Background Blood glucose (BG) management is crucial for type-1 diabetes patients resulting in the necessity of reliable artificial pancreas or insulin infusion systems. In recent years, deep learning techniques have been utilized for a more accurate BG level prediction system. However, continuous glucose monitoring (CGM) readings are susceptible to sensor errors. As a result, inaccurate CGM readings would affect BG prediction and make it unreliable, even if the most optimal machine learning model is used. Methods In this work, we propose a novel approach to predicting blood glucose level with a stacked Long short-term memory (LSTM) based deep recurrent neural network (RNN) model considering sensor fault. We use the Kalman smoothing technique for the correction of the inaccurate CGM readings due to sensor error. Results For the OhioT1DM (2018) dataset, containing eight weeks’ data from six different patients, we achieve an average RMSE of 6.45 and 17.24 mg/dl for 30 min and 60 min of prediction horizon (PH), respectively. Conclusions To the best of our knowledge, this is the leading average prediction accuracy for the ohioT1DM dataset. Different physiological information, e.g., Kalman smoothed CGM data, carbohydrates from the meal, bolus insulin, and cumulative step counts in a fixed time interval, are crafted to represent meaningful features used as input to the model. The goal of our approach is to lower the difference between the predicted CGM values and the fingerstick blood glucose readings—the ground truth. Our results indicate that the proposed approach is feasible for more reliable BG forecasting that might improve the performance of the artificial pancreas and insulin infusion system for T1D diabetes management.

No abstract

No abstract

Without effective cryptographic mechanisms, the wireless channel between the USB/uploader and insulin pump frequently suffers from vulnerabilities. Either eavesdropping or therapy manipulation attacks would put the patients in a life-threatening situation. Towards tackling this problem, we propose an access control scheme by introducing feature fusion and voiceprint. Featured by the anti-replay speaker verification and voiceprint-based key agreement, it secures communications over the wireless channel. Through a cascaded fusion of speaker verification and anti-replay countermeasure, the anti-replay speaker verification guarantees that the pump can only be accessed after the verification. When defending against zero-effort and replay impostors with our scheme, the equal error rate can be reduced to 2.22%. Furthermore, to generate a common key for the wireless channel, in the voiceprint-based key agreement, we present a noninteractive energy-difference-based voiceprint extraction and adaptive Reed-Solomon coding based fuzzy extractor. Thus, it enhances the communication encryption which protects the pump from eavesdropping and therapy manipulation attacks. Also, with an appropriate constraint on voiceprints similarity, the key agreement lowers the risk of channel establishment from device locating outside the pump's close proximity. INDEX TERMS Wireless insulin pump, feature fusion, voiceprint, access control, acoustic channel.

Physical phenomenon in nature is generally simulated by partial differential equations. Among different sorts of partial differential equations, the problem of two-phase flow in porous media has been paid intense attention. As a promising direction, physics-informed neural networks shed new light on the solution of partial differential equations. However, current physics-informed neural networks' ability to learn partial differential equations relies on adding artificial diffusion or using prior knowledge to increase the number of training points along the shock trajectory, or adaptive activation functions. To address these issues, this study proposes a physics-informed neural network with long short-term memory and attention mechanism, an ingenious method to solve the Buckley-Leverett partial differential equations representing two-phase flow in porous media. The designed network structure overcomes the dependency on artificial diffusion terms and enhances the importance of shallow features. The experimental results show that the proposed method is in good agreement with analytical solutions. Accurate approximations are shown even when encountering shock points in saturated fields of porous media. Furthermore, experiments show our innovative method outperforms existing traditional physics-informed machine learning approaches.

No abstract

No abstract

Recent works investigated attacks on sensors by influencing analog sensor components with acoustic, light, and electromagnetic signals. Such attacks can have extensive security, reliability, and safety implications since many types of the targeted sensors are also widely used in critical process control, robotics, automation, and industrial control systems.While existing works advanced our understanding of the physical-level risks that are hidden from a digital-domain perspective, gaps exist in how the attack can be guided to achieve system-level control in real-time, continuous processes. This paper proposes an adversarial control loop-based approach for real-time attacks on process and actuation control systems relying on sensors. We study how to utilize the system feedback extracted from physical-domain signals to guide the attacks. In the attack process, injection signals are adjusted in real time based on the extracted feedback to exert targeted influence on a victim control system that is continuously affected by the injected perturbations and applying changes to the physical environment. In our case study, we investigate how an external adversarial control system can be constructed over sensor-actuator systems and demonstrate the attacks with program-controlled processes to manipulate the victim system without accessing its internal statuses.