NIMG-73. ACCURATE 3D GBM SEGMENTATION IN 30 SECONDS

The number of medical imaging devices is quickly and steadily rising, generating an increasing amount of image records day by day. The number of qualified human experts able to handle this data cannot follow this trend, so there is a strong need to develop reliable automatic segmentation and decision support algorithms. The Brain Tumor Segmentation Challenge (BraTS), first organized seven years ago, provoked a strong intensification of the development of brain tumor detection and segmentation algorithms. Beside many others, several ensemble learning solutions have been proposed lately to the above mentioned problem. This study presents an evaluation framework developed to evaluate the accuracy and efficiency of these algorithms deployed in brain tumor segmentation, based on the BraTS 2016 train data set. All evaluated algorithms proved suitable to provide acceptable accuracy in segmentation, but random forest was found the best, both in terms of precision and efficiency.

3D Freehand Ultrasound Reconstruction of the Carotid Artery.

There are two inherent obstacles to effectively using Graphics Processing Units (GPUs) for query processing in search engines: (a) the highly restricted GPU memory space, and (b) the CPU-GPU transfer latency. Previously, Ao et al. presented a GPU method for lists intersection, an essential component in AND-based query processing. However, this work assumes the whole inverted index can be stored in GPU memory and does not address document ranking. In this paper, we describe and analyze a GPU query processing method which incorporates both lists intersection and top-k ranking. We introduce a parameterized pruned posting list GPU caching method where the parameter determines how much GPU memory is used for caching. This method allows list caching for large inverted indexes using the limited GPU memory, thereby making a qualitative improvement over previous work. We also give a mathematical model which can identify an approximately optimal choice of the parameter. Experimental results indicate that this GPU approach under the pruned list caching policy achieves better query throughput than its CPU counterpart, even when the inverted index size is much larger than the GPU memory space.

The graphics processing unit (GPU) is extensively used in diverse domains, such as finance, machine learning, and image processing. The GPU can be underutilized as multiple applications may not share the same GPU concurrently owing to a memory oversubscription issue. For example, when applications that require fewer computational resources but a larger GPU memory are running instantaneously, the GPU memory may be insufficient; consequently, the number of GPU applications running simultaneously is restricted, decreasing GPU utilization. Further, it can even stop the execution of applications that are running on the GPU. To this end, we propose FlexGPU, which schedules the kernels of the GPU applications that run on the same GPU according to their features. This framework 1) schedules the kernel at the launching time according to its features to improve GPU utilization and 2) temporarily checkpoints and restores non-dependent content in the GPU memory to/from the host memory, which avoids oversubscription of the GPU when out-of-memory failure occurs and allows more kernels to run concurrently on the GPU. The experimental results show that compared to existing methods, our approach demonstrates a 7 times improvement in performance in terms of execution time and enables a 2.5 times increase in the concurrent execution of applications.

Brain tumors are one of the most common neurological complications that affect millions each year. Current studies have focused only on either the segmentation of brain tumors or healthy anatomical regions. Unfortunately, there is a lack of research on whole brain segmentation of patients with existing tumors. The segmentation of these regions is cruical for understanding patient pathologies. Due to the distortions in the brain caused by tumors, whole-brain segmentation models trained on healthy brains will not perform accurately on diseased ones. To address this issue, this paper proposes a novel unified 3D nnU-Net model for the segmentation of both brain tumor and whole-brain structures for the first time. Training a model on datasets that include both healthy and cancerous brain regions could enhance the generalizability of machine learning (ML) models for performing both whole-brain segmentation and tumor segmentation tasks. Currently, there are no datasets that are publicly available for this purpose. This paper proposes to address this challenge by augmenting data from OASIS-1 and BraTS2023 to train the baseline model. The model successfully achieves a Dice Similarity Coefficient (DSC) of 0.814 for whole brain structures and 0.874 for tumor segmentation, demonstrating its effectiveness across diverse regions.

