In this work we present the evaluation results of our 3D sonar camera system. The system consists of a matrix antenna array with 1024 single transducer elements and our in house developed DiPhAS sonar beamformer - a 128 channel FPGA-based beamforming system with a 1:8 multiplexing device for each channel. The system is designed to be applicable to ROV and AUV systems for real-time volumetric imaging in a deep sea environment. Defocused excitation of the transducer array is used to achieve a sound field opening angle of up to 40° in lateral and elevational direction. The antenna's sound field can be adjusted electronically in order to increase either the imaged area or the image contrast in a specific area of interest. Different filter algorithms working on a raw data basis have been implemented in order to suppress image artifacts which occur during the reconstruction process. Measurements on different phantoms have been performed in order to prove the real-time imaging as well as spatial resolution capabilities of the camera system

We present the newest application specific version of our beamformer platform “DiPhAS” that provides 256 parallel channels both for generation of ultrasound signals as well as digitalization of returned echos. Using ultrasound transducers with lots of elements requires high channel count electronics. Applications for such systems range from functional ultrafast imaging using high element count linear array transducers for imaging of a large field of view to real time volumetric imaging with matrix array transducers. To perform volumetric transmit beamforming with matrix transducers, lots of these matrix elements have to be controlled individually. Furthermore, many elements need to be excited in order to compensate for the small active element size and provide a sufficient overall active footprint to generate enough acoustic power for imaging with adequate SNR. The system is set up based on our platform concept with the common ultrasound research device components: mainboard, power supply, application-specific new front ends integrating 16 channels on each PCB and device software. Using 16 front ends results in a total channel count of 256. The new front ends are based on our existing 8 channel front end technology and share the same concepts with doubled channel count for both transmission and reception. The system generates transmit sequences with voltages up to 150 Vpp and digitizes with a sampling rate of up to 80 MHz. The beamformer implements the control for additional external multiplexers in the transducer probe. This has been tested with an external transducer matrix array and can be used to connect to our custom 1024 elements matrix array (32×32 elements) with a 1:4 multiplexer integrated into the probe head. Received data can be accessed as single element channel data of all 256 channels in parallel and transferred to a PC via PCI-Express. Beamforming can be done on a massively parallel computing graphics processor (GPU). The used software includes standard applications for measurements and interfaces for Matlab, C++ and C#. It is used to process, analyze and visualize data from the beamformer. This system will be scalable to an even higher channel count by connecting several beamformers to a single PC using multiple PCI-Express connections and additional synchronization over all single beamformer electronics. It is the basis of our 3D/4D ultrasound research system connected to our matrix arrays developed in-house

We developed a new mobile ultrasound device for long-term and automated bladder monitoring without user interaction consisting of 32 transmit and receive electronics as well as a 32-element phased array 3 MHz transducer. The device architecture is based on data digitization and rapid transfer to a consumer electronics device (e.g., a tablet) for signal reconstruction (e.g., by means of plane wave compounding algorithms) and further image processing. All reconstruction algorithms are implemented in the GPU, allowing real-time reconstruction and imaging. The system and the beamforming algorithms were evaluated with respect to the imaging performance on standard sonographical phantoms (CIRS multipurpose ultrasound phantom) by analyzing the resolution, the SNR and the CNR. Furthermore, ML-based segmentation algorithms were developed and assessed with respect to their ability to reliably segment human bladders with different filling levels. A corresponding CNN was trained with 253 B-mode data sets and 20 B-mode images were evaluated. The quantitative and qualitative results of the bladder segmentation are presented and compared to the ground truth obtained by manual segmentation.

Mobile and cost effective ultrasound devices are used in point of care scenarios or the drama room. To reduce the costs of such devices we already presented the possibilities of consumer devices like the Apple iPad for full signal processing of raw data for ultrasound image generation. Emerging technologies like ultrafast ultrasound imaging result in new algorithms for example for shearwave elastography or vector velocity imaging but also enable the creation of a full image with only one excitation/reception event based on plane wave imaging. This way acquisition times and power consumption of ultrasound imaging can be reduced for low power mobile devices based on consumer electronics realizing the transition from FPGA or ASIC based beamforming into more flexible software beamforming. This is usually performed on a GPU utilizing massive parallel processing (like CUDA or OpenCL) but with the development of modern processors (A7, A8 and A8X) for its smartphones and tablet s Apple introduced parallel GPU hardware and the framework 'Metal' for advanced graphics and general purpose GPU processing for the iOS platform. We use it for medical signal reconstruction in the mobile plane wave beamforming and imaging on ultrasound channel data sets measured with our research systems 'DiPhAS' in ultrafast imaging mode. We were able to integrate the beamforming reconstruction into our mobile ultrasound processing application on the iOS platform. The next step into realizing a mobile, fully software based ultrasound system was made. The beamforming can be performed at up to 62 Hz at reasonable image quality on iPad Air 2 hardware providing real time imaging including the post-processing of beamformed data into images (envelope detection and scan conversion)

No abstract