检索结果-内蒙古大学图书馆

Prompt Tuning for Parameter-efficient Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Fischer, Marc Bartler, Alexander Yang, Bin Institute of Signal Processing and System Theory University of Stuttgart Stuttgart70550 Germany

Neural networks pre-trained on a self-supervision scheme have become the standard when operating in data rich environments with scarce annotations. As such, fine-tuning a model to a downstream task in a parameter-efficient but effective way, e.g. for a new set of classes in the case of semantic segmentation, is of increasing importance. In this work, we propose and investigate several contributions to achieve a parameter-efficient but effective adaptation for semantic segmentation on two medical imaging datasets. Relying on the recently popularized prompt tuning approach, we provide a prompt-able UNet (PUNet) architecture, that is frozen after pre-training, but adaptable throughout the network by class-dependent learnable prompt tokens. We pre-train this architecture with a dedicated dense self-supervision scheme based on assignments to online generated prototypes (contrastive prototype assignment, CPA) of a student teacher combination alongside a concurrent segmentation loss on a subset of classes. We demonstrate that the resulting neural network model is able to attenuate the gap between fully fine-tuned and parameter-efficiently adapted models on CT imaging datasets. As such, the difference between fully fine-tuned and prompt-tuned variants amounts to only 3.83 pp for the TCIA/BTCV dataset and 2.67 pp for the CT-ORG dataset in the mean Dice Similarity Coefficient (DSC, in %) while only prompt tokens, corresponding to 0.85% of the pre-trained backbone model with 6.8M frozen parameters, are adjusted. The code for this work is available on https://***/marcdcfischer/PUNet. © 2022, CC BY-NC-ND.

关键词： Semantic Segmentation

CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Abdulatif, Sherif Cao, Ruizhe Yang, Bin The Institute of Signal Processing and System Theory University of Stuttgart Stuttgart70569 Germany

In this work, we further develop the conformer-based metric generative adversarial network (CMGAN) model1 for speech enhancement (SE) in the time-frequency (TF) domain. This paper builds on our previous work but takes a more in-depth look by conducting extensive ablation studies on model inputs and architectural design choices. We rigorously tested the generalization ability of the model to unseen noise types and distortions. We have fortified our claims through DNSMOS measurements and listening tests. Rather than focusing exclusively on the speech denoising task, we extend this work to address the dereverberation and super-resolution tasks. This necessitated exploring various architectural changes, specifically metric discriminator scores and masking techniques. It is essential to highlight that this is among the earliest works that attempted complex TF-domain super-resolution. Our findings show that CMGAN outperforms existing state-of-the-art methods in the three major speech enhancement tasks: denoising, dereverberation, and super-resolution. For example, in the denoising task using the Voice Bank+DEMAND dataset, CMGAN notably exceeded the performance of prior models, attaining a PESQ score of 3.41 and an SSNR of 11.10 dB. Audio samples and CMGAN implementations are available online2 Copyright © 2022, The Authors. All rights reserved.

关键词： Generative adversarial networks

Automatic Localization of an Ultrasound Probe with the Help of Magnetic Sensors

学校读者我要写书评

暂无评论

Automatic Localization of an Ultrasound Probe with the Help ...

作者： Bald, Christin Bergholz, Robert Schmidt, Gerhard Digital Signal Processing and System Theory Institute of Electrical Engineering and Information Technology Faculty of Engineering Christian-Albrechts-Universität zu Kiel Kaiserstraße 2 Kiel Germany Digital Signal Processing and System Theory Institute of Electrical Engineering and Information Technology Faculty of Engineering Christian-Albrechts-Universität zu Kiel Kiel Germany Pediatric Surgery Faculty of Medicine Christian-Albrechts-Universität zu Kiel Kiel Germany

