pseudo-CT Images Research Articles

Generating pseudo-computerized tomography (P-CT) scan images from magnetic resonance imaging (MRI) images using machine learning algorithms based on fuzzy theory for radiotherapy treatment planning.

The substitution of computerized tomography (CT) with magnetic resonance imaging (MRI) has been investigated for external radiotherapy treatment planning. The present study aims to use pseudo-CT (P-CT) images created by MRI images to calculate the dose distribution for facilitating the treatment planning process. In this work, following image segmentation with a fuzzy clustering algorithm, an adaptive neuro-fuzzy algorithm was utilized to design the Hounsfield unit (HU) conversion model based on the features vector of MRI images. The model was generated on the set of extracted features from the gray-level co-occurrence matrices and the gray-level run-length matrices for 14 arbitrarily selected patients with brain malady. The performance of the algorithm was investigated on blind datasets through dose-volume histogram and isodose curve evaluations, using the RayPlan treatment planning system (TPS), along with the gamma analysis and statistical indices. In the proposed approach, the mean absolute error within the range of 45.4 HU was found among the test data. Also, the relative dose difference between the planning target volume region of the CT and the P-CT was 0.5%, and the best gamma pass rate for 3%/3mm was 97.2%. The proposed method provides a satisfactory average error rate for the generation of P-CT data in the different parts of the brain region from a collection of MRI series. Also, dosimetric parameters evaluation shows good agreement between reference CT and related P-CT images.

Evaluation of Deep Learning-Based Approaches to Segment Bowel Air Pockets and Generate Pelvic Attenuation Maps from CAIPIRINHA-Accelerated Dixon MR Images.

Attenuation correction remains a challenge in pelvic PET/MRI. In addition to the segmentation/model-based approaches, deep learning methods have shown promise in synthesizing accurate pelvic attenuation maps (μ-maps). However, these methods often misclassify air pockets in the digestive tract, potentially introducing bias in the reconstructed PET images. The aims of this work were to develop deep learning-based methods to automatically segment air pockets and generate pseudo-CT images from CAIPIRINHA-accelerated MR Dixon images. Methods: A convolutional neural network (CNN) was trained to segment air pockets using 3-dimensional CAIPIRINHA-accelerated MR Dixon datasets from 35 subjects and was evaluated against semiautomated segmentations. A separate CNN was trained to synthesize pseudo-CT μ-maps from the Dixon images. Its accuracy was evaluated by comparing the deep learning-, model-, and CT-based μ-maps using data from 30 of the subjects. Finally, the impact of different μ-maps and air pocket segmentation methods on the PET quantification was investigated. Results: Air pockets segmented using the CNN agreed well with semiautomated segmentations, with a mean Dice similarity coefficient of 0.75. The volumetric similarity score between 2 segmentations was 0.85 ± 0.14. The mean absolute relative changes with respect to the CT-based μ-maps were 2.6% and 5.1% in the whole pelvis for the deep learning-based and model-based μ-maps, respectively. The average relative change between PET images reconstructed with deep learning-based and CT-based μ-maps was 2.6%. Conclusion: We developed a deep learning-based method to automatically segment air pockets from CAIPIRINHA-accelerated Dixon images, with accuracy comparable to that of semiautomatic segmentations. The μ-maps synthesized using a deep learning-based method from CAIPIRINHA-accelerated Dixon images were more accurate than those generated with the model-based approach available on integrated PET/MRI scanners.

Development of attenuation correction methods using deep learning in brain-perfusion single-photon emission computed tomography.

