Jingjing Luo

Lateral force blind spots frequently trigger airway injuries during robotic bronchoscopy. Traditional sensing approaches lack fine spatial resolution, and single-segment robots suffer poor posture adjustment flexibility, worsening unmonitored tissue contact. We integrate a snake-scale-inspired 64-channel gradient piezoelectric film with a dual-segment manipulator to resolve lateral blind spots for secure bronchial navigation. The conformally wrapped sensor delivers fast ( 0.98, RMSE < 0.145 N, CV < 2%) and filters bending pseudo-forces. The dual-segment design enables coordinated proximal-distal motion to avoid overcontact in tortuous lumens. Phantom and ex vivo porcine lung tests confirm real-time force localization, 5.0 N safety alarms, and autonomous tuning to keep contact forces below 3.0 N. This synergistic robot-sensor design supplies a translatable safety control strategy for flexible surgical manipulators.

With the evolution of mobile communication technologies, consumer medical electronics have emerged as a critical enabler for ubiquitous information exchange across Internet of Medical Things (IoMT). The integration of chaotic encryption and communication offers significant potential to enhance communication security by leveraging highly initial sensitivity and strong randomness of chaotic systems. This paper presents a novel hybrid chaotic encryption scheme and its integration with secure communication in consumer IoMT (CIoMT) systems. We propose a new chaotic system based on a Tabu learning neuron (TLN) and two nonlinear memristors, named chaotic TLN (CTLN), addressing the need for high security performance. The proposed CTLN is modeled by using two nonlinear memristors to describe the influence of external electromagnetic radiation on TLN. Dynamical analyses show that the CTLN has extremely complex chaotic characteristics, including butterfly attractors, multi-single-butterfly attractors, multi-double-butterfly attractors, and coexisting multi-butterfly attractors. Furthermore, in order to explore the feasibility of its hardware implementation, an analog CLTN circuit is designed and experimentally verified. Finally, we introduce a novel secure framework that integrates chaotic neural computing, DNA encoding and Compressed sensing to improve communication security in consumer IoMT. The security analysis reveals that the proposed hybrid cryptography solution exhibits robust resistance to differential attacks and high information entropy, effectively preserving the privacy and unaltered nature of sensitive medical images in the Internet of Medical Things. This paper aims to bridge the gap between information security and communication in CIoMT systems, offering insights into integrating biology-inspired technologies such as chaotic neural networks and DNA for more security consumer applications.

Precision medicine requires cuffless blood pressure (BP) estimation technologies to achieve high accuracy levels suitable for healthcare monitoring. Recent advances, notably in personalized BP estimation methods utilizing transfer learning, have partially met these requirements. However, traditional personalized fine-tuning pipelines often ignore key physiological factors across age groups, thus limiting their effectiveness. This study identifies an important fine-tuning principle: target subjects benefit most when fine-tuned using a source model pretrained on an age-matched group. We introduce a novel two-level ensemble learning method, ATTE, which integrates multiple BP models for cuffless BP estimation based on pulse waves. It first performs feature-level ensemble learning (FEL) to combine age-group-specific fine-tuned models and enrich representations, then applies an optimized Bayesian model averaging (BMA) method that adaptively weights model outputs to boost accuracy and reduce model-selection uncertainty. Using only 50 transfer samples, ATTE improved systolic and diastolic BP estimation accuracy by 6.55% and 5.58% on a public dataset, compared to the traditional fine-tuning paradigm. It also achieved improvements of 5.10% and 8.30% on a self-collected dataset, which has been preliminarily validated in an elderly care scenario. Ablation and interpretability analyses validated the age-matched fine-tuning principle, the effectiveness of ATTE, and its model-agnostic generalizability. This work represents a pioneering step in optimizing personalized BP estimation and underscores its significant potential for real-world healthcare applications.

Objective. Electroencephalography (EEG) records the spontaneous electrical activity in the brain. Despite the growing application of deep learning in EEG decoding, traditional methods still rely heavily on supervised learning, which is often limited by task specificity and dataset dependency, restricting model performance and generalization. Inspired by the success of large language models, EEG foundation models (EEG FMs) are attracting increasing attention as a unified paradigm for EEG decoding. In this study, we review a selection of representative studies on EEG FMs, aiming to extract trends and provide recommendations for future research. Approach. We provide a comprehensive analysis of recent advances in EEG FMs, with a focus on downstream tasks, benchmark datasets, model architectures, and pre-training techniques. We analyze and synthesize core FMs components, and systematically compare their performances and generalizabilities. Main results. Our review reveals that EEG FMs are pre-trained on large-scale datasets, typically involving several hundred subjects. The number of subjects can reach up to 14 987, with a maximum total duration of 27 062 h. Current EEG FMs most adopt mask-based reconstruction pre-training strategy and employ efficient transformer-based architectures. Our comparative analysis shows that EEG FMs demonstrate significant potential in advancing EEG decoding tasks, particularly in seizure detection. However, their performance in complex scenarios such as motor imagery decoding remains limited. Significance. This review summarizes the existing approaches and performance outcomes of EEG FM, offers valuable insights into their current limitations and delineates prospective avenues for future research.

The growing demand for non-wearable, long-term health monitoring solutions, especially for elderly, has driven the development of bed-based physiological monitoring systems. This letter introduces a novel non-contact, bed-based micromovement sensing system designed for accurate heart rate variability (HRV) monitoring. The system uses a polyvinylidene fluoride (PVDF) piezoelectric film sensor to detect ballistocardiogram (BCG) signals through a 10 cm thick mattress, ensuring both comfort and practicality in real-world settings. A deep learning model combining an attention mechanism, phase shift correction, and a ResUNet architecture is employed to reconstruct ECG-R waves from the BCG signals, enabling precise RR interval estimation. The system computes a comprehensive set of HRV metrics, including time-domain, frequency-domain, and non-linear indices, with high accuracy. Extensive experiments involving 20 subjects validate the system's performance, demonstrating a mean absolute error of less than 10 ms in RR interval estimation. 20 out of 23 HRV metrics evaluated had average relative errors under 15%. The results highlight the system's robustness and its ability to operate effectively through thick bedding materials, offering a reliable and practical solution for continuous health monitoring. Furthermore, the proposed technology has the potential to be integrated into future bed-based caregiving robots, enabling real-time health monitoring and autonomous care servicing, thereby enhancing caregiving efficiency and intelligence.