检索结果-内蒙古大学图书馆

Feature-based Response Prediction to Immunotherapy of late-stage Melanoma Patients Using PET/MR Imaging

学校读者我要写书评

暂无评论

Feature-based Response Prediction to Immunotherapy of late-s...

European signal processing Conference (EUSIPCO)

作者： Annika Liebgott Sergios Gatidis Viet Chau Vu Tobias Haueise Konstantin Nikolaou Bin Yang Institute of Signal Processing and System Theory University of Stuttgart Germany University Hospital of Tübingen Germany

ISBN: (数字)9789082797053

ISBN: (纸本)9781728150017

The treatment of malignant melanoma with immunotherapy is a promising approach to treat advanced stages of the disease. However, the treatment can cause serious side effects and not every patient responds to it. This means, crucial time may be wasted on an ineffective treatment. Assessment of the possible therapy response is hence an important research issue. The research presented in this study focuses on the investigation of the potential of medical imaging and machine learning to solve this task. To this end, we extracted image features from multi-modal images and trained a classifier to differentiate non-responsive patients from responsive ones.

关键词： Support vector machines Liver Melanoma signal processing Feature extraction Task analysis Immune system

Automated Multi-Organ Segmentation in Pet Images Using Cascaded Training of a 3d U-Net and Convolutional Autoencoder

学校读者我要写书评

暂无评论

Automated Multi-Organ Segmentation in Pet Images Using Casca...

IEEE International Conference on Acoustics, Speech and signal processing

作者： Annika Liebgott Charlotte Lorenz Sergios Gatidis Viet Chau Vu Konstantin Nikolaou Bin Yang Institute of Signal Processing and System Theory University of Stuttgart Germany University Hospital of Tübingen Germany

PET imaging is an important tool in clinical diagnostics, especially in oncology as it is able to visualize ongoing metabolic processes, e.g. caused by a tumor. Due to the low spatial resolution, a corresponding CT or MRI scan is normally necessary to gain knowledge about the physiological structures of a patient and especially to perform some computer-aided diagnostics methods, e.g. segmentation of structures of interest. As this transfer of information from CT/MRI to the PET domain is not always feasible, e.g. when the corresponding CT or MRI images are unavailable or corrupted by artifacts, we propose a novel approach to perform organ segmentation on the PET images directly. We utilize a CNN architecture based on a 3D U-Net combined with a convolutional autoencoder and train our model purely on PET images and corresponding ground truth masks. Our resulting Dice scores of 0.88, 0.82 an 0.59 for liver, spleen and spine, respectively, show that standalone PET organ segmentation is generally feasible.

关键词： Training Image segmentation Visualization Three-dimensional displays Convolution Magnetic resonance imaging Computed tomography

SELD-TCN: Sound Event Localization & Detection via Temporal Convolutional Networks

学校读者我要写书评

暂无评论

SELD-TCN: Sound Event Localization & Detection via Temporal ...

European signal processing Conference (EUSIPCO)

作者： Karim Guirguis Christoph Schorn Andre Guntoro Sherif Abdulatif Bin Yang Robert Bosch GmbH Renningen Germany Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany

ISBN: (数字)9789082797053

ISBN: (纸本)9781728150017

The understanding of the surrounding environment plays a critical role in autonomous robotic systems, such as self-driving cars. Extensive research has been carried out concerning visual perception. Yet, to obtain a more complete perception of the environment, autonomous systems of the future should also take acoustic information into account. Recent sound event localization and detection (SELD) frameworks utilize convolutional recurrent neural networks (CRNNs). However, considering the recurrent nature of CRNNs, it becomes challenging to implement them efficiently on embedded hardware. Not only are their computations strenuous to parallelize, but they also require high memory bandwidth and large memory buffers. In this work, we develop a more robust and hardware-friendly novel architecture based on a temporal convolutional network (TCN). The proposed framework (SELD-TCN) outperforms the state-of-the-art SELDnet performance on four different datasets. Moreover, SELD-TCN achieves 4x faster training time per epoch and 40x faster inference time on an ordinary graphics processing unit (GPU).