Computed tomography (CT)-based attenuation correction (CTAC) in single-photon emission computed tomography (SPECT) is highly accurate, but it requires hybrid SPECT/CT instruments and additional radiation exposure. To obtain attenuation correction (AC) without the need for additional CT images, a deep learning method was used to generate pseudo-CT images has previously been reported, but it is limited because of cross-modality transformation, resulting in misalignment and modality-specific artifacts. This study aimed to develop a deep learning-based approach using non-attenuation-corrected (NAC) images and CTAC-based images for training to yield AC images in brain-perfusion SPECT. This study also investigated whether the proposed approach is superior to conventional Chang's AC (ChangAC). In total, 236 patients who underwent brain-perfusion SPECT were randomly divided into two groups: the training group (189 patients; 80%) and the test group (47 patients; 20%). Two models were constructed using Autoencoder (AutoencoderAC) and U-Net (U-NetAC), respectively. ChangAC, AutoencoderAC, and U-NetAC approaches were compared with CTAC using qualitative analysis (visual evaluation) and quantitative analysis (normalized mean squared error [NMSE] and the percentage error in each brain region). Statistical analyses were performed using the Wilcoxon signed-rank sum test and Bland-Altman analysis. U-NetAC had the highest visual evaluation score. The NMSE results for the U-NetAC were the lowest, followed by AutoencoderAC and ChangAC (P<0.001). Bland-Altman analysis showed a fixed bias for ChangAC and AutoencoderAC and a proportional bias for ChangAC. ChangAC underestimated counts by 30-40% in all brain regions. AutoencoderAC and U-NetAC produced mean errors of <1% and maximum errors of 3%, respectively. New deep learning-based AC methods for AutoencoderAC and U-NetAC were developed. Their accuracy was higher than that obtained by ChangAC. U-NetAC exhibited higher qualitative and quantitative accuracy than AutoencoderAC. We generated highly accurate AC images directly from NAC images without the need for intermediate pseudo-CT images. To verify our models' generalizability, external validation is required.

Research on obtaining pseudo CT images based on stacked generative adversarial network.

To investigate the feasibility of using a stacked generative adversarial network (sGAN) to synthesize pseudo computed tomography (CT) images based on ultrasound (US) images. The pre-radiotherapy US and CT images of 75 patients with cervical cancer were selected for the training set of pseudo-image synthesis. In the first stage, labeled US images were used as the first conditional GAN input to obtain low-resolution pseudo CT images, and in the second stage, a super-resolution reconstruction GAN was used. The pseudo CT image obtained in the first stage was used as an input, following which a high-resolution pseudo CT image with clear texture and accurate grayscale information was obtained. Five cross validation tests were performed to verify our model. The mean absolute error (MAE) was used to compare each pseudo CT with the same patient's real CT image. Also, another 10 cases of patients with cervical cancer, before radiotherapy, were selected for testing, and the pseudo CT image obtained using the neural style transfer (NSF) and CycleGAN methods were compared with that obtained using the sGAN method proposed in this study. Finally, the dosimetric accuracy of pseudo CT images was verified by phantom experiments. The MAE metric values between the pseudo CT obtained based on sGAN, and the real CT in five-fold cross validation are 66.82±1.59 HU, 66.36±1.85 HU, 67.26±2.37 HU, 66.34±1.75 HU, and 67.22±1.30 HU, respectively. The results of the metrics, namely, normalized mutual information (NMI), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR), between the pseudo CT images obtained using the sGAN method and the ground truth CT (CTgt) images were compared with those of the other two methods via the paired t-test, and the differences were statistically significant. The dice similarity coefficient (DSC) measurement results showed that the pseudo CT images obtained using the sGAN method were more similar to the CTgt images of organs at risk. The dosimetric phantom experiments also showed that the dose distribution between the pseudo CT images synthesized by the new method was similar to that of the CTgt images. Compared with NSF and CycleGAN methods, the sGAN method can obtain more accurate pseudo CT images, thereby providing a new method for image guidance in radiotherapy for cervical cancer.

Imaging Study of Pseudo-CT Synthesized From Cone-Beam CT Based on 3D CycleGAN in Radiotherapy