In data-parallel computing that uses a graphic processing unit (GPU), processing of large data requires that multiple GPUs be used in the computer to increase its execution performance. Increasing processing performance by using multiple computing resources has been enabled by the development of computing frameworks based on a directed acyclic graph (DAG). However, their performance degrades in out-of-core processing, which often occurs in processing of large data on GPUs with limited memory capacity. The GPU data input/output (I/O) for data swapping between host memory and GPU memory during the execution of a user DAG is usually a performance bottleneck. A computing framework called is proposed to overcome this drawback. It uses a novel scheduler that involves two methods to minimize the total amount of GPU data I/O of data swapping. First, it performs locality-aware scheduling. When it schedules a task, it selects one that requires the minimum amount of data swapping and reuses as much of the data residing in GPU memory as possible. Second, it extends the locality-aware scheduling so that GPUs can execute data prefetching. Prefetching data that are swapped out from a GPU enables efficient use of bottleneck GPU I/O resources. To prefetch the input data of future tasks, it is required to determine the schedule of future tasks. Victream's scheduler (hereafter, the Victream scheduler) extends the locality-aware scheduling so that it can schedule future tasks to enable data prefetching that is executed in the way that minimizes the amount of data I/O of data swapping. Evaluation of a Victream prototype showed that the performance of Victream is better than that of conventional frameworks by up to 117%.

Detection and segmentation of a brain tumor such as glioblastoma multiforme (GBM) in magnetic resonance (MR) images are often challenging due to its intrinsically heterogeneous signal characteristics. A robust segmentation method for brain tumor MRI scans was developed and tested. Simple thresholds and statistical methods are unable to adequately segment the various elements of the GBM, such as local contrast enhancement, necrosis, and edema. Most voxel-based methods cannot achieve satisfactory results in larger data sets, and the methods based on generative or discriminative models have intrinsic limitations during application, such as small sample set learning and transfer. A new method was developed to overcome these challenges. Multimodal MR images are segmented into superpixels using algorithms to alleviate the sampling issue and to improve the sample representativeness. Next, features were extracted from the superpixels using multi-level Gabor wavelet filters. Based on the features, a support vector machine (SVM) model and an affinity metric model for tumors were trained to overcome the limitations of previous generative models. Based on the output of the SVM and spatial affinity models, conditional random fields theory was applied to segment the tumor in a maximum a posteriori fashion given the smoothness prior defined by our affinity model. Finally, labeling noise was removed using "structural knowledge" such as the symmetrical and continuous characteristics of the tumor in spatial domain. The system was evaluated with 20GBM cases and the BraTS challenge data set. Dice coefficients were computed, and the results were highly consistent with those reported by Zikic et al. (MICCAI 2012, Lecture notes in computer science. vol 7512, pp 369-376, 2012). A brain tumor segmentation method using model-aware affinity demonstrates comparable performance with other state-of-the art algorithms.

Purpose: Accurate and fast delineation of targets and normal structures based on daily CTs is critical for online adaptive radiotherapy (ART). Manual contouring generally takes too long to be practical. Auto-segmentation can run in a few minutes on modern computers, but it is desirable to reduce this time further. A simple, inexpensive option to accelerate auto-segmentation is to use a graphics processor unit (GPU). This work aims to evaluate the performance of a GPU-accelerated auto-segmentation. Method and Materials: A GPU card (NVidia GTX 285) was added to a computer consisting of two 3GHz four core CPUs running auto-segmentation software, ABAS (Atlas Based Auto Segmentation, Elekta CMS Software). To evaluate performance, this setup was tested using daily CTs acquired during IGRT using a CT-on-Rails (CTVision, Siemens) for 14 breast (4 prone, 10 supine) and 6 prostate cancer patients. Contours of targets and critical structures were populated from planning CTs to daily CTs using ABAS with and without the GPU. The accuracy of GPU-accelerated contouring was measured using Dice's Coefficient (DC). Results: For prostate cases, there is no noticeable difference between contours generated with and without GPU (<2% difference in DC to physician contours). The average time to segment was reduced by 30%, to 1.4 minutes, with GPU, (as low as 0.97 minutes). For prone breast cases, except for lumpectomy cavity (DC: 77 – 93%), there is also no substantial difference between contours with or without GPU (DC>95%). The average time to segment images was reduced by 33% with GPU (to 2.8 minutes). For supine breast cases, contours generated with and without GPU agreed equally with contours drawn by physicians. The GPU average segmentation time was 2.3 minutes, reduced by 47%. Conclusion: The use of GPU can substantially accelerate auto-segmentation without degrading the segmentation accuracy. This time saving is significant for online ART.

