To maximize acquisition bandwidth in zero echo time (ZTE) sequences, readout gradients are already switched on during the RF pulse, creating unwanted slice selectivity. The resulting image distortions are amplified especially when the anatomy of interest is not located at the isocenter. We aim to characterize off-center ZTE MRI of extremities such as the shoulder, knee, and hip, adjusting the carrier frequency of the RF pulse excitation for each TR.
In ZTE MRI, radial encoding schemes are used, where the distorted slice profile due to the finite RF pulse length rotates with the k-space trajectory. To overcome these modulations for objects far away from the magnet isocenter, the frequency of the RF pulse is shifted for each gradient setting so that artifacts do not occur at a given off-center target position. The sharpness of the edges in the images were calculated and the ZTE acquisition with off-center excitation was compared to an acquisition with isocenter excitation both in phantom and in vivo off-center MRI of the shoulder, knee, and hip at 1.5 and 3T MRI systems.
Distortion and blurriness artifacts on the off-center MRI images of the phantom, in vivo shoulder, knee, and hip images were mitigated with off-center excitation without time or noise penalty, at no additional computational cost.
The off-center excitation allows ZTE MRI of the shoulder, knee, and hip for high-bandwidth image acquisitions for clinical settings, where positioning at the isocenter is not possible.
Dose calculation for MR-guided radiotherapy (MRgRT) at the 0.35 T MR-Linac is currently based on deformation of planning CTs (defCT) acquired for each patient. We present a simple and robust bulk density overwrite synthetic CT (sCT) method for abdominal treatments in order to streamline clinical workflows.
Fifty-six abdominal patient treatment plans were retrospectively evaluated. All patients had been treated at the MR-Linac using MR datasets for treatment planning and plan adaption and defCT for dose calculation. Bulk density CTs (4M-sCT) were generated from MR images with four material compartments (bone, lung, air, soft tissue). The relative electron densities (RED) for bone and lung were extracted from contoured CT structure average REDs. For soft tissue, a correlation between BMI and RED was evaluated. Dose was recalculated on 4M-sCT and compared to dose distributions on defCTs assessing dose differences in the PTV and organs at risk (OAR).
Mean RED of bone was 1.17 ± 0.02, mean RED of lung 0.17 ± 0.05. The correlation between BMI and RED for soft tissue was statistically significant (p < 0.01). PTV dose differences between 4M-sCT and defCT were Dmean: −0.4 ± 1.0%, D1%: −0.3 ± 1.1% and D95%: −0.5 ± 1.0%. OARs showed D2%: −0.3 ± 1.9% and Dmean: −0.1 ± 1.4% differences. Local 3D gamma index pass rates (2%/2mm) between dose calculated using 4M-sCT and defCT were 96.8 ± 2.6% (range 89.9–99.6%).
The presented method for sCT generation enables precise dose calculation for MR-only abdominal MRgRT.
Cone-beam computed tomography (CBCT)-based online adaptation is increasingly being introduced into many clinics. Upon implementation of a new treatment technique, a prospective risk analysis is required and enhances workflow safety. We conducted a risk analysis using Failure Mode and Effects Analysis (FMEA) upon the introduction of an online adaptive treatment programme (Wegener et al., Z Med Phys. 2022).
A prospective risk analysis, lacking in-depth clinical experience with a treatment modality or treatment machine, relies on imagination and estimates of the occurrence of different failure modes. Therefore, we systematically documented all irregularities during the first year of online adaptation, namely all cases in which quality assurance detected undesired states potentially leading to negative consequences. Additionally, the quality of automatic contouring was evaluated. Based on those quantitative data, the risk analysis was updated by an interprofessional team. Furthermore, a hypothetical radiation therapist-only workflow during adaptive sessions was included in the prospective analysis, as opposed to the involvement of an interprofessional team performing each adaptive treatment.
A total of 126 irregularities were recorded during the first year. During that time period, many of the previously anticipated failure modes (almost) occurred, indicating that the initial prospective risk analysis captured relevant failure modes. However, some scenarios were not anticipated, emphasizing the limits of a prospective risk analysis. This underscores the need for regular updates to the risk analysis. The most critical failure modes are presented together with possible mitigation strategies. It was further noted that almost half of the reported irregularities applied to the non-adaptive treatments on this treatment machine, primarily due to a manual plan import step implemented in the institution’s workflow.
The recently introduced Varian Ethos system allows adjusting radiotherapy treatment plans to anatomical changes on a daily basis. The system uses artificial intelligence to speed up the process of creating adapted plans, comes with its own software solutions and requires a substantially different workflow. A detailed analysis of possible risks of the associated workflow is presented.
A prospective risk analysis of the adaptive workflow with the Ethos system was performed using Failure Modes and Effects Analysis (FMEA). An interprofessional team collected possible adverse events and evaluated their severity as well as their chance of occurrence and detectability. Measures to reduce the risks were discussed.
A total of 122 events were identified, and scored. Within the 20 events with the highest-ranked risks, the following were identified: Challenges due to the stand-alone software solution with very limited connectivity to the existing record and verify software and digital patient file, unfamiliarity with the new software and its limitations and the adaption process relying on results obtained by artificial intelligence. The risk analysis led to the implementation of additional quality assurance measures in the workflow.
The thorough analysis of the risks associated with the new treatment technique was the basis for designing details of the workflow. The analysis also revealed challenges to be addressed by both, the vendor and customers. On the vendor side, this includes improving communication between their different software solutions. On the customer side, this especially includes establishing validation strategies to monitor the results of the black box adaption process making use of artificial intelligence.
