Pseudo CT Research Articles

CT respiratory motion synthesis using joint supervised and adversarial learning

Objective. Four-dimensional computed tomography (4DCT) imaging consists in reconstructing a CT acquisition into multiple phases to track internal organ and tumor motion. It is commonly used in radiotherapy treatment planning to establish planning target volumes. However, 4DCT increases protocol complexity, may not align with patient breathing during treatment, and lead to higher radiation delivery. Approach. In this study, we propose a deep synthesis method to generate pseudo respiratory CT phases from static images for motion-aware treatment planning. The model produces patient-specific deformation vector fields (DVFs) by conditioning synthesis on external patient surface-based estimation, mimicking respiratory monitoring devices. A key methodological contribution is to encourage DVF realism through supervised DVF training while using an adversarial term jointly not only on the warped image but also on the magnitude of the DVF itself. This way, we avoid excessive smoothness typically obtained through deep unsupervised learning, and encourage correlations with the respiratory amplitude. Main results. Performance is evaluated using real 4DCT acquisitions with smaller tumor volumes than previously reported. Results demonstrate for the first time that the generated pseudo-respiratory CT phases can capture organ and tumor motion with similar accuracy to repeated 4DCT scans of the same patient. Mean inter-scans tumor center-of-mass distances and Dice similarity coefficients were 1.97 mm and 0.63, respectively, for real 4DCT phases and 2.35 mm and 0.71 for synthetic phases, and compares favorably to a state-of-the-art technique (RMSim). Significance. This study presents a deep image synthesis method that addresses the limitations of conventional 4DCT by generating pseudo-respiratory CT phases from static images. Although further studies are needed to assess the dosimetric impact of the proposed method, this approach has the potential to reduce radiation exposure in radiotherapy treatment planning while maintaining accurate motion representation. Our training and testing code can be found at https://github.com/cyiheng/Dynagan.

Personalized depth-specific neuromodulation of the human primary motor cortex via ultrasound.

Non-invasive brain stimulation has the potential to boost neuronal plasticity in the primary motor cortex (M1), but it remains unclear whether the stimulation of both superficial and deep layers of the human motor cortex can effectively promote M1 plasticity. Here, we leveraged transcranial ultrasound stimulation (TUS) to precisely target M1 circuits at depths of approximately 5mm and 16mm from the cortical surface. Initially, we generated computed tomography images from each participant's individual anatomical magnetic resonance images (MRI), which allowed for the generation of accurate acoustic simulations. This process ensured that personalized TUS was administered exactly to the targeted depths within M1 for each participant. Using long-term depression and long-term potentiation (LTD/LTP) theta-burst stimulation paradigms, we examined whether TUS over distinct depths of M1 could induce LTD/LTP plasticity. Our findings indicated that continuous theta-burst TUS-induced LTD-like plasticity with both superficial and deep M1 stimulation, persisting for at least 30min. In comparison, sham TUS did not significantly alter M1 excitability. Moreover, intermittent theta-burst TUS did not result in the induction of LTP- or LTD-like plasticity with either superficial or deep M1 stimulation. These findings suggest that the induction of M1 plasticity can be achieved with ultrasound stimulation targeting distinct depths of M1, which is contingent on the characteristics of TUS. KEY POINTS: The study integrated personalized transcranial ultrasound stimulation (TUS) with electrophysiology to determine whether TUS targeting superficial and deep layers of the human motor cortex (M1) could elicit long-term depression (LTD) or long-term potentiation (LTP) plastic changes. Utilizing acoustic simulations derived from individualized pseudo-computed tomography scans, we ensured the precision of TUS delivery to the intended M1 depths for each participant. Continuous theta-burst TUS targeting both the superficial and deep layers of M1 resulted in the emergence of LTD-like plasticity, lasting for at least 30min. Administering intermittent theta-burst TUS to both the superficial and deep layers of M1 did not lead to the induction of LTP- or LTD-like plastic changes. We suggest that theta-burst TUS targeting distinct depths of M1 can induce plasticity, but this effect is dependent on specific TUS parameters.

Correcting CT misregistration in data-driven gated (DDG) PET with PET self-gating and deformable image registration.