PurposeTo propose a synthesis method of pseudo-CT (CTCycleGAN) images based on an improved 3D cycle generative adversarial network (CycleGAN) to solve the limitations of cone-beam CT (CBCT), which cannot be directly applied to the correction of radiotherapy plans.MethodsThe improved U-Net with residual connection and attention gates was used as the generator, and the discriminator was a full convolutional neural network (FCN). The imaging quality of pseudo-CT images is improved by adding a 3D gradient loss function. Fivefold cross-validation was performed to validate our model. Each pseudo CT generated is compared against the real CT image (ground truth CT, CTgt) of the same patient based on mean absolute error (MAE) and structural similarity index (SSIM). The dice similarity coefficient (DSC) coefficient was used to evaluate the segmentation results of pseudo CT and real CT. 3D CycleGAN performance was compared to 2D CycleGAN based on normalized mutual information (NMI) and peak signal-to-noise ratio (PSNR) metrics between the pseudo-CT and CTgt images. The dosimetric accuracy of pseudo-CT images was evaluated by gamma analysis.ResultsThe MAE metric values between the CTCycleGAN and the real CT in fivefold cross-validation are 52.03 ± 4.26HU, 50.69 ± 5.25HU, 52.48 ± 4.42HU, 51.27 ± 4.56HU, and 51.65 ± 3.97HU, respectively, and the SSIM values are 0.87 ± 0.02, 0.86 ± 0.03, 0.85 ± 0.02, 0.85 ± 0.03, and 0.87 ± 0.03 respectively. The DSC values of the segmentation of bladder, cervix, rectum, and bone between CTCycleGAN and real CT images are 91.58 ± 0.45, 88.14 ± 1.26, 87.23 ± 2.01, and 92.59 ± 0.33, respectively. Compared with 2D CycleGAN, the 3D CycleGAN based pseudo-CT image is closer to the real image, with NMI values of 0.90 ± 0.01 and PSNR values of 30.70 ± 0.78. The gamma pass rate of the dose distribution between CTCycleGAN and CTgt is 97.0% (2%/2 mm).ConclusionThe pseudo-CT images obtained based on the improved 3D CycleGAN have more accurate electronic density and anatomical structure.

Conditional generative adversarial networks to generate pseudo low monoenergetic CT image from a single-tube voltage CT scanner.

To generate pseudo low monoenergetic CT images of the abdomen from 120-kVp CT images with cGAN. We retrospectively included 48 patients who underwent contrast-enhanced abdominal CT using dual-energy CT. We reconstructed paired data sets of 120 kVp CT images and virtual low monoenergetic (55-keV) CT images. cGAN was prepared to generate pseudo 55-keV CT images from 120-kVp CT images. The pseudo 55keV CT images in epoch 10, 50, 100, and 500 were compared to the 55keV images generated using peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM). The PSNRs were 28.0, 28.5, 28.6, and 28.8 at epochs 10, 50, 100, and 500, respectively. The SSIM was approximately constant from epochs 50 to 500. Pseudo low monoenergetic abdominal CT images were generated from 120-kVp CT images using cGAN, and the images had good quality similar to that of monochromatic images obtained with DECT software.

Attenuation correction using deep Learning and integrated UTE/multi-echo Dixon sequence: evaluation in amyloid and tau PET imaging.

Purpose:PET measures of amyloid and tau pathologies are powerful biomarkers for the diagnosis and monitoring of Alzheimer’s disease (AD). Because cortical regions are close to bone, quantitation accuracy of amyloid and tau PET imaging can be significantly influenced by errors of attenuation correction (AC). This work presents an MR-based AC method that combines deep learning with a novel ultrashort time-to-echo (UTE)/Multi-Echo Dixon (mUTE) sequence for amyloid and tau imaging.Methods:Thirty-five subjects that underwent both 11C-PiB and 18F-MK6240 scans were included in this study. The proposed method was compared with Dixon-based atlas method as well as other magnetization-prepared rapid acquisition with gradient echo (MPRAGE)- or Dixon- based deep learning methods. The Dice coefficient and validation loss of the generated pseudo-CT images were used for comparison. PET error images regarding standardized uptake value ratio (SUVR) were quantified through regional and surface analysis to evaluate the final AC accuracy.Results:The Dice coefficients of the deep learning methods based on MPRAGE, Dixon and mUTE images were 0.84 (0.91), 0.84 (0.92), and 0.87 (0.94) for the whole-brain (above-eye) bone regions, respectively, higher than the atlas method of 0.52 (0.64). The regional SUVR error for the atlas method was around 6%, higher than the regional SUV error. The regional SUV and SUVR errors for all deep learning methods were below 2%, with mUTE-based deep learning method performing the best. As for the surface analysis, the atlas method showed the largest error (> 10%) near vertices inside superior frontal, lateral occipital, superior parietal and inferior temporal cortices. The mUTE-based deep learning method resulted in the least number of regions with error higher than 1%, with the largest error (> 5%) showing up near the inferior temporal and medial orbitofrontal cortices.Conclusion:Deep learning with mUTE can generate accurate AC for amyloid and tau imaging in PET/MR.