关键词： Training Recurrent neural networks Convolution Memory management Graphics processing units Task analysis Visual perception

A Concept for Myocardial Current Density Estimation with Magnetoelectric Sensors

学校读者我要写书评

暂无评论

A Concept for Myocardial Current Density Estimation with Mag...

作者： Schmidt, Gerhard Engelhardt, Erik Elzenheimer, Eric Hoffman, Johannes Schmidt, Tobias Zaman, Adrian Frey, Norbert Digital Signal Processing and System Theory Institute of Electrical Engineering and Information Technology Faculty of Engineering Kiel University Kaiserstr. 2 Kiel24143 Germany Medical Department III Specialised in Cardiology and Angiology University Medical Center Schleswig-Holstein Arnold-Heller-Str. 3 / House K3 Kiel24105 Germany Internal Medicine III: Heart Vascular and Lung University Medical Center Heidelberg Im Neuenheimer Feld 410 Heidelberg69120 Germany

In this paper we present a novel noninvasive approach to estimate current densities in the heart from magnetocardiography. The proposed algorithm uses nested optimization to model current densities in equally-sized voxels of myocardial tissue. First-order Thiran all-pass filters are used to describe the propagation between *** demonstrate feasibility of the algorithm for a noise-free single-layer simulation. However, challenges remain, such as addressing measurement noise and optimizing propagation velocity. Overall, this approach has the potential to complement or replace invasive catheter-based electrophysiological studies for localization of arrhythmogenic tissue. © 2023 the author(s), published by Walter de Gruyter Berlin/Boston.

关键词： Magnetocardiography

Improving uncertainty of deep learning-based object classification on radar spectra using label smoothing

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Patel, Kanil Beluch, William Rambach, Kilian Pfeiffer, Michael Yang, Bin Bosch Center for Artificial Intelligence Renningen Germany Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany

Object type classification for automotive radar has greatly improved with recent deep learning (DL) solutions, however these developments have mostly focused on the classification accuracy. Before employing DL solutions in safety-critical applications, such as automated driving, an indispensable prerequisite is the accurate quantification of the classifiers’ reliability. Unfortunately, DL classifiers are characterized as black-box systems which output severely over-confident predictions, leading downstream decision-making systems to false conclusions with possibly catastrophic consequences. We find that deep radar classifiers maintain high-confidences for ambiguous, difficult samples, e.g. small objects measured at large distances, under domain shift and signal corruptions, regardless of the correctness of the predictions. The focus of this article is to learn deep radar spectra classifiers which offer robust real-time uncertainty estimates using label smoothing during training. Label smoothing is a technique of refining, or softening, the hard labels typically available in classification datasets. In this article, we exploit radar-specific know-how to define soft labels which encourage the classifiers to learn to output high-quality calibrated uncertainty estimates, thereby partially resolving the problem of over-confidence. Our investigations show how simple radar knowledge can easily be combined with complex data-driven learning algorithms to yield safe automotive radar perception. Copyright © 2021, The Authors. All rights reserved.

关键词： Decision making

On-manifold Adversarial Data Augmentation Improves Uncertainty Calibration

学校读者我要写书评

暂无评论

On-manifold Adversarial Data Augmentation Improves Uncertain...

International Conference on Pattern Recognition

作者： Kanil Patel William Beluch Dan Zhang Michael Pfeiffer Bin Yang Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany Bosch Center for Artificial Intelligence Renningen Germany

ISBN: (纸本)9781728188089;9781728188096

Uncertainty estimates help to identify ambiguous, novel, or anomalous inputs, but the reliable quantification of uncertainty has proven to be challenging for modern deep networks. To improve uncertainty estimation, we propose On-Manifold Adversarial Data Augmentation or OMADA, which specifically attempts to generate challenging examples by following an on-manifold adversarial attack path in the latent space of an autoencoder that closely approximates the decision boundaries between classes. On a variety of datasets and for multiple network architectures, OMADA consistently yields more accurate and better calibrated classifiers than baseline models, and outperforms competing approaches such as Mixup, as well as achieving similar performance to (at times better than) postprocessing calibration methods such as temperature scaling. Variants of OMADA can employ different sampling schemes for ambiguous on-manifold examples based on the entropy of their estimated soft labels, which exhibit specific strengths for generalization, calibration of predicted uncertainty, or detection of out-of-distribution inputs.