Misregistration between CT and PET data can result in mis-localization and inaccurate quantification of functional uptake in whole body PET/CT imaging. This problem is exacerbated when an abnormal inspiration occurs during the free-breathing helical CT (FB CT) used for attenuation correction of PET data. In data-driven gated (DDG) PET, the data selected for reconstruction is typically derived from the end-expiration (EE) phase of the breathing cycle, making this potential issue worse. The objective of this study is to develop a deformable image registration (DIR)-based respiratory motion model to improve the registration and quantification between misregistered FB CT and PET. Twenty-two whole-body 18 F-FDG PET/CT scans encompassing 48 lesions in misregistered regions were analyzed in this study. End-inspiration (EI) and EE PET data were derived from -10% to 15% and 30% to 80% of the breathing cycle, respectively. DIR was used to estimate a motion model from the EE to EI phase of the PET data. The model was then used to generate PET images at any phase of up to four times the amplitude of motion between EE and EI for correlation with the misregistered FB CT. Once a matched phase of the FB CT was determined, FB CT was deformed to a pseudo CT at the EE phase (DIR CT). DIR CT was compared with the ground truth DDG CT for AC and localization of the DDG PET. Between DDG PET/FB CT and DDG PET/DIR CT, a significant increase in ∆%SUV was observed (p<0.01), with median values elevating from 26.7% to 42.4%. This new method was most effective for lesions ≤3cm proximal to the diaphragm (p<0.001) but showed decreasing efficacy as the distance increased. When FB CT was severely misregistered with DDG PET (>3cm), DDG PET/DIR CT outperformed DDG PET/FB CT alone (p<0.05). Even when patients showed varied breathing patterns during the PET/CT scan, DDG PET/DIR CT still surpassed the efficiency of DDG PET/FB CT (p<0.01). Though DDG PET/DIR CT couldn't match the performance of the DDG PET/CT ground truth (42.4%vs. 53.6%, p<0.01), it reached 84% of its quantification, demonstrating good agreement and a strong overall correlation (regression coefficient of 0.94, p<0.0001). In some cases, anatomical distortion and blurring, and misregistration error were observed in DIR CT, rendering it still unable to correct inaccurate localization near the boundaries of two organs. Based on the motion model derived from gated PET data, DIR CT can significantly improve the quantification and localization of DDG PET. This approach can achieve a performance level of about 84% of the ground truth established by DDG PET/CT. These results show that self-gated PET and DIR CT may offer an alternative clinical solution to DDG PET and FB CT for quantification without the need for additional cine-CT imaging. DIR CT was at times inferior to DDG CT due to some distortion and blurring of anatomy and misregistration error.

Pseudo-Computed Tomography Generation from Noisy Magnetic Resonance Imaging with Deep Learning Algorithm

Purpose: Magnetic Resonance Imaging (MRI) applications offer superior soft tissue contrast compared with Computed Tomography (CT) for accurate radiotherapy planning although MRI images suffer from poor image quality and lack electron density for radiation dose calculation. The present study aims to use the Deep Learning (DL) approach to 1) enhance the quality of MRI images and 2) generate synthetic CT images using MRI images for more accurate radiotherapy planning.&#x0D; Materials and Methods: In this paper, the pix2pix Generative Adversarial Network was utilized to synthesize CT images from noisy MRI images of 20 arbitrarily patients with brain disease. The standard statistical measurements investigated the accuracy comparison of the modeled Hounsfield Unit (HU) value from MRI images and referenced CT of each patient. The famous quality metrics that were used to compare synthetic CTs and referenced CTs were the Mean Absolute eError (MAE), the structural similarity index (SSIM), and the Peak Signal-to-Noise Ratio (PSNR).&#x0D; Results: The higher quality measurements between the synthetic pseudo-CT and the referenced CT images as PSNR and SSIM should correlate with the lower MAE value. For the overall brain among blind test data, the measured peak signal-to-noise ratio, mean absolute error, and structural similarity index values were about 16.5, 28.13, and 93.46, respectively.&#x0D; Conclusion: The proposed method provides an acceptable level of statistical measurements computed on the Pseudo-CT and referenced CT, and it could be concluded that the p-CT can be implemented in radiotherapy treatment planning with acceptable accuracy.

Deep learning-based attenuation correction method in 99mTc-GSA SPECT/CT hepatic imaging: a phantom study.