Ultrasound measurements are a widely used instrument in clinical practice. For later traceability of the images, the position (and orientation) of the ultrasound probe must be recorded during the measurement. Until now this has to be done manually by the physician. An easier and more accurate approach would be the automatic tracking of the ultrasound probe. This contribution shows a first approach for automatically localizing the ultrasonic head during measurement. The proposed method is based on coils surrounding the patient bed and a 3D magnetic sensor placed on the ultrasound head. Besides some pre- and postprocessing steps, the proposed localization algorithm is based on trilateration followed by a least mean squares approach for refinement of the estimation. In a first proof-of-concept measurement with fixed positions and orientations of the ultrasound head a mean accuracy of 2.85 cm and 8.94◦ was achieved. Additionally, a measurement with a moving ultrasound head is presented to demonstrate the real-time capability of the system. Finally, future steps for improving the automatic measurement are discussed, including a graphical user interface for the physician and the use of magnetoelectric sensors for measurement. © 2022 The Author(s), published by De Gruyter.

关键词： Ultrasonics

MULTI-CLASS UNCERTAINTY CALIBRATION VIA MUTUAL INFORMATION MAXIMIZATION-BASED BINNING 9

学校读者我要写书评

暂无评论

MULTI-CLASS UNCERTAINTY CALIBRATION VIA MUTUAL INFORMATION M...

9th International Conference on Learning Representations, ICLR 2021

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

Post-hoc multi-class calibration is a common approach for providing high-quality confidence estimates of deep neural network predictions. Recent work has shown that widely used scaling methods underestimate their calibration error, while alternative Histogram Binning (HB) methods often fail to preserve classification accuracy. When classes have small prior probabilities, HB also faces the issue of severe sample-inefficiency after the conversion into K one-vs-rest class-wise calibration problems. The goal of this paper is to resolve the identified issues of HB in order to provide calibrated confidence estimates using only a small holdout calibration dataset for bin optimization while preserving multi-class ranking accuracy. From an information-theoretic perspective, we derive the I-Max concept for binning, which maximizes the mutual information between labels and quantized logits. This concept mitigates potential loss in ranking performance due to lossy quantization, and by disentangling the optimization of bin edges and representatives allows simultaneous improvement of ranking and calibration performance. To improve the sample efficiency and estimates from a small calibration set, we propose a shared class-wise (sCW) calibration strategy, sharing one calibrator among similar classes (e.g., with similar class priors) so that the training sets of their class-wise calibration problems can be merged to train the single calibrator. The combination of sCW and I-Max binning outperforms the state of the art calibration methods on various evaluation metrics across different benchmark datasets and models, using a small calibration set (e.g., 1k samples for ImageNet). © 2021 ICLR 2021 - 9th International Conference on Learning Representations. All rights reserved.

关键词： Calibration

Deep Regression for Biological Age Estimation in Multiple Organs: Investigations on 40,000 Subjects of the UK Biobank

学校读者我要写书评

暂无评论

Deep Regression for Biological Age Estimation in Multiple Or...

International Conference on Acoustics, Speech, and signal processing (ICASSP)

作者： Veronika Ecker Marcel Früh Bin Yang Sergios Gatidis Thomas Küstner Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany Department of Radiology MIDAS.lab University Hospital of Tübingen Tübingen Germany

Age plays an important role in shaping medical decisions, but the biological changes associated with aging do not solely depend on the chronological age. Genetics, lifestyle, and environment cause variations in age-related characteristics, even within the same chronological age group. Biological age (BA) was introduced to better capture an individual’s actual aging process, although its imprecise definition remains a challenge. Organ systems can age at different rates, necessitating organ-specific BA evaluation. To our knowledge, there have been no studies assessing age regionally across multiple organ systems. These age estimations, reflecting various body parts, enable a comprehensive patient-based analysis. We conducted brain, heart, kidney, liver, spleen, pancreas, and retinal fundus age estimations using MRI and OCT scans in 40,000 subjects of the UK Biobank with an uncertainty-aware ResNet-based network. Our results demonstrate the feasibility of organ-specific age estimation with cross-organ correlations of age-related changes. We achieve a mean age difference between predicted and chronological age of 2.72 years across all organs and an averaged Pearson correlation coefficient of 0.87.

关键词：

Indoor Positioning Based on Active Radar Sensing and Passive Reflectors: Reflector Placement Optimization

学校读者我要写书评

暂无评论

Indoor Positioning Based on Active Radar Sensing and Passive...

