The COVID-19 pandemic has highlighted the need for a tool to speed up triage in ultrasound scans and provide clinicians with fast access to relevant information. To this end, we propose a new unsupervised reinforcement learning (RL) framework with novel rewards to facilitate unsupervised learning by avoiding tedious and impractical manual labelling for summarizing ultrasound videos. The proposed framework is capable of delivering video summaries with classification labels and segmentations of key landmarks which enhances its utility as a triage tool in the emergency department (ED) and for use in telemedicine. Using an attention ensemble of encoders, the high dimensional image is projected into a low dimensional latent space in terms of: a) reduced distance with a normal or abnormal class (classifier encoder), b) following a topology of landmarks (segmentation encoder), and c) the distance or topology agnostic latent representation (autoencoders). The summarization network is implemented using a bi-directional long short term memory (Bi-LSTM) which utilizes the latent space representation from the encoder. Validation is performed on lung ultrasound (LUS), that typically represent potential use cases in telemedicine and ED triage acquired from different medical centers across geographies (India and Spain). The proposed approach trained and tested on 126 LUS videos showed high agreement with the ground truth with an average precision of over 80% and average $\boldsymbol{F_{1}}$ score of well over $ \boldsymbol{44 \pm 1.7 \%}$ . The approach resulted in an average reduction in storage space of 77% which can ease bandwidth and storage requirements in telemedicine.
Analogous to how aerial imagery of above-ground environments transformed our understanding of the earth’s landscapes, remote underwater imaging systems could provide us with a dramatically expanded view of the ocean. However, maintaining high-fidelity imaging in the presence of ocean surface waves is a fundamental bottleneck in the real-world deployment of these airborne underwater imaging systems. In this work, we introduce a sensor fusion framework which couples multi-physics airborne sonar imaging with a water surface imager. Accurately mapping the water surface allows us to provide complementary multi-modal inputs to a custom image reconstruction algorithm, which counteracts the otherwise detrimental effects of a hydrodynamic water surface. Using this methodology, we experimentally demonstrate three-dimensional imaging of an underwater target in hydrodynamic conditions through a lab-based proof-of-concept, which marks an important milestone in the development of robust, remote underwater sensing systems.
7 November 2022 in IEEE Microwave and Wireless Technology Letters
This letter presents a double-tuned dual input transmitter coil operating at 13.56 and 40.68 MHz industrial, scientific, and medical (ISM) bands for multisite biomedical applications. The proposed system removes the need for two separate coils, which reduces system size and unwanted couplings. The design and analysis of the double-tuned transmitter coil using a lumped element frequency trap are discussed in this letter. The transmitter achieves measured matching of $-$ 26.2 and $-$ 21.5 dB and isolation of $-$ 17.7 and $-$ 11.7 dB at 13.56 and 40.68 MHz, respectively. A 3 $\times$ 15 mm flexible coil is used as an implantable receiver. This letter shows synchronized multisite stimulation of two flexible implants at a distance of 2 cm while covered with 1 cm chicken breast.
This article presents a low-power, battery-less biopotential-recording integrated circuit (IC) designed for implantable leadless cardiac monitoring applications, addressing full autonomy based on wireless energy harvesting and closed-loop control. A key technical feature is the closed-loop capability with integration of a dual-band radio frequency (RF) energy harvester and communication system utilizing industrial, scientific, and medical (ISM) bands at 40.68 MHz for downlink and 915 MHz for uplink, ensuring efficient wireless power transfer and robust bidirectional data communication. The RF harvester achieves a peak conversion efficiency exceeding 40%, a critical design challenge in maintaining consistent performance under variable power conditions. Also, the proposed IC overcomes the need for a local oscillator (LO) by utilizing an externally provided clock, reducing system complexity and power consumption. Operating with a clock frequency of 100 kHz, the IC consumes only $8.8~\mu $ W, a major achievement in energy efficiency for wireless, battery-less systems. Fabricated in a 180-nm complementary metaloxidesemiconductor (CMOS) process, the IC successfully demonstrates cardiac signal measurement and heart rate (HR) extraction under a fully battery-less environment with air medium, providing a promising solution for next-generation leadless cardiac monitoring systems.
This paper presents an innovative dual-mode system as well as design techniques to realize full-duplex data communication and wireless power transfer over a single inductive link. An LC tank implemented on a compact printed circuit board is shared between a rectifier that supplies the batteryless device and a power oscillator that modulates the uplink data in a time division duplex fashion. Designed to operate efficiently in the loose coupling regime, the proposed design ensures reliable performance over a substantial distance and is resilient to interference from the wireless power transfer signal. A miniaturized prototype is implemented on a printed circuit board package with a diameter of 19 mm, which is powered by a 180-nm CMOS chipset. At a data rate of 10 kbps, the coin-sized prototype achieves powering and uplink distances of 140 mm and 110 mm, respectively, offering a fivefold improvement in normalized operating range compared to recent single-coil designs of similar size. The performance of the proposed sensing device is verified by inserting a thick biological tissue between the coils and recording ex-vivo biopotential signals.