This study aimed to evaluate a deep learning-based attenuation correction (AC) method to generate pseudo-computed tomography (CT) images from non-AC single-photon emission computed tomography images (SPECTNC) for AC in 99mTc-galactosyl human albumin diethylenetriamine pentaacetic acid (GSA) scintigraphy and to reduce patient dosage. A cycle-consistent generative network (CycleGAN) model was used to generate pseudo-CT images. The training datasets comprised approximately 850 liver phantom images obtained from SPECTNC and real CT images. The training datasets were then input to CycleGAN, and pseudo-CT images were output. SPECT images with real-time CT attenuation correction (SPECTCTAC) and pseudo-CT attenuation correction (SPECTGAN) were acquired. The difference in liver volume between real CT and pseudo-CT images was evaluated. Total counts and uniformity were then used to evaluate the effects of AC. Additionally, the similarity coefficients of SPECTCTAC and SPECTGAN were assessed using a structural similarity (SSIM) index. The pseudo-CT images produced a lower liver volume than the real CT images. SPECTCTAC exhibited a higher total count than SPECTNC and SPECTGAN, which were approximately 60% and 7% lower, respectively. The uniformities of SPECTCTAC and SPECTGAN were better than those of SPECTNC. The mean SSIM value for SPECTCTAC and SPECTGAN was 0.97. We proposed a deep learning-based AC approach to generate pseudo-CT images from SPECTNC images in 99mTc-GSA scintigraphy. SPECTGAN with AC using pseudo-CT images was similar to SPECTCTAC, demonstrating the possibility of SPECT/CT examination with reduced exposure to radiation.

A multi-centric evaluation of self-learning GAN based pseudo-CT generation software for low field pelvic magnetic resonance imaging.

An artificial intelligence-based pseudo-CT from low-field MR images is proposed and clinically evaluated to unlock the full potential of MRI-guided adaptive radiotherapy for pelvic cancer care. In collaboration with TheraPanacea (TheraPanacea, Paris, France) a pseudo-CT AI-model was generated using end-to-end ensembled self-supervised GANs endowed with cycle consistency using data from 350 pairs of weakly aligned data of pelvis planning CTs and TrueFisp-(0.35T)MRIs. The image accuracy of the generated pCT were evaluated using a retrospective cohort involving 20 test cases coming from eight different institutions (US: 2, EU: 5, AS: 1) and different CT vendors. Reconstruction performance was assessed using the organs at risk used for treatment. Concerning the dosimetric evaluation, twenty-nine prostate cancer patients treated on the low field MR-Linac (ViewRay) at Montpellier Cancer Institute were selected. Planning CTs were non-rigidly registered to the MRIs for each patient. Treatment plans were optimized on the planning CT with a clinical TPS fulfilling all clinical criteria and recalculated on the warped CT (wCT) and the pCT. Three different algorithms were used: AAA, AcurosXB and MonteCarlo. Dose distributions were compared using the global gamma passing rates and dose metrics. The observed average scaled (between maximum and minimum HU values of the CT) difference between the pCT and the planning CT was 33.20 with significant discrepancies across organs. Femoral heads were the most reliably reconstructed (4.51 and 4.77) while anal canal and rectum were the less precise ones (63.08 and 53.13). Mean gamma passing rates for 1%1mm, 2%/2mm, and 3%/3mm tolerance criteria and 10% threshold were greater than 96%, 99% and 99%, respectively, regardless the algorithm used. Dose metrics analysis showed a good agreement between the pCT and the wCT. The mean relative difference were within 1% for the target volumes (CTV and PTV) and 2% for the OARs. This study demonstrated the feasibility of generating clinically acceptable an artificial intelligence-based pseudo CT for low field MR in pelvis with consistent image accuracy and dosimetric results.

