Recently, miniature luminescence-based blood gas monitors have emerged as noninvasive, compact devices for continuous monitoring. Although these sensors demonstrate enhanced stability compared to traditional electrochemical-based sensors, they still face challenges such as photobleaching, temperature fluctuations, and material degradation, albeit at a reduced rate. This study introduces an algorithm aimed at compensating for drift in luminescence-based sensing. We implemented the algorithm on a custom evaluation board, incorporating an STM32WB35CC microcontroller and an ADPD4101 analog front end for luminescence lifetime sampling. The setup utilized a commercial platinum porphyrin-based luminescence sensor film. Experiments measured the partial pressure of oxygen in ambient air over three-day periods, comparing uncompensated measurements with those adjusted in real-time using the proposed compensation algorithm. The results indicated a ±0.3% change in luminescence lifetime over three days with the compensation method, in contrast to a drift of -2.6% to -3.6% without compensation. The proposed method enhances the long-term reliability and accuracy of luminescence-based oxygen sensing, with potential applications in extended home care and clinical monitoring.

Acute mountain sickness (AMS) is a potentially life-threatening condition that affects many individuals traveling to high altitudes. Early diagnosis is crucial, especially for travelers who may not have immediate access to medical resources. While traditional machine learning (ML) methods have been used to detect AMS using biomedical data (e.g., heart rate, blood oxygen saturation, respiration rate, blood pressure, and body temperature), hyperdimensional computing (HDC) has yet to be explored for this purpose using the few of biomedical data. Previous classification methods fall short of balancing accuracy with low hardware complexity, but HDC offers a promising solution. HDC provides a hardware-efficient alternative solution, making it well-suited for resource-constrained environments, such as wearable devices. Its lightweight architecture and efficient memory management make it ideal for embedded systems, enabling real-time AMS detection with accuracy comparable to traditional ML models. We introduce AMS-HD, a novel framework that leverages custom feature engineering and quasi-random hyper-vector encoding to further enhance the efficiency and accuracy of HDC for AMS detection. The proposed framework demonstrates the potential for seamless integration into wearable biomedical devices for on-the-go health monitoring.

Transcutaneous oxygen sensing is a noninvasive method for continuous monitoring of partial pressure of oxygen diffused through the skin that closely correlates with changes in arterial blood gases. A method for measuring transcutaneous oxygen, suitable for a wearable device, is luminescence oxygen sensing. This study aims to determine the optimum values of the system parameters for power, accuracy, and precision for the development of transcutaneous oxygen wearable based on a platinum-porphyrin film and the ADPD4101 analog front-end. For this purpose, we conducted several experiments using the CN0503 optical measurement system from Analog Devices, Inc. Based on the optimization experiments, we determined a 100 mA and 100 $\mu$s LED pulse intensity and width, and a 2 $\mu$s sampling period gave a good compromise between power and signal integrity. Using these settings, we observed a discernible $\tau$ response to changes in oxygen pressure from 0 to 418 mmHg.

The landscape of miniaturized wireless biomedical devices (MWBDs), including various injectables, ingestibles, implantables, and wearables, is rapidly expanding as proactive mobile healthcare proliferates. While the growth of MWBDs increases the flexibility of medical services, the adoption of these technologies poses privacy and security risks to their users. As a result, while being restricted in resources (size, power, processing, and storage), these devices require trust and must be at least minimally secure in the face of evolving threats. Making MWBDs secure begins with threat modeling. Therefore, this research reviews and summarizes the information on threat modeling applicable to MWBDs. Then, we propose a domain-specific qualitative-quantitative threat model that aims to help the designers and manufacturers of MWBDs to identify threats and embed security in their designs in the premarket phase of the lifecycle of an MWBD. This model is tailored to a wide range of MWBDs. Among the different stakeholders, this model focuses on the user. It also prioritizes noninvasive direct attacks against telemetry interfaces. To discuss the advantages and disadvantages of the proposed model, it is compared to some other threat models. To illustrate how the model can be adopted by a threat-modeling team, it is then applied to representative case studies from each category of MWBDs. The outcomes of the performed risk analysis reveal that the model is easy to apply and sufficient to disclose threats.

General graphs are difficult for learning due to their irregular structures. Existing works employ message passing along graph edges to extract local patterns using customized graph kernels, but few of them are effective for the integration of such local patterns into global features. In contrast, in this paper we study the methods to transfer the graphs into trees so that explicit orders are learned to direct the feature integration from local to global. To this end, we apply the breadth first search (BFS) to construct trees from the graphs, which adds direction to the graph edges from the center node to the peripheral nodes. In addition, we proposed a novel projection scheme that transfer the trees to image representations, which is suitable for conventional convolution neural networks (CNNs) and recurrent neural networks (RNNs). To best learn the patterns from the graph-tree-images, we propose TreeRNN, a 2D RNN architecture that recurrently integrates the image pixels by rows and columns to help classify the graph categories. We evaluate the proposed method on several graph classification datasets, and manage to demonstrate comparable accuracy with the state-of-the-art on MUTAG, PTC-MR and NCI1 datasets.

In most optical-based oxygen sensors, fluorescent quenching is used as the sensing principle. Intensity and lifetime are two known measurement techniques that are employed to quantify the measured parameter. The lifetime measurement technique has superior features such as little or no sensitivity to changes in the optical path, degradation of the film or photo-bleaching. In this study, we conducted an experiment to analyze the performance of both the intensity and lifetime measurement techniques of a platinum-porphyrin fluorescent film, the sensing material, under different concentrations of oxygen. A prototype that deploys an analog front-end, a light emitting diode driver, and a power management block is implemented on a printed circuit board with commercial off-the-shelf components. The system resolves changes in oxygen pressure from 0.5 mmHg to 500 mmHg with a power consumption of 132 mW. This system could be potentially used to measure the concentration of oxygen diffused through the skin, also known as transcutaneous measurement of oxygen. These measurements are key to developing more sophisticated read-out circuits for reliable and sensitive medical purpose transcutaneous oxygen sensing.

Battery-less wireless systems generate their supply voltage from the incoming RF signal, and need to deal with the problem of generating sufficient voltage level required by their various blocks from weak input signals, while maintaining high efficiency. The presented voltage multiplier seeks to maximize the power conversion efficiency (PCE) by changing the RF-to-DC converter topology in the voltage multiplier based on the incoming signal. Besides topology, the number of stages in the voltage multiplier can change for the desired output DC voltage from the varying RF input signal. A prototype is fabricated in 0.35-μm standard CMOS process, operating at 13.56 MHz with a 2 kΩ load. The core element in the voltage multiplier yields a peak PCE of 76% with the cross-coupled topology at low input power, while the hybrid topology provides 70% PCE at higher input power. This reconfigurable voltage multiplier can be used in RFID or Internet-of-Things (IoT) mobile applications to regulate power against changes in the distance between the power source and the target.