Objective: Wearable ultrasound is emerging as a new paradigm of real-time imaging in freely moving humans and has wide applications from cardiovascular health monitoring to human gesture recognition. However, current wearable ultrasound devices have typically employed pulse-echo imaging which requires high excitation voltages and sampling rates, posing safety risks, and requiring specialized hardware. Our objective was to develop and evaluate a wearable ultrasound system based on time delay spectrometry (TDS) that utilizes low-voltage excitation and significantly simplified instrumentation. Methods: We developed a TDS-based ultrasound system that utilizes continuous, frequency-modulated sweeps at low excitation voltages. By mixing the transmit and receive signals, the system digitizes the ultrasound signal at audio frequency (kHz) sampling rates. Wearable ultrasound transducers were developed, and the system was characterized in terms of imaging performance, acoustic output, thermal characteristics, and applications in musculoskeletal imaging. Results: The prototype TDS system is capable of imaging up to 6 cm of depth with signal-to-noise ratio of up to 42 dB at a spatial resolution of 0.33 mm. Acoustic and thermal radiation measurements were within clinically safe limits for continuous ultrasound imaging. We demonstrated the ability to use a 4-channel wearable system for dynamic imaging of muscle activity. Conclusion: We developed a wearable ultrasound imaging system using TDS to mitigate challenges with pulse echo-based wearable ultrasound imaging systems. Our device is capable of high-resolution, dynamic imaging of deep-seated tissue structures and is safe for long-term use. Significance: This work paves the way for low-voltage wearable ultrasound imaging devices with significantly reduced hardware complexity.

Rehabilitation from musculoskeletal injuries focuses on reestablishing and monitoring muscle activation patterns to accurately produce force. The aim of this study is to explore the use of a novel low-powered wearable distributed Simultaneous Musculoskeletal Assessment with Real-Time Ultrasound (SMART-US) device to predict force during an isometric squat task. Participants (N = 5) performed maximum isometric squats under two medical imaging techniques; clinical musculoskeletal motion mode (m-mode) ultrasound on the dominant vastus lateralis and SMART-US sensors placed on the rectus femoris, vastus lateralis, medial hamstring, and vastus medialis. Ultrasound features were extracted, and a linear ridge regression model was used to predict ground reaction force. The performance of ultrasound features to predict measured force was tested using either the Clinical M-mode, SMART-US sensors on the vastus lateralis (SMART-US: VL), rectus femoris (SMART-US: RF), medial hamstring (SMART-US: MH), and vastus medialis (SMART-US: VMO) or utilized all four SMART-US sensors (Distributed SMART-US). Model training showed that the Clinical M-mode and the Distributed SMART-US model were both significantly different from the SMART-US: VL, SMART-US: MH, SMART-US: RF, and SMART-US: VMO models (p < 0.05). Model validation showed that the Distributed SMART-US model had an R² of 0.80 ± 0.04 and was significantly different from SMART-US: VL but not from the Clinical M-mode model. In conclusion, a novel wearable distributed SMART-US system can predict ground reaction force using machine learning, demonstrating the feasibility of wearable ultrasound imaging for ground reaction force estimation.

Rehabilitation from musculoskeletal injuries focuses on reestablishing and monitoring muscle activation patterns to accurately produce force. The aim of this study is to explore the use of a novel low-powered wearable distributed Simultaneous Musculoskeletal Assessment with Real-Time Ultrasound (SMART-US) device to predict force during an isometric squat task. Participants (N=5) performed maximum isometric squats under two medical imaging techniques; clinical musculoskeletal motion mode (m-mode) ultrasound on the dominant vastus lateralis and SMART-US sensors placed on the rectus femoris, vastus lateralis, medial hamstring, and vastus medialis. Ultrasound features were extracted, and a linear ridge regression model was used to predict ground reaction force. The performance of ultrasound features to predict measured force was tested using either the Clinical M-mode, SMART-US sensors on the vastus lateralis (SMART-US: VL), rectus femoris (SMART-US: RF), medial hamstring (SMART-US: MH), and vastus medialis (SMART-US: VMO) or utilized all four SMART-US sensors (Distributed SMART-US). Model training showed that the Clinical M-mode and the Distributed SMART-US model were both significantly different from the SMART-US: VL, SMART-US: MH, SMART-US:RF, and SMART-US:VMO models (p<0.05). Model validation showed that the Distributed SMART-US model had an R² of 0.80 ±0.04 and was significantly different from SMART-US: VL but not from the Clinical M-mode model. In conclusion, a novel wearable distributed SMART-US system can predict ground reaction force using machine learning, demonstrating the feasibility of wearable ultrasound imaging for ground reaction force estimation.

Frequency Modulated Continuous Wave (FMCW) ultrasound is an emerging imaging method adapted from radar that can be implemented in wearable ultrasound systems due to the reduced size, power, and signal processing requirements. Acoustic reflections are encoded in the frequency domain, and thus evaluating signal quality currently requires extracting frequency information from a time-varying acquisition signal, which requires additional signal processing. Question: Can we differentiate FMCW ultrasound signal quality using an acquired FMCW time-series? We collected FMCW ultrasound images with a custom ultrasound transducer contacting and losing contact with tissue, to simulate moments of poor signal quality. We used two potential time-domain signal quality metrics: the mean absolute value (MAV) of the signal, and Sample Entropy. Sample Entropy is an information theory method for assessing complexity in a time-series. Both MAV and Sample Entropy could differentiate when the transducer was in good contact with the tissue. MAV required less processing time, but Sample Entropy had less noise when the transducer lost contact. FMCW ultrasound imaging is a promising imaging method that has application in wearable ultrasound systems, and FMCW ultrasound signal quality can be assessed in the time-domain without extracting frequency information. [Work supported by NIH 5U01EB027601-02.]

Doppler ultrasound imaging has become a key tool for clinicians and researchers to measure the velocity of blood and tissues within the human body. Current clinical instruments utilize pulsed wave or continuous wave transmissions for tracking phase shifts or frequency shifts respectively to estimate scatterer motion. However, these two techniques can also be considered as ends of a continuum along which other motion tracking techniques exist. In this paper, we explore frequency modulated semi-continuous wave (FMSCW) Doppler, an underexplored Doppler imaging technique that has particular benefits for emerging wearable ultrasound applications due to significant simplification of instrumentation and lowered data rates. We present a full technical overview of this technique, simulations exploring the impact of selected imaging parameters as well as how noise impacts imaging, and present three exemplary imaging applications in human participants using FMSCW Doppler implemented on a wearable ultrasound system. The results of the simulations evaluating the impact of imaging parameters demonstrate the trade-off between resolution in the range and velocity axis as well as the impact of scatterer range and speed on the measurement error. The three sets of images collected from human participants serve to demonstrate FMSCW Doppler used in a variety of wearable ultrasound applications including cardiovascular monitoring, rehabilitation assessment, and prosthetic device control. Furthermore, we discuss the relationship between FMSCW Doppler and isomorphic techniques from radar as well as potential inspiration for improvements for FMSCW Doppler inspired by the echolocation of bats.