Background: Reminiscence therapy (RT) is a commonly used nonpharmaceutical treatment for cognitive impairment. Artifacts or conversations are used in RT to recall individuals’ memories and past experiences. Virtual reality (VR) has increasingly been used as an assistive technology during RT. However, the effects of VR-based RT (VR-RT) methods remain unclear, and insights into the related benefits and challenges are urgently needed. Objective: The study aims to systematically review the effects of VR-RTs for older adults with cognitive impairment. Methods: Seven databases (MEDLINE, Academic Search Premier, CINAHL, Web of Science, PubMed, the Cochrane Central Register of Controlled Trials, and ScienceDirect) were searched to identify relevant articles published from inception to August 10, 2023. Peer-reviewed publications that assessed the effect of VR-RTs (ie, using virtual clues to evoke participants’ memories or past experiences) on cognitive-related outcomes were included. Two independent researchers conducted the literature search, review, and data extraction processes. A narrative synthesis approach was used to analyze the extracted data. Results: Of the 537 identified articles, 22 were ultimately included in the data analysis. The results revealed that VR-RTs could maintain cognitive status (4/4, 100%) and reduce anxiety (2/2, 100%) in older adults with cognitive impairment. Nevertheless, one study found a cognitive improvement after VR-RTs, whereas cognitive degradation was observed at a 3- to 6-month follow-up measure. Around 88% (7/8) of the included studies indicated that VR-RTs improved memory; however, the evidence regarding the beneficial effects of VR-RTs was limited in improving quality of life (1/4, 25%) and reducing apathy (0/2, 0%) and depression (1/3, 33%). The results indicated that VR-RTs are safe, engaging, acceptable, and satisfying for older adults with cognitive impairment. In VR scenarios, personalized stimulus materials related to the users’ youth experiences were more effective for treating cognitive impairment than other stimulus materials. Conclusions: The results of this systematic review demonstrate the potential benefits of VR-RT for older adults with cognitive impairment, especially in improving emotion and memory and maintaining cognitive status. VR-RT is also safe and enjoyable for older adults. However, due to the trial heterogeneity of included studies, we can only provide qualitative results instead of performing meta-analysis to quantify the effect size of VR-RTs. Thus, more randomized controlled trials are required to examine the designs and effects of VR-RTs for groups of older adults with specific needs.

With the popularization of humans working in tandem with robots and artificial intelligence (AI) by Industry 5.0, ultrasonic non-destructive testing (NDT)) technology has been increasingly used in quality inspections in the industry. As a crucial part of handling ultrasonic testing results–signal processing, the current approach focuses on professional training to perform signal discrimination but automatic and intelligent signal optimization and estimation lack systematic research. Though the automated and intelligent framework for ultrasonic echo signal processing has already exhibited essential research significance for diagnosing defect locations, the real-time applicability of the algorithm for the time-of-flight (ToF) estimation is rarely considered, which is a very important indicator for intelligent detection. This paper conducts a systematic comparison among different ToF algorithms for the first time and presents the auto-diagnosis of the ToF approach based on the Defect Peaks Tracking Model (DPTM). The proposed DPTM is used for ultrasonic echo signal processing and recognition for the first time. The DPTM using the Hilbert transform was verified to locate the defect with the size of 2–10 mm, in which the wavelet denoising method was adopted. With the designed mechanical fixture through 3D printing technology on the pipeline to inspect defects, the difficulty of collecting sufficient data could be conquered. The maximum auto-diagnosis error could be reduced to 0.25% and 1.25% for steel plate and pipeline under constant pressure, respectively, which were much smaller than those with the DPTM adopting the cross-correlation. The real-time auto-diagnosis identification feature of DPTM has the potential to be combined with AI in future work, such as machine learning and deep learning, to achieve more intelligent approaches for industrial health inspection.

