Search

To address the problem of business process prediction in smart healthcare involving the fusion of multimodal data from interactions among multiple systems, this paper proposes an object-centered explainable prediction method (OCPPP). The approach comprises three modules: 1) Utilizing control flow and data flow constraints between activities for process modelling, Algorithm 1 constructs a Petri net centered on the object to be predicted; 2) Recognizing that business activities share resources during system interactions, individual activities are analysed using AI models, with distinct models applied to different modalities of data in activity logs (e.g., multi-object detection for image data, time-series forecasting for text data); 3) Employing coloured Petri nets and Algorithm 2 for predicting activity durations, integrating outputs from various intelligent models to formulate predictions, including those for low-frequency events. The experimental outcomes indicate that during both individual component analysis and sequential forecasting phases, a versatile selection of AI models enables effective operation. Integrating Petri Nets for system-level predictions enhances explainability through six distinct service compositions. Furthermore, the introduction of precursor transitions and a waiting threshold facilitates the anticipation of infrequent behaviours, thereby augmenting the system’s predictive capabilities for a broader spectrum of occurrences.

Brain-Computer Interface (BCI) is a revolutionary technique that employs wearable electroencephalography (EEG) sensors and artificial intelligence (AI) to monitor and decode brain activity. EEG-based motor imagery (MI) brain signal is widely utilized in various BCI fields including intelligent healthcare, robot control, and smart homes. Yet, the limited capability of decoding brain signals remains a significant obstacle to BCI techniques expansion. In this study, we describe an architecture known as the dual-branch attention temporal convolutional network (DB-ATCNet) for EEG-based MI classification. DB-ATCNet improves MI classification performance with relatively fewer parameters by utilizing a dual-branch convolutional network and channel attention. The DB-ATCNet model consists of two primary modules: attention dual-branch convolution (ADBC) and attention temporal fusion convolution (ATFC). The ADBC module utilizes a dual-branch convolutional network to extract low-level MI-EEG features and incorporates channel attention to improve spatial feature extraction. ATFC employs sliding windows with self-attention to obtain the high-level temporal features, and utilizes feature fusion strategies to minimize information loss. The DB-ATCNet achieved subject-independent accuracies of 87.33% and 69.58% in two-class and four-class classification tasks, respectively, on the PhysioNet dataset. On the BCI Competition IV-2a dataset, it achieved an accuracy of 71.34% and 87.54% for subject-independent and subject-dependent evaluations, respectively, surpassing existing methods. The code is available at https://github.com/zk-xju/DB-ATCNet.

This paper proposes a posture recognition system that can be applied for medical surveillance. The proposed method estimates human posture using mobilenetV2 and long short-term memory (LSTM) to extract the important features of an image. The output of the system was a fully estimated skeleton. We used seven human indoor postures, including lying, sitting, crouching, standing, walking, fighting, and falling, and classified them. The output results are the extraction of the human skeleton and the corresponding labels for the poses. We first experiment with classification using machine learning. The system only achieves approximately 88% accuracy because it is not able to classify similar postures, such as standing and walking. This difference can be caused by the extraction of features for static images, and the machine learning classification algorithm has not reached accuracy with training data. Therefore, we proposed the integration of the LSTM model into the proposed system. LSTM learns the features of the skeleton and provides classification results for postures. As a result, our system improved the accuracy by up to 99%. Similar postures, such as standing and walking, have improved accuracy by up to 7%. In addition, we performed the system on the Jetson Nano hardware. The results show that it can run on a low-profile (44% CPU and 2.1 frames per second) that is capable of applications for remote patient monitoring devices.

6G technology is now attracting many scientific researchers due to its prominent features including high mobility, high data rate, high operating frequency, and ultra-low end-to-end delay. Compared to the 5G, 6G has certain advantages that can transform smart healthcare to a new advanced intelligent healthcare system, where multiple issues and concerns of the services (e.g., slow ambulance response) are effectively resolved. In such a communication system, patients and service providers communicate with each other via an open internet channel. Since the healthcare data is important and very sensitive, security and privacy in the healthcare network become prominent. In this paper, we introduce a 6G-aided intelligent healthcare environment. Our work also proposes a solution called Centerless User-Controlled Single Sign-On (CL-UCSSO) for achieving a convenient and cost-saving communication in a multi-server system constructed. A three-factor mechanism (combining smart card, password and biometrics) and time bound property are integrated to design the protocol with fast authentication that allows patients and providers to efficiently establish secure communications. Security proof of the proposed protocol is provided using well-known verification tools including RoR model, AVISPA simulation and BAN logic. Results of performance comparisons on various aspects show that our work provides more functionalities and incurs less cost compared with the related works.

With the rapid economic development and the increasing pressure of work and life, people are increasingly eager to obtain real-time intelligent healthcare to monitor their sub-healthy bodies. Healthcare wireless body area network (HWBAN) is one of the key infrastructures to realize intelligent healthcare. Health data is extremely private information, which makes adversaries prefer to carry out data-related interception, modification, and destruction attacks against the HWBAN system. To provide secure protection for HWBAN, many researchers have proposed various HWBAN schemes, but there are still deficiencies in security, performance, and availability. Our paper proposes a lightweight and amplified secure scheme for HWBAN by using fewer Elliptic Curve Cryptography(ECC) operations and Physically Unclonable Function(PUF) to improve security and efficiency at the same time. Furthermore, we allow users to know their health status in real-time through their mobile phones without initiating additional requests to the medical server in our system to achieve better availability. In addition, we provide strict formal security proof to demonstrate the proposed scheme meets the security and reliability requirements in this paper. The detailed comparative analysis illustrates that the proposed scheme has certain advantages in computing, communication, and security.