Deep learning advanced to one of the most important technologies in almost all medical fields. Especially in areas, related to medical imaging it plays a big role. However, in interventional radiotherapy (brachytherapy) deep learning is still in an early phase. In this review, first, we investigated and scrutinised the role of deep learning in all processes of interventional radiotherapy and directly related fields. Additionally, we summarised the most recent developments. For better understanding, we provide explanations of key terms and approaches to solving common deep learning problems. To reproduce results of deep learning algorithms both source code and training data must be available. Therefore, a second focus of this work is on the analysis of the availability of open source, open data and open models. In our analysis, we were able to show that deep learning plays already a major role in some areas of interventional radiotherapy, but is still hardly present in others. Nevertheless, its impact is increasing with the years, partly self-propelled but also influenced by closely related fields. Open source, data and models are growing in number but are still scarce and unevenly distributed among different research groups. The reluctance in publishing code, data and models limits reproducibility and restricts evaluation to mono-institutional datasets. The conclusion of our analysis is that deep learning can positively change the workflow of interventional radiotherapy but there is still room for improvements when it comes to reproducible results and standardised evaluation methods.
Multiple sclerosis (MS) is a complex neurodegenerative disorder that affects the brain and spinal cord. In this study, we applied a deep learning-based approach using the StyleGAN model to explore patterns related to MS and predict disease progression in magnetic resonance images (MRI).
We trained the StyleGAN model unsupervised using T1-weighted GRE MR images and diffusion-based ADC maps of MS patients and healthy controls. We then used the trained model to resample MR images from real input data and modified them by manipulations in the latent space to simulate MS progression. We analyzed the resulting simulation-related patterns mimicking disease progression by comparing the intensity profiles of the original and manipulated images and determined the brain parenchymal fraction (BPF).
Our results show that MS progression can be simulated by manipulating MR images in the latent space, as evidenced by brain volume loss on both T1-weighted and ADC maps and increasing lesion extent on ADC maps.
Overall, this study demonstrates the potential of the StyleGAN model in medical imaging to study image markers and to shed more light on the relationship between brain atrophy and MS progression through corresponding manipulations in the latent space.
During the radiation treatment planning process, one of the time-consuming procedures is the final high-resolution dose calculation, which obstacles the wide application of the emerging online adaptive radiotherapy techniques (OLART). There is an urgent desire for highly accurate and efficient dose calculation methods. This study aims to develop a dose super resolution-based deep learning model for fast and accurate dose prediction in clinical practice.
A Multi-stage Dose Super-Resolution Network (MDSR Net) architecture with sparse masks module and multi-stage progressive dose distribution restoration method were developed to predict high-resolution dose distribution using low-resolution data. A total of 340 VMAT plans from different disease sites were used, among which 240 randomly selected nasopharyngeal, lung, and cervix cases were used for model training, and the remaining 60 cases from the same sites for model benchmark testing, and additional 40 cases from the unseen site (breast and rectum) was used for model generalizability evaluation. The clinical calculated dose with a grid size of 2 mm was used as baseline dose distribution. The input included the dose distribution with 4 mm grid size and CT images. The model performance was compared with HD U-Net and cubic interpolation methods using Dose-volume histograms (DVH) metrics and global gamma analysis with 1%/1 mm and 10% low dose threshold. The correlation between the prediction error and the dose, dose gradient, and CT values was also evaluated.
The prediction errors of MDSR were 0.06–0.84% of Dmean indices, and the gamma passing rate was 83.1–91.0% on the benchmark testing dataset, and 0.02–1.03% and 71.3–90.3% for the generalization dataset respectively. The model performance was significantly higher than the HD U-Net and interpolation methods (p < 0.05). The mean errors of the MDSR model decreased (monotonously by 0.03–0.004%) with dose and increased (by 0.01–0.73%) with the dose gradient. There was no correlation between prediction errors and the CT values.
The proposed MDSR model achieved good agreement with the baseline high-resolution dose distribution, with small prediction errors for DVH indices and high gamma passing rate for both seen and unseen sites, indicating a robust and generalizable dose prediction model. The model can provide fast and accurate high-resolution dose distribution for clinical dose calculation, particularly for the routine practice of OLART.
An accurate prognosis of renal function decline in Autosomal Dominant Polycystic Kidney Disease (ADPKD) is crucial for early intervention. Current biomarkers used are height-adjusted total kidney volume (HtTKV), estimated glomerular filtration rate (eGFR), and patient age. However, manually measuring kidney volume is time-consuming and subject to observer variability. Additionally, incorporating automatically generated features from kidney MRI images, along with conventional biomarkers, can enhance prognostic improvement. To address these issues, we developed two deep-learning algorithms. Firstly, an automated kidney volume segmentation model accurately calculates HtTKV. Secondly, we utilize segmented kidney volumes, predicted HtTKV, age, and baseline eGFR to predict chronic kidney disease (CKD) stages =3A, =3B, and a 30% decline in eGFR after 8 years from the baseline visit. Our approach combines a convolutional neural network (CNN) and a multi-layer perceptron (MLP). Our study included 135 subjects and the AUC scores obtained were 0.96, 0.96, and 0.95 for CKD stages =3A, =3B, and a 30% decline in eGFR, respectively. Furthermore, our algorithm achieved a Pearson correlation coefficient of 0.81 between predicted and measured eGFR decline. We extended our approach to predict distinct CKD stages after eight years with an AUC of 0.97. The proposed approach has the potential to enhance monitoring and facilitate prognosis in ADPKD patients, even in the early disease stages.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：