Estimation and evaluation of pseudo-CT images using linear regression models and texture feature extraction from MRI images in the brain region to design external radiotherapy planning.

The aim of this study is to construct and evaluate Pseudo-CT images (P-CTs) for electron density calculation to facilitate external radiotherapy treatment planning. Despite numerous benefits, computed tomography (CT) scan does not provide accurate information on soft tissue contrast, which often makes it difficult to precisely differentiate target tissues from the organs at risk and determine the tumor volume. Therefore, MRI imaging can reduce the variability of results when registering with a CT scan. In this research, a fuzzy clustering algorithm was used to segment images into different tissues, also linear regression methods were used to design the regression model based on the feature extraction method and the brightness intensity values. The results of the proposed algorithm for dose-volume histogram (DVH), Isodose curves, and gamma analysis were investigated using the RayPlan treatment planning system, and VeriSoft software. Furthermore, various statistical indices such as Mean Absolute Error (MAE), Mean Error (ME), and Structural Similarity Index (SSIM) were calculated. The MAE of a range of 45-55 was found from the proposed methods. The relative difference error between the PTV region of the CT and the Pseudo-CT was 0.5, and the best gamma rate was 95.4% based on the polar coordinate feature and proposed polynomial regression model. The proposed method could support the generation of P-CT data for different parts of the brain region from a collection of MRI series with an acceptable average error rate by different evaluation criteria.

MR-based Attenuation Correction for Brain PET Using 3D Cycle-Consistent Adversarial Network.

Attenuation correction (AC) is important for the quantitative merits of positron emission tomography (PET). However, attenuation coefficients cannot be derived from magnetic resonance (MR) images directly for PET/MR systems. In this work, we aimed to derive continuous AC maps from Dixon MR images without the requirement of MR and computed tomography (CT) image registration. To achieve this, a 3D generative adversarial network with both discriminative and cycle-consistency loss (Cycle-GAN) was developed. The modified 3D U-net was employed as the structure of the generative networks to generate the pseudo CT/MR images. The 3D patch-based discriminative networks were used to distinguish the generated pseudo CT/MR images from the true CT/MR images. To evaluate its performance, datasets from 32 patients were used in the experiment. The Dixon segmentation and atlas methods provided by the vendor and the convolutional neural network (CNN) method which utilized registered MR and CT images were employed as the reference methods. Dice coefficients of the pseudo-CT image and the regional quantification in the reconstructed PET images were compared. Results show that the Cycle-GAN framework can generate better AC compared to the Dixon segmentation and atlas methods, and shows comparable performance compared to the CNN method.

Independent attenuation correction of whole body [18F]FDG-PET using a deep learning approach with Generative Adversarial Networks

BackgroundAttenuation correction (AC) of PET data is usually performed using a second imaging for the generation of attenuation maps. In certain situations however—when CT- or MR-derived attenuation maps are corrupted or CT acquisition solely for the purpose of AC shall be avoided—it would be of value to have the possibility of obtaining attenuation maps only based on PET information. The purpose of this study was to thus develop, implement, and evaluate a deep learning-based method for whole body [18F]FDG-PET AC which is independent of other imaging modalities for acquiring the attenuation map.MethodsThe proposed method is investigated on whole body [18F]FDG-PET data using a Generative Adversarial Networks (GAN) deep learning framework. It is trained to generate pseudo CT images (CTGAN) based on paired training data of non-attenuation corrected PET data (PETNAC) and corresponding CT data. Generated pseudo CTs are then used for subsequent PET AC. One hundred data sets of whole body PETNAC and corresponding CT were used for training. Twenty-five PET/CT examinations were used as test data sets (not included in training). On these test data sets, AC of PET was performed using the acquired CT as well as CTGAN resulting in the corresponding PET data sets PETAC and PETGAN. CTGAN and PETGAN were evaluated qualitatively by visual inspection and by visual analysis of color-coded difference maps. Quantitative analysis was performed by comparison of organ and lesion SUVs between PETAC and PETGAN.ResultsQualitative analysis revealed no major SUV deviations on PETGAN for most anatomic regions; visually detectable deviations were mainly observed along the diaphragm and the lung border. Quantitative analysis revealed mean percent deviations of SUVs on PETGAN of − 0.8 ± 8.6% over all organs (range [− 30.7%, + 27.1%]). Mean lesion SUVs showed a mean deviation of 0.9 ± 9.2% (range [− 19.6%, + 29.2%]).ConclusionIndependent AC of whole body [18F]FDG-PET is feasible using the proposed deep learning approach yielding satisfactory PET quantification accuracy. Further clinical validation is necessary prior to implementation in clinical routine applications.