International Conference on Indoor Positioning and Indoor Navigation (IPIN)

作者： Sven Hinderer Pascal Schlachter Zhibin Yu Xiaofeng Wu Bin Yang Institute of Signal Processing and System Theory University of Stuttgart Stuttgart Germany Munich Research Center Huawei Technologies Duesseldorf GmbH Munich Germany

We extend our work on a novel indoor positioning system (IPS) for autonomous mobile robots (AMRs) based on radar sensing of local, passive radar reflectors. Through the combination of simple reflectors and a single-channel frequency modulated continuous wave (FMCW) radar, high positioning accuracy at low system cost can be achieved. Further, a multi-objective (MO) particle swarm optimization (PSO) algorithm is presented that optimizes the 2D placement of radar reflectors in complex room settings.

关键词：

Feature-based response prediction to immunotherapy of late-stage melanoma patients using PET/MR imaging 28

学校读者我要写书评

暂无评论

Feature-based response prediction to immunotherapy of late-s...

28th European signal processing Conference, EUSIPCO 2020

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

ISBN: (纸本)9789082797053

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. © 2021 European signal processing Conference, EUSIPCO. All rights reserved.

关键词： Support vector machines

Fine-Tuning Strategies for Continual Online EEG Motor Imagery Decoding: Insights from a Large-Scale Longitudinal Study

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wimpff, Martin Aristimunha, Bruno Chevallier, Sylvain Yang, Bin Institute of Signal Processing and System Theory University of Stuttgart Stuttgart70569 Germany Inria TAU Team LISN-CNRS Universite Paris-Saclay Orsay91400 France Federal University of ABC Santo Andre Brazil

This study investigates continual fine-tuning strategies for deep learning in online longitudinal electroencephalography (EEG) motor imagery (MI) decoding within a causal setting involving a large user group and multiple sessions per participant. We are the first to explore such strategies across a large user group, as longitudinal adaptation is typically studied in the single-subject setting with a single adaptation strategy, which limits the ability to generalize findings. First, we examine the impact of different fine-tuning approaches on decoder performance and stability. Building on this, we integrate online test-time adaptation (OTTA) to adapt the model during deployment, complementing the effects of prior fine-tuning. Our findings demonstrate that fine-tuning that successively builds on prior subject-specific information improves both performance and stability, while OTTA effectively adapts the model to evolving data distributions across consecutive sessions, enabling calibration-free operation. These results offer valuable insights and recommendations for future research in longitudinal online MI decoding and highlight the importance of combining domain adaptation strategies for improving BCI performance in real-world applications. Clinical Relevance—Our investigation enables more stable and efficient long-term motor imagery decoding, which is critical for neurorehabilitation and assistive technologies. © 2025, CC BY.

关键词： Electroencephalography

HALS: A Height-Aware Lidar Super-Resolution Framework for Autonomous Driving

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Eskandar, George Sudarsan, Sanjeev Guirguis, Karim Palaniswamy, Janaranjani Somashekar, Bharath Yang, Bin University of Stuttgart Institute of Signal Processing and System Theory Stuttgart Germany Robert Bosch GmbH Renningen Germany

Lidar sensors are costly yet critical for understanding the 3D environment in autonomous driving. High-resolution sensors provide more details about the surroundings because they contain more vertical beams, but they come at a much higher cost, limiting their inclusion in autonomous vehicles. Upsampling lidar pointclouds is a promising approach to gain the benefits of high resolution while maintaining an affordable cost. Although there exist many pointcloud upsampling frameworks, a consistent comparison of these works against each other on the same dataset using unified metrics is still missing. In the first part of this paper, we propose to benchmark existing methods on the Kitti dataset. In the second part, we introduce a novel lidar upsampling model, HALS: Height-Aware Lidar Super-resolution. HALS exploits the observation that lidar scans exhibit a height-aware range distribution and adopts a generator architecture with multiple upsampling branches of different receptive fields. HALS regresses polar coordinates instead of spherical coordinates and uses a surface-normal loss. Extensive experiments show that HALS achieves state-of-the-art performance on 3 real-world lidar datasets. Copyright © 2022, The Authors. All rights reserved.

关键词： Optical radar