Search for: All records

Precise monitoring of respiratory rate in premature newborn infants is essential to initiating medical interventions as required. Wired technologies can be invasive and obtrusive to the patients. We propose a deep-learning-enabled wearable monitoring system for premature newborn infants, where respiratory cessation is predicted using signals that are collected wirelessly from a non-invasive wearable Bellypatch put on the infant’s body. We propose a five-stage design pipeline involving data collection and labeling, feature scaling, deep learning model selection with hyperparameter tuning, model training and validation, and model testing and deployment. The model used is a 1-D convolutional neural network (1DCNN) architecture with one convolution layer, one pooling layer, and three fully-connected layers, achieving 97.15% classification accuracy. To address the energy limitations of wearable processing, several quantization techniques are explored, and their performance and energy consumption are analyzed for the respiratory classification task. Results demonstrate a reduction of energy footprints and model storage overhead with a considerable degradation of the classification accuracy, meaning that quantization and other model compression techniques are not the best solution for respiratory classification problem on wearable devices. To improve accuracy while reducing the energy consumption, we propose a novel spiking neural network (SNN)-based respiratory classification solution, which can be implemented on event-driven neuromorphic hardware platforms. To this end, we propose an approach to convert the analog operations of our baseline trained 1DCNN to their spiking equivalent. We perform a design-space exploration using the parameters of the converted SNN to generate inference solutions having different accuracy and energy footprints. We select a solution that achieves an accuracy of 93.33% with 18x lower energy compared to the baseline 1DCNN model. Additionally, the proposed SNN solution achieves similar accuracy as the quantized model with a 4× lower energy. 

Currently, wired respiratory rate sensors tether patients to a location and can potentially obscure their body from medical staff. In addition, current wired respiratory rate sensors are either inaccurate or invasive. Spurred by these deficiencies, we have developed the Bellyband, a less invasive smart garment sensor, which uses wireless, passive Radio Frequency Identification (RFID) to detect bio-signals. Though the Bellyband solves many physical problems, it creates a signal processing challenge, due to its noisy, quantized signal. Here, we present an algorithm by which to estimate respiratory rate from the Bellyband. The algorithm uses an adaptively parameterized Savitzky-Golay (SG) filter to smooth the signal. The adaptive parameterization enables the algorithm to be effective on a wide range of respiratory frequencies, even when the frequencies change sharply. Further, the algorithm is three times faster and three times more accurate than the current Bellyband respiratory rate detection algorithm and is able to run in real time. Using an off-the-shelf respiratory monitor and metronome-synchronized breathing, we gathered 25 sets of data and tested the algorithm against these trials. The algorithm’s respiratory rate estimates diverged from ground truth by an average Root Mean Square Error (RMSE) of 4.1 breaths per minute (BPM) over all 25 trials. Further, preliminary results suggest that the algorithm could be made as or more accurate than widely used algorithms that detect the respiratory rate of non-ventilated patients using data from an Electrocardiogram (ECG) or Impedance Plethysmography (IP). 

Future advances in the medical Internet of Things (IoT) will require sensors that are unobtrusive and passively powered. With the use of wireless, wearable, and passive knitted smart garment sensors, we monitor infant respiratory activity. We improve the utility of multi-tag Radio Frequency Identification (RFID) measurements via fusion learning across various features from multiple tags to determine the magnitude and temporal information of the artifacts. In this paper, we develop an algorithm that classifies and separates respiratory activity via a Regime Hidden Markov Model compounded with higher-order features of Minkowski and Mahalanobis distances. Our algorithm improves respiratory rate detection by increasing the Signal to Noise Ratio (SNR) on average from 17.12 dB to 34.74 dB. The effectiveness of our algorithm in increasing SNR shows that higher-order features can improve signal strength detection in RFID systems. Our algorithm can be extended to include more feature sources and can be used in a variety of machine learning algorithms for respiratory data classification, and other applications. Further work on the algorithm will include accurate parameterization of the algorithm's window size.