Generation of Pseudo-CT using High-Degree Polynomial Regression on Dual-Contrast Pelvic MRI Data

Increasing interests in using magnetic resonance imaging only in radiation therapy require methods for predicting the computed tomography numbers from MRI data. Here we propose a simple voxel method to generate the pseudo-CT (pCT) image using dual-contrast pelvic MRI data. The method is first trained with the CT data and dual-contrast MRI data (two sets of MRI with different sequences) of multiple patients, where the anatomical structures in the images after deformable image registration are segmented into several regions, and after MRI intensity normalizations a regression analysis is used to determine a two-variable polynomial function for each region to relate a voxel’s two MRI intensity values to its CT number. We first evaluate the accuracy via the Hounsfield unit (HU) difference between the pseudo-CT and reference-CT (rCT) images and obtain the average mean absolute error as 40.3 ± 2.9 HU from leave-one-out-cross-validation (LOOCV) across all six patients, which is better than most previous results and comparable to another study using the more complicated atlas-based method. We also perform a dosimetric evaluation of the treatment plans based on pCT and rCT images and find the average passing rate within 2% dose difference to be 95.4% in point-to-point dose comparisons. Therefore, our method shows encouraging results in predicting the CT numbers. This polynomial method needs less computer storage than the interpolation method and can be readily extended to the case of more than two MRI sequences.

Pseudo CT Image Synthesis and Bone Segmentation From MR Images Using Adversarial Networks With Residual Blocks for MR-Based Attenuation Correction of Brain PET Data

For photon attenuation correction, current positron emission tomography systems combined with magnetic resonance imaging (PET/MR) imaging systems typically use methods based on MR image segmentation with subsequent assignment of empirical attenuation coefficients in PET image reconstruction. Delineation of bone in MR images has been challenging, especially in the head and neck areas, due to the difficulty of separating bone from air. In this article, we study deep learning techniques that assist the MR-based attenuation correction (MRAC) process for PET/MR systems, with focus on the brain region. We use a generative adversarial network (GAN) with residual blocks in a conditional setting for this task. We studied the performance of the designed network on image translation and segmentation tasks, which are essential for MRAC. For both tasks, the network generates pseudo CT images that resemble real CT images with normalized pixel value difference of around 5% and structural similarity (SSIM) index of around 0.8.

Pseudo-CT generation from multi-parametric MRI using a novel multi-channel multi-path conditional generative adversarial network for nasopharyngeal carcinoma patients.

