Search

Amyotrophic lateral sclerosis (ALS) patients suffer a great inconvenience in their daily lives due to the gradual loss of their motion abilities. In order to help ALS patients regain their self-care abilities at a certain level, an eye-tracking and brain–computer interface (BCI)-based human–environment interactive (HEI) system is proposed. Through a single-channel EEG recorder and a pair of eye-tracking glasses, the user’s attention levels and gaze points are detected. Based on the two types of intention information, the target objects are recognized by a You Only Look Once v5 (YOLOv5) model. Then, according to voice instructions, users can control the household electric appliances. The system performed a high success rate of 89.3% experimentally, providing a reliable and capable assistive solution for ALS patients.

Cyber security is of significance in today’s e-commerce applications. In this article, we present a piezoelectric touch sensing supported keystroke dynamics based identity authentication technique, for providing a secure access manner to smartphones. Here, the polyvinylidene fluoride (PVDF) based piezoelectric touch panel can learn detailed force touch habits of users. With a support vector machine (SVM) algorithm, our proposed frequency domain features experimentally demonstrate a better authentication accuracy of 98.3%, compared to the traditional time domain features. The work showcases a feasible method of combining functional materials and artificial intelligence (AI) for satisfying highly secure requirements.

Sleep apnea syndrome (SAS) is a high incidence respiratory disorder disease with potential mortality risk. Thus, researchers continuously present diverse techniques to monitor sleeping posture and respiratory rate (RR) which are closely correlated to the occurrence of SAS. Current techniques can implement high detection accuracy in both sleeping posture and RR, nevertheless, some undesired attributes still remain, such as high device cost, personal privacy leakage, and heavy burden on hardware. To satisfy the expectations, in this paper, we develop an integrated sleep monitoring system, based on flexible piezoresistive architectures and machine learning algorithms, in a low-cost and privacy protection manner. The experimental results successfully demonstrate the feasibility of the proposed technique, by showing a high sleeping posture classification accuracy of 98.1% and RR estimation accuracy of 97.5%, indicating a strong potential in practical utilization for SAS patients.

Teleoperation has been intensively used in diverse application scenarios in recent years. However, the frequently reported fatigue issue limits its successful use in tasks requiring long term and precise operations. One of key factors giving rise to the fatigue issue is the posture misalignment between the user’s arm and the robotic arm. Although conventional computer vision (CV) based solutions can implement postures alignment, they introduce potential privacy leakage problems at the same time. To address this issue, in this article we present an inertial measurement unit based technique for learning the posture status of the user’s arm and the robotic arm. Through a smart mapping algorithm, the proposed technique is proved to be a feasible manner to synchronize the two postures, potentially reducing the mental fatigue and preventing the privacy leakage possibilities.