Pseudo CT Synthesis Using Cone-Beam CT of Cervical Cancer with GAN-Based Neural Network Model

Cervical cancer (CC) is a tumor disease that threatens the health of women. As an important treatment of CC, radiotherapy has been widely used in clinic. With the rapid development of radiotherapy technology, adaptive radiotherapy has received much attention. Adaptive radiotherapy means more accurate radiation dose and more accurate radiation area, which can effectively protect normal tissue. It is significant to improve the local control rate of tumor and the quality of life of patients. However, the Cone-Beam CT (CBCT) images collected during radiotherapy are of poor quality and cannot provide real-time radiation effect information, resulting in timely and effective adjustment of radiation dose and radiation area in the process of radiotherapy for cervical cancer. To alleviate this issue, this study will establish a model to leverage CC CBCT images to synthetize pseudo computed tomography (CT) images with high quality, so as to achieve the purpose of quality improvement. This study included the data of 20 patients with CC in ** hospital. The planning CT and CBCT scan data of each patient before radiotherapy were collected, and the interval between the two kinds of image data was required to be less than one week. After data preprocessing, a total of 1206 pairs of images were trained and tested. The generative adversarial network (GAN) is constructed. In order to ensure the similarity between the input image and the output image, the L1 loss function is leveraged. And the full supervision method is used to train the model to achieve a better effect of image synthesis and improve the quality of CBCT image. Peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) were used as evaluation indexes. Using five-fold cross-validation, the values of PSNR between the pseudo-CT (sCT) and the planning CT (pCT) image and between the CBCT and the pCT image are calculated. The results are 26.9 and 22.6, respectively. The sCT obtained from the GAN model increases the peak signal-to-noise ratio by 19% compared with the original CBCT, which means that the proposed model built in this study can improve the useful information of the CBCT image. The SSIM values between sCT and pCT and between CBCT and pCT are also calculated, and the average values of them are 0.89 and 0.63, respectively. Therefore, in this experiment, the structure of the sCT obtained by the proposed model is closer to pCT. And the SSIM increases by 41.2% compared with the original CBCT, which means that the sCT by the proposed model is more similar to the pCT in structure. These results could make a more accurate judgment on the effect of radiotherapy. In this study, the pseudo-CT synthesis method based on GAN can improve the quality of CC CBCT image. The results show this method makes the structure clearer and could assist doctors to adjust the radiation dose and radiation area in time. This study is able to facilitate the development of adaptive radiotherapy for CC.

Pseudo CT Images Synthesized from CBCT Based on Unsupervised DDPM in Cervical Cancer Radiotherapy

The low image quality and inaccurate HU value of cone beam computed tomography (CBCT) restrict its further application in cervical cancer radiotherapy. A new unsupervised based denoising diffusion probabilistic model (DDPM) was proposed to synthesize pseudo-CT images from CBCT. CBCT and CT images of 120 patients with cervical cancer were selected. The proposed DDPM network with condition and iterative mechanism was used for data training and testing between two image domains. In the training process, the model first obtained coarse pseudo-CT images with Gaussian noise through a diffusion process. Then, with the real CT images as the training target, the noise images were nonlinearly mapped to the domain of the CT images through the inverse diffusion process, and the fine pseudo-CT images were obtained. We repeated the above steps and gradually increased the number of noise diffusion. When the image difference value is less than the threshold, the training of the model is terminated and the pseudo-CT images are output. In the testing stage, each pseudo-CT generated was compared against the real CT image of the same patient based on the metrics of peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) and normalized mean absolute error (NMAE). In terms of anatomical verification, the PSNR (dB), SSIM (%) and NCC values between the pseudo-CT images obtained based on DDPM and the real CT images were presented as mean (standard deviation), which were 31.92(0.46), 86.69(4.55) and 0.0106(0.002) respectively. Compared with CBCT, the accuracy of the three metric values has been improved 15.6%, 14.2% and 23.3% respectively. For the metric values of the pseudo-CT images obtained based on the U-Net and CycleGAN models, the results synthesized based on the proposed model were paired with T-tests, the p values were all less than 0.05, and the differences were statistically significant. The pseudo-CT images obtained based on the DDPM were close to the real CT images in anatomy. The pseudo-CT images synthesized by the proposed DDPM network have good application prospects in cervical cancer radiotherapy.

