{"id":26335,"date":"2025-10-21T14:33:04","date_gmt":"2025-10-21T12:33:04","guid":{"rendered":"https:\/\/sano.science\/?post_type=seminars&p=26335"},"modified":"2026-02-05T16:47:43","modified_gmt":"2026-02-05T15:47:43","slug":"software-defined-ultrasound-innovations","status":"publish","type":"seminars","link":"https:\/\/sano.science\/seminars\/software-defined-ultrasound-innovations\/","title":{"rendered":"169. Software-Defined Ultrasound Innovations"},"content":{"rendered":"\n

Abstract<\/h2>\n\n\n\n
<\/div>\n\n\n\n

Software-defined ultrasound (SDU) is a design paradigm where raw ultrasound signals are acquired and processed with software, instead of dedicated hardware. This approach facilitates the rapid prototyping of new ultrasound modalities and enables the introduction of advanced signal processing algorithms that would be difficult and laborious to implement in hardware.<\/p>\n\n\n\n

We present our fully programmable SDU solution composed of a versatile ultrasound front-end hardware and software framework. An open-source ARRUS API provides a general programming and data model, enabling users to define and execute ultrasound TX\/RX sequences, acquire raw channel data, and execute data transformation graphs on selected processing devices. The API communicates with the low-level session manager, responsible for device resource management and communication with GPU-accelerated processing devices and the us4us ultrasound platforms.<\/p>\n\n\n\n

ARRUS incorporates a data processing framework based on a directed acyclic graph (DAG) of operations, where nodes represent algorithms executed at specific processing stages, and edges define input\/output data dependencies. The framework supports configurable operation-to-device assignments, allowing users to select a processing placement by adjusting a single parameter within a subgraph. Under the hood, the processing graph is translated into device-specific streams of processing kernels.<\/p>\n\n\n\n

I will present a few practical use-cases of advanced ultrasound modalities implemented on our SDU platform (including AI\/ML methods).<\/p>\n\n\n\n

The platform has proven its usefulness in research as well as for implementation of real medical devices.<\/p>\n\n\n\n

We are open to R&D academic collaboration and mutual projects.<\/p>\n\n\n\n

<\/div>\n\n\n\n

About the author:<\/h2>\n\n\n\n
<\/div>\n\n\n\n

Dr Marcin Lewandowski marcin@us4us.eu<\/a>, CEO\/co-founder of us4us Ltd.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Since receiving a master\u2019s degree in physics followed by a PhD in Electronic Engineering (2009), Marcin Lewandowski has headed many R&D projects, product designs, and medical devices developments. He has authored numerous publications in scientific journals on the medical and industrial applications of ultrasound. Over the course of 30+ years working in ultrasound, electronics and software development, Marcin has strived to apply his research expertise in projects with a strong potential for innovation and commercialization. Today, he balances work in research with his role as CEO at us4us Ltd., who produces original ultrasound solutions for research, biomedical and industrial applications.<\/p>\n\n\n\n

<\/div>\n\n\n\n
\n