The seismocardiogram (SCG) is a recording of a human heart's mechanical activity. It captures fine-grained cardiovascular events such as the opening and closing of heart valves and the contraction and relaxation of heart chambers. Today, SCG recordings are obtained by strapping an accelerometer at the apex of the heart to measure chest wall vibrations. These recordings can be used to diagnose and monitor various cardiovascular conditions including myocardial infarction (heart attack), coronary heart disease, and ischemia. This paper introduces RF-SCG, a system that can capture SCG recordings without requiring any contact with the human body. The system operates by analyzing the reflections of millimeter-wave radar signals off the human body. RF-SCG can reconstruct the SCG waveform, and it can time 5 cardiovascular events within individual heartbeats with high accuracy. Our design is based on a hybrid architecture that combines signal processing with deep learning. The pipeline includes a 4D Cardiac Beamformer that can focus on the reflections of the human heart and a deep learning pipeline (RFto-SCG Translator) that can transform these reflections into SCG waveforms. Empirical evaluation with 40,000 heartbeats from 21 healthy subjects demonstrates RF-SCG's ability to robustly time five key cardiovascular events (aortic valve opening, aortic valve closing, mitral valve opening, mitral valve closing, and isovolumetric contraction) with a median error between 0.26%-1.29%. CCS CONCEPTS • Applied computing → Life and medical sciences; • Humancentered computing → Ubiquitous and mobile computing systems and tools;

Purpose To develop and evaluate MyoMapNet, a rapid myocardial T1 mapping approach that uses fully connected neural networks (FCNN) to estimate T1 values from four T1-weighted images collected after a single inversion pulse in four heartbeats (Look-Locker, LL4). Method We implemented an FCNN for MyoMapNet to estimate T1 values from a reduced number of T1-weighted images and corresponding inversion-recovery times. We studied MyoMapNet performance when trained using native, post-contrast T1, or a combination of both. We also explored the effects of number of T1-weighted images (four and five) for native T1. After rigorous training using in-vivo modified Look-Locker inversion recovery (MOLLI) T1 mapping data of 607 patients, MyoMapNet performance was evaluated using MOLLI T1 data from 61 patients by discarding the additional T1-weighted images. Subsequently, we implemented a prototype MyoMapNet and LL4 on a 3 T scanner. LL4 was used to collect T1 mapping data in 27 subjects with inline T1 map reconstruction by MyoMapNet. The resulting T1 values were compared to MOLLI. Results MyoMapNet trained using a combination of native and post-contrast T1-weighted images had excellent native and post-contrast T1 accuracy compared to MOLLI. The FCNN model using four T1-weighted images yields similar performance compared to five T1-weighted images, suggesting that four T1 weighted images may be sufficient. The inline implementation of LL4 and MyoMapNet enables successful acquisition and reconstruction of T1 maps on the scanner. Native and post-contrast myocardium T1 by MOLLI and MyoMapNet was 1170 ± 55 ms vs. 1183 ± 57 ms (P = 0.03), and 645 ± 26 ms vs. 630 ± 30 ms (P = 0.60), and native and post-contrast blood T1 was 1820 ± 29 ms vs. 1854 ± 34 ms (P = 0.14), and 508 ± 9 ms vs. 514 ± 15 ms (P = 0.02), respectively. Conclusion A FCNN, trained using MOLLI data, can estimate T1 values from only four T1-weighted images. MyoMapNet enables myocardial T1 mapping in four heartbeats with similar accuracy as MOLLI with inline map reconstruction.

Today, cameras and digital image processing are transforming industries and the human environment with rich, informative sensing. However, image processing is not utilized nearly as much in homes where concerns about image privacy dominate. In a preliminary study with 200 participants, we found 21% would reject a camera based system even if the system was designed to not report images as they could still be collected if the camera system was hacked. In this paper, we demonstrate a hardware-based approach for privacy-preserving image processing: the ability to automatically extract information from imaging sensors without the risk of compromising image privacy, even if the system is hacked. The basic idea is to limit both the memory available on board the camera and the data rate of camera communication to prevent a full image from ever being extracted. As a proof of concept, we prototype a system, called Lethe, that tracks and identifies individuals by height with a thermal camera as they move from room to room. Our results show that Lethe can detect the presence of individuals with 96.9% accuracy and determine their direction of travel with 99.7% accuracy. Additionally, Lethe can identify individuals 96.0% of the time with a 5cm (~2in) or greater difference in walking height and 92.9% with a 2.5cm (~1in) or greater difference. Finally, Lethe performs this processing with only 33 bytes of memory (or 0.69% of the full thermal image).

The objective of the current study was to investigate the performance of various deep learning (DL) architectures for MyoMapNet, a DL model for T 1 estimation using accelerated cardiac T 1 mapping from four T 1 -weighted images collected after a single inversion pulse (Look-Locker 4 [LL4]). We implemented and tested three DL architectures for MyoMapNet: (a) a fully connected neural network (FC), (b) convolutional neural networks (VGG19, ResNet50), and (c) encoder-decoder networks with skip connections (ResUNet, U-Net). Modified Look-Locker inversion recovery (MOLLI) images from 749 patients at 3 T were used for training, validation, and testing. The first four T 1 -weighted images from MOLLI5(3)3 and/or MOLLI4(1)3(1)2 protocols were extracted to create accelerated cardiac T 1 mapping data. We also prospectively