conclusion: This study demonstrated the feasibility of generating realistic ED maps of the pelvis by using MRI only. The method has the potential to become an essential component of emerging applications such as MR‐only‐based radiation therapy planning. All authors have the following relevant financial interest or relationship to disclose with regard to the subject matter of this presentation: Company name: Philips Research; Type of relationship: Employee

Methods 5 control patients (4M/1F, age 44±21 y) with normal left ventricular (LV) function and no history of heart disease, 5 patients with ICM (3M/2F, 63±2 y), and 16 patients with HCM (10M/6F, age 51±15y) were recruited into an IRB approved protocol. All subjects were imaged at 1.5T (Achieva XR, Philips). T1 mapping was performed using MOLLI and LL at 2 LV short axis levels (basaland midcavity) before and after contrast agent injection (Magnevist; 0.2 mmol/kg). Post gadopentetate dimeglumine (Magnevist; 0.2 mmol/kg) injection time was about 15 min for HCM and ICM patients and was shorter (about 5 min) for the controls due to organizational reasons. For each subject, TI measurements were made in 6 ROIs per level.

Purpose: The construction of a digitally reconstructed radiograph (DRR) from a magnetic resonance image (MRI) is possible if the cortical bone signal can be acquired and separated from air and soft tissue. This may be accomplished by subtracting a long echo-time, in-phase, gradient echo (GRE) image volume from an ultra-short echo time free induction decay (FID) image to produce a bone-enhanced (BE) image that reveals cortical bone. One limitation of this approach is the length of time required for data acquisition, which can limit the quality of the DRRs due to patient and organ motion. This study aimed to significantly reduce the acquisition time without compromising DRR quality. Methods: Brain data were acquired from two volunteers using a 3T MR scanner (Ingenia, Philips Healthcare). The FID and GRE images were acquired in a single acquisition using a 3D radial readout sequence with the following parameters: TE1=0.142ms (ultra-short), TE2=2.197ms (nearly in-phase), 2*2*2mm3 isotropic voxels, 250*250*250mm3 FOV. To reduce the acquisition time, k-space was sampled at 75, 50 and 25% of a full 3D sphere . The TE2 image was subtracted from the TE1 image to generate the BE images. The BE images were used to generate DRRs using the Pinnacle treatment planning system (Philips-version 9.2). The quality of the DRRs was evaluated qualitatively by 5 board certified medical physicists for clinical usefulness. Results: The acquisition time for 75, 50 and 25% sampling schemes were 219s, 146s, and 73s, respectively, the latter of which was a four-fold reduction in scan time compared to a 300s fully-sampled acquisition. All DRRs obtained were of acceptable quality and were shown to have sufficient information for clinical 2D image matching. Conclusion: Undersampling k-space while maintaining the same range of frequency information results in significantly reduced scan time and clinically acceptable DRR image quality. Drs. B Traughber and R Muzic have research support from Philips Healthcare. Drs. M Traughber and L Hu are employees of Philips Healthcare.

Background This study describes initial testing and evaluation of a vertical-field open Magnetic Resonance Imaging (MRI) scanner for the purpose of simulation in radiation therapy for prostate cancer. We have evaluated the clinical workflow of using open MRI as a sole modality for simulation and planning. Relevant results related to MRI alignment (vs. CT) reference dataset with Cone-Beam CT (CBCT) for daily localization are presented. Methods Ten patients participated in an IRB approved study utilizing MRI along with CT simulation with the intent of evaluating the MRI-simulation process. Differences in prostate gland volume, seminal vesicles, and penile bulb were assessed with MRI and compared to CT. To evaluate dose calculation accuracy, bulk-density-assignments were mapped onto respective MRI datasets and treated IMRT plans were re-calculated. For image localization purposes, 400 CBCTs were re-evaluated with MRI as the reference dataset and daily shifts compared against CBCT-to-CT registration. Planning margins based on MRI/CBCT shifts were computed using the van Herk formalism. Results Significant organ contour differences were noted between MRI and CT. Prostate volumes were on average 39.7% (p = 0.002) larger on CT than MRI. No significant difference was found in seminal vesicle volumes (p = 0.454). Penile bulb volumes were 61.1% higher on CT, without statistical significance (p = 0.074). MRI-based dose calculations with assigned bulk densities produced agreement within 1% with heterogeneity corrected CT calculations. The differences in shift positions for the cohort between CBCT-to-CT registration and CBCT-to-MRI registration are −0.15 ± 0.25 cm (anterior-posterior), 0.05 ± 0.19 cm (superior-inferior), and −0.01 ± 0.14 cm (left-right). Conclusions This study confirms the potential of using an open-field MRI scanner as primary imaging modality for prostate cancer treatment planning simulation, dose calculations and daily image localization.