In the research of breast morphology, numerous breast features are measured, whereas only a few parameters are adopted for classification. Therefore, how to extract the key variables from the multi-dimensional features in a rational way is an issue that is focused upon. This study aimed to reduce the complexity of the dimensionality reduction for further improving the objectivity and interpretability of the selected breast features. Since the random forest (RF) algorithm can quantify the feature importance during training, the method was adopted to determine the optimal breast features for classification and recognition in this paper. Firstly, the anthropometric data of 360 females from northwestern China aged from 19 to 27 years were measured by non-contact three-dimensional body scanning technology and the contact manual measurement method. Then, the k-means clustering was applied to categorize breast shapes, and the RF algorithm was utilized to quantify and rank the importance of 25 breast features. Finally, to verify the availability of the RF algorithm on breast feature selection, the t-distributed stochastic neighbor embedding method was adopted to visualize the distribution of breast shape clusters into two dimensions. Meanwhile, four neural networks were determined to recognize the breast morphology. The results demonstrate that fewer breast features can effectively increase the accuracy of breast shape classification and recognition. The best performance of breast shape classification and recognition is obtained when the number of breast features is 13. In this case, the average Hamming loss of four neural networks is the smallest (0.1136). Interestingly, the bust circumference and the horizontal curve of breasts across the bust points are found to be the most important of the 25 breast features in this paper. The importance of the breast curve features is higher than that of the breast cross-sectional features, while the breast positioning features have the lowest importance. Meanwhile, the RF algorithm is verified to be more effective than traditional dimensionality reduction methods, such as principal component analysis, hierarchical clustering, and recursive feature elimination. The approach developed in this paper can be generalized to the dimensionality reduction of other body morphology.

As the world grapples with the challenges posed by climate change and depleting energy resources, achieving sustainability in the construction and operation of buildings has become a paramount concern. The construction and operation of buildings account for a substantial portion of global energy consumption and carbon emissions. Hence, the accurate prediction of building energy consumption is indispensable for reducing energy waste, minimizing greenhouse gas emissions, and fostering sustainable urban development. The aspiration to achieve predicted outcomes with remarkable accuracy has emerged as a pivotal objective, coinciding with the burgeoning popularity of deep learning techniques. This paper presents an auto-evaluation model for building energy consumption prediction via Long Short-Term Memory with modified Kalman filtering (LSTM-MKF). Results gleaned from data validation activities evince a notable transformation—a reduction of the maximal prediction error from an initial 83% to a markedly ameliorated 24% through the intervention of the proposed model. The LSTM-MKF model, a pioneering contribution within this paper, clearly exhibits a distinct advantage over the other models in terms of predictive accuracy, as underscored by its superior performance in all three key metrics, including mean absolute error, root mean square error, and mean square error. The model presents excellent potential as a valuable tool for enhancing the precision of predictions of building energy consumption, a pivotal aspect in energy efficiency, smart city development, and the formulation of informed energy policy.

Gait instability is critical in medicine and healthcare, as it has associations with balance disorder and physical impairment. With the development of sensor technology, despite the fact that numerous wearable gait detection and recognition systems have been designed to monitor users’ gait patterns, they commonly spend a lot of time and effort to extract gait metrics from signal data. This study aims to design an artificial intelligence-empowered and economic-friendly gait monitoring system. A pair of intelligent shoes with a single inertial sensor and a smartphone application were developed as a gait monitoring system to detect users’ gait cycle, stand phase time, swing phase time, stride length, and foot clearance. We recruited 30 participants (24.09 ± 1.89 years) to collect gait data and used the Vicon motion capture system to verify the accuracy of the gait metrics. The results show that the gait monitoring system performs better on the assessment of the gait metrics. The accuracy of stride length and foot clearance is 96.17% and 92.07%, respectively. The artificial intelligence-empowered gait monitoring system holds promising potential for improving gait analysis and monitoring in the medical and healthcare fields.

In this work, the deformation of bilateral breasts was investigated with an established hybrid model to predict the nipple movement specifically for senior women during yoga exercise. A motion capture system was used to collect the displacement of 10 markers on the breasts from 11 senior women (average age of 62) during yoga practice and then the data were analyzed by integrating the absolute grey relation analysis (AGRA) and extreme learning machine (ELM). The right and left breasts had the maximum motion amplitude in the horizontal direction but they were respectively featured with contraction and extension during yoga practice. AGRA showed that the nipple motion was highly associated with the vertical region above the nipple for the left breast but the parallel region along with the nipple for the right breast. The ELM model is able to predict the nipple movement within tolerable error (∼0.0037). This study lays a foundation for a better understanding of ageing breast kinematics during yoga poses with limited practical experiments. Besides, the accurate and efficient results can be used not only for yoga pose instruction but also for ergonomic sports bra design.