Feedback control has become a crucial tool in the research on magnetic confinement of plasmas for achieving controlled nuclear fusion. This thesis presents a novel plasma feedback control system that, for the first time, employs a Graphics Processing Unit (GPU) for microsecond-latency, real-time control computations. This novel application area for GPU computing is opened up by a new system architecture that is optimized for low-latency computations on less than kilobyte sized data samples as they occur in typical plasma control algorithms. In contrast to traditional GPU computing approaches that target complex, high-throughput computations with massive amounts of data, the architecture presented in this thesis uses the GPU as the primary processing unit rather than as an auxiliary of the CPU, and data is transferred from A-D/D-A converters directly into GPU memory using peer-to-peer PCI Express transfers. The described design has been implemented in a new, GPU-based control system for the High-Beta Tokamak – Extended Pulse (HBT-EP) device. The system is built from commodity hardware and uses an NVIDIA GeForce GPU and D-TACQ A-D/D-A converters providing a total of 96 input and 64 output channels. The system is able to run with sampling periods down to 4 μs and latencies down to 8 μs. The GPU provides a total processing power of 1.5 x 1012 floating point operations per second. To illustrate the performance and versatility of both the general architecture and concrete implementation, a new control algorithm has been developed. The algorithm is designed for the control of multiple rotating magnetic perturbations in situations where the plasma equilibrium is not known exactly and features an adaptive system model: instead of requiring the rotation frequencies and growth rates embedded in the system model to be set a priori, the adaptive algorithm derives these parameters from the evolution of the perturbation amplitudes themselves. This results in non-linear control computations with high computational demands, but is handled easily by the GPU based system. Both digital processing latency and an arbitrary multi-pole response of amplifiers and control coils is fully taken into account for the generation of control signals. To separate sensor signals into perturbed and equilibrium components without knowledge of the equilibrium fields, a new separation method based on biorthogonal decomposition is introduced and used to derive a filter that performs the separation in real-time. The control algorithm has been implemented and tested on the new, GPU-based feedback control system of the HBT-EP tokamak. In this instance, the algorithm was set up to control four rotating n = 1 perturbations at different poloidal angles. The perturbations were treated as coupled in frequency but independent in amplitude and phase, so that the system effectively controls a helical n = 1 perturbation with unknown poloidal spectrum. Depending on the plasma's edge safety factor and rotation frequency, the control system is shown to be able to suppress the amplitude of the dominant 8 kHz mode by up to 60% or amplify the saturated amplitude by a factor of up to two. Intermediate feedback phases combine suppression and amplification with a speed up or slow down of the mode rotation frequency. Increasing feedback gain results in the excitation of an additional, slowly rotating 1.4 kHz mode without further effects on the 8 kHz mode. The feedback performance is found to exceed previous results obtained with an FPGA- and Kalman-filter based control system without requiring any tuning of system model parameters. Experimental results are compared with simulations based on a combination of the Boozer surface current model and the Fitzpatrick-Aydemir model. Within the subset of phenomena that can be represented by the model as well as determined experimentally, qualitative agreement is found.

Graphics processing units (GPUs) are important components of modern computing devices for not only graphics rendering, but also efficient parallel computations. However, their security problems are ignored despite their importance and popularity. In this paper, we first perform an in-depth security analysis on GPUs to detect security vulnerabilities. We observe that contemporary, widely-used GPUs, both NVIDIA's and AMD's, do not initialize newly allocated GPU memory pages which may contain sensitive user data. By exploiting such vulnerabilities, we propose attack methods for revealing a victim program's data kept in GPU memory both during its execution and right after its termination. We further show the high applicability of the proposed attacks by applying them to the Chromium and Firefox web browsers which use GPUs for accelerating webpage rendering. We detect that both browsers leave rendered webpage textures in GPU memory, so that we can infer which web pages a victim user has visited by analyzing the remaining textures. The accuracy of our advanced inference attack that uses both pixel sequence matching and RGB histogram matching is up to 95.4%.