To develop and evaluate a novel method for pseudo-CT generation from multi-parametric MR images using multi-channel multi-path generative adversarial network (MCMP-GAN). Pre- and post-contrast T1-weighted (T1-w), T2-weighted (T2-w) MRI, and treatment planning CT images of 32 nasopharyngeal carcinoma (NPC) patients were employed to train a pixel-to-pixel MCMP-GAN. The network was developed based on a 5-level Residual U-Net (ResU-Net) with the channel-based independent feature extraction network to generate pseudo-CT images from multi-parametric MR images. The discriminator with five convolutional layers was added to distinguish between the real CT and pseudo-CT images, improving the nonlinearity and prediction accuracy of the model. Eightfold cross validation was implemented to validate the proposed MCMP-GAN. The pseudo-CT images were evaluated against the corresponding planning CT images based on mean absolute error (MAE), peak signal-to-noise ratio (PSNR), Dice similarity coefficient (DSC), and Structural similarity index (SSIM). Similar comparisons were also performed against the multi-channel single-path GAN (MCSP-GAN), the single-channel single-path GAN (SCSP-GAN). It took approximately 20h to train the MCMP-GAN model on a Quadro P6000, and less than 10s to generate all pseudo-CT images for the subjects in the test set. The average head MAE between pseudo-CT and planning CT was 75.7±14.6Hounsfield Units (HU) for MCMP-GAN, significantly (P-values<0.05) lower than that for MCSP-GAN (79.2±13.0HU) and SCSP-GAN (85.8±14.3HU). For bone only, the MCMP-GAN yielded a smaller mean MAE (194.6±38.9HU) than MCSP-GAN (203.7±33.1HU), SCSP-GAN (227.0±36.7HU). The average PSNR of MCMP-GAN (29.1±1.6) was found to be higher than that of MCSP-GAN (28.8±1.2) and SCSP-GAN (28.2±1.3). In terms of metrics for image similarity, MCMP-GAN achieved the highest SSIM (0.92±0.02) but did not show significantly improved bone DSC results in comparison with MCSP-GAN. We developed a novel multi-channel GAN approach for generating pseudo-CT from multi-parametric MR images. Our preliminary results in NPC patients showed that the MCMP-GAN method performed apparently superior to the U-Net-GAN and SCSP-GAN, and slightly better than MCSP-GAN.

Generation of the Pseudo CT Image Based on the Deep Learning Technique Aimed for the Attenuation Correction of the PET Image

Computed tomography (CT) is used for the attenuation correction (AC) of [F-18] fluoro-deoxy-glucose positron emission tomography (PET) image. However, acquisition of a CT image for this purpose requires increasing the radiation dose of the patient. To generate a pseudo-image, a generative adversarial network (GAN) based on deep learning is adopted. The purpose of this study was to generate a pseudo-CT image, using a GAN, for the AC of the PET image, with the aim of reducing the dose of the patient. A set of approximately 15,000 no-AC PET and CT images was used as the training sample, and the CycleGAN was employed as the image generation model. The training samples were inputted in the CycleGAN, and the hyperparameters, i.e., the learning rate, batch size, and number of epochs were set to 0.0001, 1, and 300, respectively. A pseudo-PET image was obtained using a pseudo-CT image, which was used for the AC of the no-AC PET image. The coefficient of similarity between the real and generated pseudo-images was estimated using the peak signal-to-noise ratio (PSNR) , the structural similarity (SSIM), and the dice similarity coefficient (DSC). The average values of PSNR, SSIM, and DSC of the pseudo-CT were 31.0 dB, 0.87, and 0.89, and those of the pseudo-PET were 35.9 dB, 0.90, and 0.95, respectively. The AC for the whole-body PET image could be accomplished using the pseudo-CT image generated via the GAN. The proposed method would be established as the CT-less PET/CT examination.

Method for a motion model based automated 4D dose calculation

The Vero system can treat intra-fractionally moving tumors with gimbaled dynamic tumor tracking (DTT) by rotating the treatment beam so that it follows the motion of the tumor. However, the changes in the beam geometry and the constant breathing motion of the patient influence the dose applied to the patient. This study aims to perform a full 4D dose reconstruction for thirteen patients treated with DTT at the Vero system at the Universitätsklinikum Erlangen and investigates the temporal resolution required to perform an accurate 4D dose reconstruction. For all patients, a 4DCT was used to train a 4D motion model, which is able to calculate pseudo-CT images for arbitrary breathing phases. A new CT image was calculated for every 100 ms of treatment and a dose calculation was performed according to the current beam geometry (i.e. the rotation of the treatment beam at this moment in time) by rotating according to the momentary beam rotation, which is extracted from log-files. The resulting dose distributions were accumulated on the planning CT and characteristic parameters were extracted and compared. -evaluations of dose accumulations with different spatial-temporal resolutions were performed to determine the minimal required resolution. In total 173 700 dose calculations were performed. The accumulated 4D dose distributions show a reduced mean GTV dose of 0.77% compared to the static treatment plan. For some patients larger deviations were observed, especially in the presence of a poor 4DCT quality. The -evaluation showed that a temporal resolution of 500 ms is sufficient for an accurate dose reconstruction. If the tumor motion is regarded as well, a spatial-temporal sampling of 1400 ms and 2 mm yields accurate results, which reduces the workload by 84%.