关键词： Manifolds Training Uncertainty Temperature Estimation Stochastic processes Network architecture

Investigation of uncertainty of deep learning-based object classification on radar spectra

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Patel, Kanil Beluch, William Rambach, Kilian Cozma, Adriana-Eliza Pfeiffer, Michael Yang, Bin Bosch Center for Artificial Intelligence Renningen Germany Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany

Deep learning (DL) has recently attracted increasing interest to improve object type classification for automotive radar. In addition to high accuracy, it is crucial for decision making in autonomous vehicles to evaluate the reliability of the predictions;however, decisions of DL networks are non-transparent. Current DL research has investigated how uncertainties of predictions can be quantified, and in this article, we evaluate the potential of these methods for safe, automotive radar perception. In particular we evaluate how uncertainty quantification can support radar perception under (1) domain shift, (2) corruptions of input signals, and (3) in the presence of unknown objects. We find that in agreement with phenomena observed in the literature, deep radar classifiers are overly confident, even in their wrong predictions. This raises concerns about the use of the confidence values for decision making under uncertainty, as the model fails to notify when it cannot handle an unknown situation. Accurate confidence values would allow optimal integration of multiple information sources, e.g. via sensor fusion. We show that by applying state-of-the-art post-hoc uncertainty calibration, the quality of confidence measures can be significantly improved, thereby partially resolving the over-confidence problem. Our investigation shows that further research into training and calibrating DL networks is necessary and offers great potential for safe automotive object classification with radar sensors. Copyright © 2021, The Authors. All rights reserved.

关键词： Decision making

Benchmarking Dependence Measures to Prevent Shortcut Learning in Medical Imaging

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Müller, Sarah Fay, Louisa Koch, Lisa M. Gatidis, Sergios Küstner, Thomas Berens, Philipp Hertie Institute for AI in Brain Health University of Tübingen Germany Medical Image and Data Analysis University Hospital of Tübingen Germany Institute of Signal Processing and System Theory University of Stuttgart Germany Department of Diabetes Endocrinology Nutritional Medicine and Metabolism UDEM Inselspital Bern University Hospital University of Bern Switzerland Stanford University Department of Radiology Stanford United States

Medical imaging cohorts are often confounded by factors such as acquisition devices, hospital sites, patient backgrounds, and many more. As a result, deep learning models tend to learn spurious correlations instead of causally related features, limiting their generalizability to new and unseen data. This problem can be addressed by minimizing dependence measures between intermediate representations of task-related and non-task-related variables. These measures include mutual information, distance correlation, and the performance of adversarial classifiers. Here, we benchmark such dependence measures for the task of preventing shortcut learning. We study a simplified setting using Morpho-MNIST and a medical imaging task with CheXpert chest radiographs. Our results provide insights into how to mitigate confounding factors in medical imaging. The project’s code is publicly available1 Copyright © 2024, The Authors. All rights reserved.

关键词： Adversarial machine learning

SLPC: A VRNN-based approach for stochastic lidar prediction and completion in autonomous driving

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Eskandar, George Braun, Alexander Meinke, Martin Armanious, Karim Yang, Bin University of Stuttgart Institute of Signal Processing and System Theory Stuttgart Germany Robert Bosch GmbH Lidar Perception Stuttgart Germany

Predicting future 3D LiDAR pointclouds is a challenging task that is useful in many applications in autonomous driving such as trajectory prediction, pose forecasting and decision making. In this work, we propose a new LiDAR prediction framework that is based on generative models namely Variational Recurrent Neural Networks (VRNNs), titled Stochastic LiDAR Prediction and Completion (SLPC). Our algorithm is able to address the limitations of previous video prediction frameworks when dealing with sparse data by spatially inpainting the depth maps in the upcoming frames. Our contributions can thus be summarized as follows: we introduce the new task of predicting and completing depth maps from spatially sparse data, we present a sparse version of VRNNs and an effective self-supervised training method that does not require any labels. Experimental results illustrate the effectiveness of our framework in comparison to the state of the art methods in video prediction. Copyright © 2021, The Authors. All rights reserved.

关键词： Autonomous vehicles