Brain tumors are considered as most fatal cancers. To reduce the risk of death, early identification of the disease is required. One of the best available methods to evaluate brain tumors is Magnetic resonance Images (MRI). Brain tumor detection and segmentation are tough as brain tumors may vary in size, shape, and location. That makes manual detection of brain tumors by exploring MRI a tedious job for radiologists and doctors’. So an automated brain tumor detection and segmentation is required. This work suggests a Region-based Convolution Neural Network (RCNN) approach for automated brain tumor identification and segmentation using MR images, which helps solve the difficulties of brain tumor identification efficiently and accurately. Our methodology is based on the accurate and efficient selection of tumorous areas. That reduces computational complexity and time. We have validated the designed experimental setup on a standard dataset, BraTS 2020. We used binary evaluation matrices based on Dice Similarity Coefficient (DSC) and Mean Average Precision (mAP). The segmentation results are compared with state-of-the-art methodologies to demonstrate the effectiveness of the proposed method. The suggested approach attained an average DSC of 0.92 and mAP 0.92 for 10 patients, while on the whole dataset, the scores are DSC 0.89 and mAP 0.90. The following results clearly show the performance efficiency of the proposed methodology.

Intraoperative ultrasound (iUS) is increasingly used in pediatric brain tumor surgery. However, accurate tumor tissue identification and localization within the patient with iUS remains challenging. Automatic tumor segmentation may overcome these difficulties by providing quicker, more consistent, and reliable image analysis. This study evaluated the feasibility of automatic 3-dimensional segmentation of pediatric brain tumors using navigated iUS. A retrospective analysis was conducted on 109 navigated iUS acquisitions from 60 pediatric patients who underwent neurosurgical tumor resection. 3-dimensional nnU-Net was trained on 79 preresectional iUS acquisitions with corresponding tumor segmentations. Evaluation was performed on 30 iUS acquisitions against segmentations validated by 2 neurosurgeons. Performance benchmarks were set at a minimum Dice Similarity Coefficient (DSC) of 0.85 and a maximum mean surface distance and 95th percentile Hausdorff distance of 5 mm. The trained algorithm predicted tumor segmentations for all acquisitions in the test set, which met the predefined benchmark values with a median DSC of 0.89 (IQR: 0.13), a median mean surface distance of 1.1 mm (IQR: 1.0 mm), and a median 95th percentile Hausdorff distance of 5.6 mm (IQR: 3.5 mm). The lowest quartile of DSC values consisted of 5 patients, including craniopharyngiomas (n = 3) and high-grade gliomas (n = 2). Oversegmentation of the lateral ventricle as tumor was observed in 2 patients. In 1 high-grade glioma case, edema was misclassified as tumor because of unclear tumor boundaries. This study demonstrates that automatic segmentation of pediatric brain tumors using navigated iUS is feasible and achieved high performance in most cases, with few outliers that showed moderate performance in tumor prediction. Future efforts will focus on real-time segmentation of residual tumor tissue during tumor resection. Ultimately, real-time visualization of automatically segmented tumors in 3 dimensions could improve intraoperative navigation and support neurosurgical decision making.