Australian perspectives on artificial intelligence in veterinary practice

While artificial intelligence (AI) and recent developments in deep learning (DL) have sparked interest in medical imaging, there has been little commentary on the impact of AI on the veterinarian and veterinary imaging technologists. This survey study aimed to understand the attitudes, applications, and concerns among veterinarians and radiography professionals in Australia regarding the rapidly emerging applications of AI. An anonymous online survey was circulated to the members of three Australian veterinary professional organizations. The survey invitations were shared via email and social media with the survey open for 5 months. Among the 84 respondents, there was a high level of acceptance of lower order tasks (e.g., patient registration, triage, and dispensing) and less acceptance of high order task automation (e.g., surgery and interpretation). There was a low priority perception for the role of AI in higher order tasks (e.g., diagnosis, interpretation, and decision making) and high priority for those applications that automate complex tasks (e.g., quantitation, segmentation, reconstruction) or improve image quality (e.g., dose/noise reduction and pseudo CT for attenuation correction). Medico-legal, ethical, diversity, and privacy issues posed moderate or high concern while there appeared to be no concern regarding AI being clinically useful and improving efficiency. Mild concerns included redundancy, training bias, transparency, and validity. Australian veterinarians and veterinary professionals recognize important applications of AI for assisting with repetitive tasks, performing less complex tasks, and enhancing the quality of outputs in medical imaging. There are concerns relating to ethical aspects of algorithm development and implementation.

Pseudo Computed Tomography estimation from brain MRI using anatomic signature and joint dictionary learning

Most research on using pseudo-computed tomography (pCT) on brain-imaging techniques relies on in-house methods. As performance as a whole increase, they pay particular attention when using MRI imaging. Methodologies for predicting CT values from MRI data are needed in radiation treatment (RT). Although the employment of dictionary-learning-based approaches for defining picture patches has not been considered, it has been found that Deep Learning (DL) offers increased opportunities in the medical domains. The stages of this paper CT estimation from MRI using Anatomic Signature and Joint Dictionary Learning (ASJDL) are as follows: a) data gathering from the RIRE image data and b) image pre-processing to remove anomalies. c) Using Gabor Filters for feature extraction to extract significant features d) The choice of anatomic signature traits is used to identify strong, illuminating characteristics that classify and identify objects. e) Development of the pCT through shared dictionary learning. Using cutting-edge techniques like Accelerated Simplified Swarm Optimization (ASSO), Particle Swarm Optimization (PSO), Simplified Swarm Optimization (SSO), SLA12, Intensity-based, Fast-patch based, and Coupled dictionary over measures like peak S/N ratio, MAE, normalized cross-correlation, SSIM, Accuracy, and Computation time, the evaluation of both feature selection and classification methods is compared. With values of 23.825.08 (PSNR), 83.327.05 (MAE(HU)), 0.920.03 (NCC), and 0.860.03 (SSIM), the suggested framework (ASJDL) performs better than all other methods.

Deep learning with multiresolution handcrafted features for brain MRI segmentation

The segmentation of magnetic resonance (MR) images is a crucial task for creating pseudo computed tomography (CT) images which are used to achieve positron emission tomography (PET) attenuation correction. One of the main challenges of creating pseudo CT images is the difficulty to obtain an accurate segmentation of the bone tissue in brain MR images. Deep convolutional neural networks (CNNs) have been widely and efficiently applied to perform MR image segmentation. The aim of this work is to propose a segmentation approach that combines multiresolution handcrafted features with CNN-based features to add directional properties and enrich the set of features to perform segmentation. The main objective is to efficiently segment the brain into three tissue classes: bone, soft tissue, and air. The proposed method combines non subsampled Contourlet (NSCT) and non subsampled Shearlet (NSST) coefficients with CNN’s features using different mechanisms. The entropy value is calculated to select the most useful coefficients and reduce the input’s dimensionality. The segmentation results are evaluated using fifty clinical brain MR and CT images by calculating the precision, recall, dice similarity coefficient (DSC), and Jaccard similarity coefficient (JSC). The results are also compared to other methods reported in the literature. The DSC of the bone class is improved from 0.6179 ± 0.0006 to 0.6416 ± 0.0006. The addition of multiresolution features of NSCT and NSST with CNN’s features demonstrates promising results. Moreover, NSST coefficients provide more useful information than NSCT coefficients.

