A5011272546- Ultrasound Imaging and Elastography
- Photoacoustic and Ultrasonic Imaging
- Non-Invasive Vital Sign Monitoring
- Medical Image Segmentation Techniques
- Seismic Imaging and Inversion Techniques
- Wireless Body Area Networks
- Ultrasonics and Acoustic Wave Propagation
- Medical Imaging and Analysis
- Advanced MRI Techniques and Applications
- Ultrasound and Hyperthermia Applications
- Antenna Design and Analysis
- Body Composition Measurement Techniques
- Brain Tumor Detection and Classification
- Pain Management and Placebo Effect
- Flow Measurement and Analysis
- Infrared Thermography in Medicine
- Energy Harvesting in Wireless Networks
- Myofascial pain diagnosis and treatment
- Stroke Rehabilitation and Recovery
- Image and Signal Denoising Methods
- Musculoskeletal pain and rehabilitation
- Ultrasound in Clinical Applications
Fraunhofer Institute for Digital Medicine
2020-2025
Mediri (Germany)
2022
Universität Hamburg
2019
Hamburg University of Technology
2019
Deep learning for ultrasound image formation is rapidly garnering research support and attention, quickly rising as the latest frontier in formation, with much promise to balance both quality display speed. Despite this promise, one challenge identifying optimal solutions absence of unified evaluation methods datasets that are not specific a single group. This article introduces largest known international database channel data describes associated were initially developed on beamforming...
Background/Objectives: Chronic low back pain (CLBP) is prevalent among older adults and leads to significant functional limitations reduced quality of life. Segmental stabilization exercises (SSEs) are commonly used treat CLBP, but the selective activation deep abdominal muscles during these can be challenging for patients. To support muscle activation, physiotherapists use biofeedback methods such as palpation ultrasound imaging. This randomized controlled pilot study aimed compare...
Reliable acoustic coupling in a non-handheld mode and reducing the form factor of electronics are specific challenges making ultrasound wearable. Applications relying on large field view (such as tracking muscles) induce need for element count to achieve high image quality. In our work, we developed 256-element linear array imaging abdominal muscles with four integrated custom-developed 8:32 multiplexer Integrated Circuits (ICs), allowing be driven by compact 32 ch electronics. The system is...
The emergence of data driven approaches such as Deep Learning has led to novel application various aspects science and engineering. It recently entered the field ultrasound image beamforming. In this work we investigate neural networks tailored create images quality multiple compounded plane wave excitations from central angle (0°) excitation only. proposed network is used produce pixel-wise weights weigh a standard delay-and-sum all channel available pixel. found higher than classical...
We investigate the feasibility of reconstructing ultrasound images directly from raw channel data using a deep learning network. Starting data, we present network full measurement information, allowing for more generic reconstruction to form, as compared common reconstructions constrained by physical models fixed speed sound assumptions.We propose U-Net-like architecture given task. Additional layers with strided convolutions downsample data. Hyperparameter optimization was used find...
For wireless capsule endoscopy, high quality images need to be transmitted from inside the digestive tract an on-body receiver. Ultra wideband transmission offers possibility achieve much larger data rates than achievable with today's technology. To design such ultra system a comprehensible channel model is needed for simulation of propagation behavior through human abdomen. In this paper we present stochastic model, that includes variation radio depending on location receive antenna body as...
Wearable ultrasound devices can be used to continuously monitor muscle activity. One possible application is provide real-time feedback during physiotherapy, show a patient whether an exercise performed correctly. Algorithms which automatically analyze the data of importance overcome need for manual assessment and annotations speed up evaluations especially when considering video sequences. They even could present in understandable manner patients home-use scenario. The following work...
Abstract Purpose Computed tomography (CT) is widely used to identify anomalies in brain tissues because their localization important for diagnosis and therapy planning. Due the insufficient soft tissue contrast of CT, division into anatomical meaningful regions challenging commonly done with magnetic resonance imaging (MRI). Methods We propose a multi-atlas registration approach propagate information from standard MRI atlas CT scans. This translation will enable detailed automated reporting...
Classical ultrasound reconstruction applies model driven approaches to obtain images from raw data. With the emergence of Deep Learning however data become feasible and can be explored. These used take shortcuts in reconstruction, directly learning relationship between image Even more, entirely new target contrasts pursued. In this work we present an approach train a neural network reconstruct classical novel MR-like contrast same
For real-time ultrasound reconstruction a global speed of sound (SoS) has to be assumed. A common strategy determine an estimate for this SoS is scan range candidate and choose the one yielding highest image quality with respect specific metric. However, usually does not consider secondary geometric effects changes in which affect grid as well transmitted angles plane wave imaging. This paper investigates how correct these what influence such corrections can have on optimization SoS. To end,...
The quality of ultrasound plane wave imaging benefits from compounding multiple angle acquisitions to reconstruct an image. However, the acquisition additional data lowers frame rate and – in presence motion integrity. This work presents approach high-quality images a reduced set angles making use artificial deep neural networks (DNNs). Unlike existing approaches that utilize DNNs for transforming beamformed into image directly, presented DNN is trained produce per-pixel angular weighting...