Independent brain 18F-FDG PET attenuation correction using a deep learning approach with Generative Adversarial Networks.

Attenuation correction (AC) of positron emission tomography (PET) data poses a challenge when no transmission data or computed tomography (CT) data are available, e.g. in stand alone PET scanners or PET/magnetic resonance imaging (MRI). In these cases, external imaging data or morphological imaging data are normally used for the generation of attenuation maps. Newly introduced machine learning methods however may allow for direct estimation of attenuation maps from non attenuation-corrected PET data (PETNAC). Our purpose was thus to establish and evaluate a method for independent AC of brain fluorine-18-fluorodeoxyglucose (18F-FDG) PET images only based on PETNAC using Generative Adversarial Networks (GAN). After training of the deep learning GAN framework on a paired training dataset of PETNAC and the corresponding CT images of the head from 50 patients, pseudo-CT images were generated from PETNAC of 40 validation patients, of which 20 were used for technical validation and 20 stemming from patients with CNS disorders were used for clinical validation. Pseudo-CT was used for subsequent AC of these validation data sets resulting in independently attenuation-corrected PET data. Visual inspection revealed a high degree of resemblance of generated pseudo-CT images compared to the acquired CT images in all validation data sets, with minor differences in individual anatomical details. Quantitative analyses revealed minimal underestimation below 5% of standardized uptake value (SUV) in all brain regions in independently attenuation-corrected PET data compared to the reference PET images. Color-coded error maps showed no regional bias and only minimal average errors around ±0%. Using independently attenuation-corrected PET data, no differences in image-based diagnoses were observed in 20 patients with neurological disorders compared to the reference PET images. Independent AC of brain 18F-FDG PET is feasible with high accuracy using the proposed, easy to implement deep learning framework. Further evaluation in clinical cohorts will be necessary to assess the clinical performance of this method.

Pseudo-CT Generation for External Radiotherapy Planning from Magnetic Resonance Imaging Data on Brain Regions

Background: Nowadays, medical imaging has an important role in radiotherapy and treatment planning process. Despite the increasing usage of the magnetic resonance imaging (MRI) in the external radiotherapy (RT) design process, computed tomography (CT) remains a basic imaging modality in radiotherapy because of its relation with electron density value. In conventional external radiotherapy, MRI is used in functional tissue structures with registration on the CT image, which causes systematic errors during the registration of MRI and CT images. Objectives: The main purpose of this paper was to investigate the possibility of removing the CT simulator and replacing it with pseudo-CT for the electron density calculation in radiotherapy treatment planning. Materials and Methods: The pseudo-CT images were generated for 10 randomly chosen patients with brain disease. The data consisted of image voxels chosen within the segmented area of the brain in both MRI and CT images. The relationship between electron density information in CT images and MRI intensity value was derived from simple polynomial models. Results: It was found that the MRI intensity value is related to the Hounsfield unit (HU) value within different parts including the skull bone, sinus, and brain in CT images. The mean errors of the conversion model were 0.7479, 0.334 and -0.043 HU in the brain, skull, and air regions respectively, with the simple polynomial model in the upper slice. Conclusion: The proposed method generated pseudo-CT images from different segmented parts of the brain of MRI series. The application requires segmentation of the cortical bones in MR images. The result indicated that finding a complex model does not necessarily lead to achieving a better result.

Flexible Prediction of CT Images From MRI Data Through Improved Neighborhood Anchored Regression for PET Attenuation Correction.