Personalized Modeling to Improve Pseudo–Computed Tomography Images for Magnetic Resonance Imaging–Guided Adaptive Radiation Therapy

Magnetic resonance imaging-guided adaptive radiation therapy (MRIgART) greatly improves daily tumor localization and enables online replanning to obtain maximum dosimetric benefits. However, accurately predicting patient-specific electron density maps for adaptive radiation therapy planning remains a challenge. Therefore, this study proposes a personalized modeling framework for generating pseudo-computed tomography (pCT) in MRIgART. Eighty-three patients who received MRIgART were included and computed tomography (CT) simulations were performed on all the patients. Daily T2-weighted 1.5 T magnetic resonance imaging (MRI) was acquired using the Unity MR-linac for adaptive planning. Pairs of coregistered CT and daily MRI images of the randomly selected training set (68 patients) were inputted into a generative adversarial network to establish a population model. The personalized model for each patient in the test set (15 patients) was acquired using model fine-tuning, which adopted the pair of the deformable-registered CT and the first daily MRI to fine-tune the population model. The pCT quality was quantitatively evaluated in the second and the last fractions with 3 metrics: intensity accuracy using mean absolute error; anatomic structure similarity using dice similarity coefficient; and dosimetric consistency using gamma-passing rate. The image generation speed was 65 slices/s. For the last fractions, and for head-neck, thoracoabdominal, and pelvic cases, the average mean absolute errors were 76.8 HU versus 123.6 HU, 38.1 HU versus 52.0 HU, and 29.5 HU versus 39.7 HU, respectively. Furthermore, the average dice similarity coefficients of bone were 0.92 versus 0.80, 0.85 versus 0.73, and 0.94 versus 0.88; and the average gamma-passing rates (1%/1 mm) were 95.5% versus 84.7%, 97.7% versus 92.8%, and 95.5% versus 88.7%, for personalized versus population models, respectively. The results of the second fractions were similar. The proposed personalized modeling framework remarkably improved pCT quality for multiple treatment sites and was well suited for the MRIgART clinical setting.

Australian perspectives on artificial intelligence in medical imaging.

IntroductionWhile artificial intelligence (AI) and recent developments in deep learning (DL) have sparked interest in medical imaging, there has been little commentary on the impact of AI on imaging technologists. The aim of this survey was to understand the attitudes, applications and concerns among nuclear medicine and radiography professionals in Australia with regard to the rapidly emerging applications of AI.MethodsAn anonymous online survey with invitation to participate was circulated to nuclear medicine and radiography members of the Rural Alliance in Nuclear Scintigraphy and the Australian Society of Medical Imaging and Radiation Therapy. The survey invitations were sent to members via email and as a push via social media with the survey open for 10 weeks. All information collected was anonymised and there is no disclosure of personal information as it was de‐identified from commencement.ResultsAmong the 102 respondents, there was a high level of acceptance of lower order tasks (e.g. patient registration, triaging and dispensing) and less acceptance of high order task automation (e.g. surgery and interpretation). There was a low priority perception for the role of AI in higher order tasks (e.g. diagnosis, interpretation and decision making) and high priority for those applications that automate complex tasks (e.g. quantitation, segmentation, reconstruction) or improve image quality (e.g. dose / noise reduction and pseudo CT for attenuation correction). Medico‐legal, ethical, diversity and privacy issues posed moderate or high concern while there appeared to be no concern regarding AI being clinically useful and improving efficiency. Mild concerns included redundancy, training bias, transparency and validity.ConclusionAustralian nuclear medicine technologists and radiographers recognise important applications of AI for assisting with repetitive tasks, performing less complex tasks and enhancing the quality of outputs in medical imaging. There are concerns relating to ethical aspects of algorithm development and implementation.

Brain MR images segmentation using 3D CNN with features recalibration mechanism for segmented CT generation

The segmentation of MR (magnetic resonance) images is a simple approach to create Pseudo CT images which are useful for many medical imaging analysis applications. One of the main challenges of this process is the bone segmentation of brain MR images. Deep convolutional neural networks (CNNs) have been widely and efficiently applied to perform MR images segmentation. The aim of this work is to propose a novel excitation-based CNN by recalibrating the network features adaptively to enhance the bone segmentation by segmenting the brain MR images into three tissue classes: bone, soft tissue, and air. The proposed method combines two types of features excitation mechanisms namely: (1) spatial squeeze and channel excitation block (cSE) and (2) channel squeeze and spatial excitation block (sSE). The two blocks are combined sequentially and integrated seamlessly into a 3D convolutional encoder decoder network. The novelty of this work emerges in the combination of the two excitation blocks sequentially to improve the segmentation performance and reduce the model complexity. The proposed approach is evaluated through a comparison with computed tomography (CT) images as ground truth and validated with other methods in the literature that applied deep CNN approaches to perform MR image segmentation for PET attenuation correction. Brain MR and CT datasets which consist of 50 patients are used to evaluate the proposed method. The segmentation performance of the three brain classes is evaluated using precision, recall, dice similarity coefficient (DSC), and Jaccard index. The presented method improves the bone tissue segmentation compared to the baseline model and other methods in the literature where the DSC is improved from 0.6278 ± 0.0006 to 0.6437 ± 0.0006 with an improvement percentage of 2.53% for bone class. The proposed excitation-based segmentation network architecture demonstrates promising and competitive results compared with other methods in the literature and reduces the model complexity thanks to the sequential combination of the two excitation blocks.