Glioblastoma is an aggressive type of cancer that can develop in the brain or spinal cord. Magnetic Resonance Imaging (MRI) is key to diagnosing and tracking brain tumors in clinical settings. Brain tumor segmentation in MRI is required for disease diagnosis, surgical planning, and prognosis. As these tumors are heterogeneous in shape and appearance, their segmentation becomes a challenging task. The performance of automated medical image segmentation has considerably improved because of recent advances in deep learning. Introducing context encoding with deep CNN models has shown promise for semantic segmentation of brain tumors. In this work, we use a 3D UNet-Context Encoding (UNCE) deep learning network for improved brain tumor segmentation. Further, we introduce epistemic and aleatoric Uncertainty Quantification (UQ) using Monte Carlo Dropout (MCDO) and Test Time Augmentation (TTA) with the UNCE deep learning model to ascertain confidence in tumor segmentation performance. We build our model using the training MRI image sets of RSNA-ASNR-MICCAI Brain Tumor Segmentation (BraTS) Challenge 2021. We evaluate the model performance using the validation and test images from the BraTS challenge dataset. Online evaluation of validation data shows dice score coefficients (DSC) of 0.7787, 0.8499, and 0.9159 for enhancing tumor (ET), tumor core (TC), and whole tumor (WT), respectively. The dice score coefficients of the test datasets are 0.6684 for ET, 0.7056 for TC, and 0.7551 for WT, respectively.

In this paper, a hybrid approach using sliding window mechanism followed by fuzzy c means clustering is proposed for the automated brain tumour extraction. The proposed method consists three phases. The first phase is used for detecting the tumorous brain MR scans by implementing pre-processing techniques followed by texture features extraction and classification. Further, this phase also compares the performance of different classifiers. The second phase consists of the localization of the tumorous region using sliding window mechanism, in which a sized window sweeps through the whole tumorous MR scan and the window is classified as tumorous or non-tumorous. The third phase consists of fuzzy c means clustering to get the exact location of the tumour by removing the misclassified windows obtained from Phase 2. 2D single-spectral anatomical FLAIR MRI scans are considered for experiment. Outcomes demonstrate significant results in terms of sensitivity, specificity, accuracy, dice similarity coefficient in comparison with the other existing methods.

Background Missing MRI sequences represent an obstacle in the development and use of deep learning (DL) models that require multiple inputs. Purpose To determine if synthesizing brain MRI scans using generative adversarial networks (GANs) allows for the use of a DL model for brain lesion segmentation that requires T1-weighted images, postcontrast T1-weighted images, fluid-attenuated inversion recovery (FLAIR) images, and T2-weighted images. Materials and Methods In this retrospective study, brain MRI scans obtained between 2011 and 2019 were collected, and scenarios were simulated in which the T1-weighted images and FLAIR images were missing. Two GANs were trained, validated, and tested using 210 glioblastomas (GBMs) (Multimodal Brain Tumor Image Segmentation Benchmark [BRATS] 2017) to generate T1-weighted images from postcontrast T1-weighted images and FLAIR images from T2-weighted images. The quality of the generated images was evaluated with mean squared error (MSE) and the structural similarity index (SSI). The segmentations obtained with the generated scans were compared with those obtained with the original MRI scans using the dice similarity coefficient (DSC). The GANs were validated on sets of GBMs and central nervous system lymphomas from the authors' institution to assess their generalizability. Statistical analysis was performed using the Mann-Whitney, Friedman, and Dunn tests. Results Two hundred ten GBMs from the BRATS data set and 46 GBMs (mean patient age, 58 years ± 11 [standard deviation]; 27 men [59%] and 19 women [41%]) and 21 central nervous system lymphomas (mean patient age, 67 years ± 13; 12 men [57%] and nine women [43%]) from the authors' institution were evaluated. The median MSE for the generated T1-weighted images ranged from 0.005 to 0.013, and the median MSE for the generated FLAIR images ranged from 0.004 to 0.103. The median SSI ranged from 0.82 to 0.92 for the generated T1-weighted images and from 0.76 to 0.92 for the generated FLAIR images. The median DSCs for the segmentation of the whole lesion, the FLAIR hyperintensities, and the contrast-enhanced areas using the generated scans were 0.82, 0.71, and 0.92, respectively, when replacing both T1-weighted and FLAIR images; 0.84, 0.74, and 0.97 when replacing only the FLAIR images; and 0.97, 0.95, and 0.92 when replacing only the T1-weighted images. Conclusion Brain MRI scans generated using generative adversarial networks can be used as deep learning model inputs in case MRI sequences are missing. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Zhong in this issue. An earlier incorrect version of this article appeared online. This article was corrected on April 12, 2021.