Given the complicated relationship between the magnetic resonance imaging (MRI) signals and the attenuation values, the attenuation correction in hybrid positron emission tomography (PET)/MRI systems remains a challenging task. Currently, existing methods are either time-consuming or require sufficient samples to train the models. In this paper, an efficient approach for predicting pseudo computed tomography (CT) images from T1- and T2-weighted MRI data with limited data is proposed. The proposed approach uses improved neighborhood anchored regression (INAR) as a baseline method to pre-calculate projected matrices to flexibly predict the pseudo CT patches. Techniques, including the augmentation of the MR/CT dataset, learning of the nonlinear descriptors of MR images, hierarchical search for nearest neighbors, data-driven optimization, and multi-regressor ensemble, are adopted to improve the effectiveness of the proposed approach. In total, 22 healthy subjects were enrolled in the study. The pseudo CT images obtained using INAR with multi-regressor ensemble yielded mean absolute error (MAE) of 92.73 ± 14.86 HU, peak signal-to-noise ratio of 29.77 ± 1.63 dB, Pearson linear correlation coefficient of 0.82 ± 0.05, dice similarity coefficient of 0.81 ± 0.03, and the relative mean absolute error (rMAE) in PET attenuation correction of 1.30 ± 0.20% compared with true CT images. Moreover, our proposed INAR method, without any refinement strategies, can achieve considerable results with only seven subjects (MAE 106.89 ± 14.43 HU, rMAE 1.51 ± 0.21%). The experiments prove the superior performance of the proposed method over the six innovative methods. Moreover, the proposed method can rapidly generate the pseudo CT images that are suitable for PET attenuation correction.

Planification de la radiothérapie du cancer de la prostate par l’imagerie par résonance magnétique

PurposeMagnetic resonance imaging (MRI) plays an increasing role in radiotherapy dose planning. Indeed, MRI offers superior soft tissue contrast compared to computerized tomography (CT) and therefore could provide a better delineation of target volumes and organs at risk than CT for radiotherapy. Furthermore, an MRI-only radiotherapy workflow would suppress registration errors inherent to the registration of MRI with CT. However, the estimation of the electronic density of tissues using MRI images is still a challenging issue. The purpose of this work was to design and evaluate a pseudo-CT generation method for prostate cancer treatments. Materials and methodsA pseudo-CT was generated for ten prostate cancer patients using an elastic deformation based method. For each patient, dose delivered to the patient was calculated using both the planning CT and the pseudo-CT. Dose differences between CT and pseudo-CT were investigated. ResultsMean dose relative difference in the planning target volume is 0.9% on average and ranges from 0.1% to 1.7%. In organs at risks, this value is 1.8%, 0.8%, 0.8% and 1% on average in the rectum, the right and left femoral heads, and the bladder respectively. ConclusionThe dose calculated using the pseudo-CT is very close to the dose calculated using the CT for both organs at risk and PTV. These results confirm that pseudo-CT images generated using the proposed method could be used to calculate radiotherapy treatment doses on MRI images.

Magnetic resonance-driven pseudo CT image using patch-based multi-modal feature extraction and ensemble learning with stacked generalisation

In recent years, research has been showing a great interest in using magnetic resonance imaging (MRI) as the only modality for radiation therapy (RT) for its superior soft-tissue visualisation and non-ionizing proprieties. Furthermore, MRI-only RT would be of great benefit to eliminate image registration errors, reduce cost and workload. In addition, machine-learning algorithms have been taking the lead in many fields. For instance, in MRI-only RT, machine learning is showing a notable performance compared to other methods owed to the flexibility of these methods towards data regardless of model complexity. In this paper, we present an ensemble learning approach with stacked generalisation to simulate a CT scan from multi-modal MR images from which patch-based shape, texture and spatial features were considered. Feature extraction, fusion and reduction were performed to get the most descriptive and informative features. The ensemble learning model was constructed with two levels of learning were the basic level consisted of three base learners namely: artificial neural networks (ANN), random forests (RF) and k-nearest neighbours (kNN) and the second level representing the stacking learner that takes predictions from the base learner and generates the final predictions. Multiple linear regression (MLR) was used for the stacked generalisation. The proposed ensemble learning with stacked generalisation (ES) approach produced an average mean absolute error (MAE) of 87.60 ± 19.70 and an average mean error (ME) of −4.68 ± 16.43 outperforming the RF method, which produced an average MAE of 106.88 ± 33.20 and an average ME of −5.38 ± 20.77. In addition, average Pearson correlation was 0.92 for the proposed approach compared to 0.89 for RF. Evaluation of the proposed approach shows that stacked generalisation can greatly improve prediction accuracy and reduce bias in electron density estimation.