Compensating Positron Range Effects of Ga-68 in Preclinical PET Imaging by Using Convolutional Neural Network: A Monte Carlo Simulation Study.

This study aimed to investigate the feasibility of positron range correction based on three different convolutional neural network (CNN) models in preclinical PET imaging of Ga-68. The first model (CNN1) was originally designed for super-resolution recovery, while the second model (CNN2) and the third model (CNN3) were originally designed for pseudo CT synthesis from MRI. A preclinical PET scanner and 30 phantom configurations were modeled in Monte Carlo simulations, where each phantom configuration was simulated twice, once for Ga-68 (CNN input images) and once for back-to-back 511-keV gamma rays (CNN output images) with a 20 min emission scan duration. The Euclidean distance was used as the loss function to minimize the difference between CNN input and output images. According to our results, CNN3 outperformed CNN1 and CNN2 qualitatively and quantitatively. With regard to qualitative observation, it was found that boundaries in Ga-68 images became sharper after correction. As for quantitative analysis, the recovery coefficient (RC) and spill-over ratio (SOR) were increased after correction, while no substantial increase in coefficient of variation of RC (CVRC) or coefficient of variation of SOR (CVSOR) was observed. Overall, CNN3 should be a good candidate architecture for positron range correction in Ga-68 preclinical PET imaging.

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.

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.

A Preliminary Simulation Study of Dose-Guided Adaptive Radiotherapy Based on Halcyon MV Cone-Beam CT Images With Retrospective Data From a Phase II Clinical Trial.

Background and purpose: To evaluate the feasibility of dose-guided adaptive radiotherapy (ART) based on deformable image registration (DIR) using fractional megavoltage cone-beam CT (MVCBCT) images from Halcyon system that uses identical beams for treatment and imaging and to retrospectively investigate the influence of anatomic changes on target coverage and organ-at-risk (OAR) sparing across various tumor sites.Materials and Methods: Four hundred twenty-two MVCBCT images from 16 patients (three head and neck, seven thoracic, three abdominal, and three pelvic cases) treated in a phase II clinical trial for Halcyon were selected. DIR between the planning CT and daily MVCBCT image was implemented by Velocity software to create pseudo CT. To investigate the accuracy of dose calculation on pseudo CT, three evaluation patients with rescanned CT and adaptive plans were selected. Dose distribution of adaptive plans calculated on pseudo CT was compared with that calculated on the rescanned planning CT on the three evaluation patients. To investigate the impact of inter-fractional anatomic changes on target dose coverage and dose to OARs of the 16 patients, fractional dose was calculated and accumulated incrementally based on deformable registration between planning CT and daily MVCBCT images.Results: Passing rates using 3 mm/3%/10% threshold local gamma analysis were 93.04, 96.00, and 91.68%, respectively, for the three evaluation patients between the reconstructed dose on pseudo CT (MVCBCT) and rescanned CT, where accumulated dose deviations of over 97% voxels were smaller than 0.5 Gy. Planning target volume (PTV) D95% and D90% (the minimum dose received by at least 95/90% of the volume) of the accumulated dose could be as low as 93.8 and 94.5% of the planned dose, respectively. OAR overdose of various degrees were observed in the 16 patients relative to the planned dose. In most cases, OARs' dose volume histogram (DVH) lines of accumulated and planned dose were very close to each other if not overlapping. Among cases with visible deviations, the differences were bilateral without apparent patterns specific to tumor sites or organs.Conclusion: As a confidence building measure, this simulation study suggested the possibility of ART for Halcyon based on DIR between planning CT and MVCBCT. Preliminary clinical data suggested the benefit of patient-specific dose reconstruction and ART to avoid unacceptable target underdosage and OAR overdosage.