Physical inactivity and sedentary behavior (SB) have attracted growing attention globally since they relate to noninfectious chronic diseases (NCDs) and could further result in the loss of life. This systematic literature review aimed to identify existing evidence on the efficacy of mobile health (mHealth) technology in inducing physical activity and reducing sedentary behavior for physically inactive people. Studies were included if they used a smartphone app in an intervention to improve physical activity and/or sedentary behavior for physically inactive individuals. Interventions could be stand-alone interventions or multi-component interventions, including an app as one of several intervention components. A total of nine studies were included, and all were randomized controlled trials. Two studies involved interventions delivered solely via a mobile application (stand-alone intervention) and seven studies involved interventions that used apps and other intervention strategies (multi-component intervention). Methodological quality was assessed, and the overall quality of the studies was ensured. The pooled data favored intervention in improving physical activity and reducing sedentary behavior. This review provided evidence that mobile health intervention improved physical activity and reduced sedentary behavior among inactive individuals. More beneficial effects can be guaranteed when interventions include multiple components. Further studies that maintain the effectiveness of such interventions are required to maximize user engagement and intervention efficacy.

Purpose: Artificial intelligence (AI) has been touted as one of the viable solutions to address urban mobility issues. Despite a growing body of research on AI across various sectors, its use in the mobility sector remains underexplored. This study addresses this limitation by investigating AI applications and identifying the AI material properties and use cases that offer mobility-specific affordances.Design/methodology/approach: Although AI applications in mobility are growing, academic research on the subject has yet to catch up. Therefore, we follow a systematic review and analysis of practitioner literature. We conducted a comprehensive search for relevant documents through Advanced Google and OECD databases and identified 173 sources. We selected 40 sources published between 2015 and 2022 and analysed the corpus of evidence through abductive qualitative analysis technique.Findings: The analysis reveals that mobility organisations are implementing various AI technologies and systems such as cameras, sensors, IoT, computer vision, natural language processing, robotic process automation, machine learning, deep learning and neural networks. These technologies offer material properties for sensing mobility objects and events, comprehending mobility data, automating mobility activities and learning from mobility data. By exploiting these material properties, mobility organisations are integrating urban mobility management, personalising and automating urban mobility, enabling the smartification of infrastructure and asset management, developing better urban transport planning and management, and enabling automatic driving.Originality/value: The study contributes a mid-range theory of the affordances of AI for mobility (AI4M) at the infrastructure, operation and service levels. This contribution extends the existing understanding of AI and offers an interconnected perspective of AI affordances for further research. For practitioners, the study provides insights on how to explore AI in alignment with organisational goals to collectively transform urban mobility to be affordable, efficient and sustainable.

The 3D quantitative analysis of facial morphology is of importance in plastic surgery (PS), which could help surgeons design appropriate procedures before conducting the surgery. We propose a system to simulate and guide the shaping effect analysis, which could produce a similar but more harmonious face simulation. To this end, first, the depth camera based on structured light coding is employed for facial 3D data acquisition, from which the point cloud data of multiple facial perspectives could be obtained. Next, the cascade regression tree algorithm is used to extract the esthetic key points of the face model and to calculate the facial features composed of the key points, such as the nose, chin, and eyes. Quantitative facial esthetic indexes are offered to doctors to simulate PS. Afterward, we exploit a face mesh metamorphosis based on finite elements. We design several morphing operators, including augmentation, cutting, and lacerating. Finally, the regional deformation is detected, and the operation effect is quantitatively evaluated by registering the 3D scanning model before and after the operation. The test of our proposed system and the simulation of PS operations find that the measurement error of facial geometric features is 0.458 mm, and the area is 0.65 mm². The ratings of the simulation outcomes provided by panels of PS prove that the system is effective. The manipulated 3D faces are deemed more beautiful compared to the original faces respecting the beauty canons such as facial symmetry and the golden ratio. The proposed algorithm could generate realistic visual effects of PS simulation. It could thus assist the preoperative planning of facial PS.

The morphological characteristics of retinal vessels in the fundus serve as the primary basis for diagnosing and assessing the risk of ophthalmic diseases. An effective segmentation scheme for retinal vessels can aid in the early diagnosis and treatment of these diseases, as well as help prevent their progression. However, accurate vessel segmentation is challenging due to the low contrast of fundus images and the complexity of the vessels’ morphological structure. To address the low sensitivity and poor generalization ability of the existing methods in vascular extraction, a Dual-decoder Network based on Attention-enhancing and multi-scale Feature Fusion (AFFD-Net) is proposed. AFFD-Net inherits the codec concept of U-Net. To improve the performance of our U-Net model, we made two modifications. Firstly, we reduced the number of convolution kernel filters in each layer, thereby significantly reducing the number of training parameters. This helps to avoid overfitting and improves the model’s ability to generalize. Secondly, we added a Multi-scale Feature Extraction (MFE) module and an M/A intermediate decoder to enhance the model’s sensitivity. MFE is designed as the first encoding unit of AFFD-Net to obtain rich vascular features in the complex anatomical background and adapt to the large-scale variations of vessels. The M/A intermediate decoder is composed of the Multi-scale Feature Fusion (MFF) module and the Attention-enhancing Hybrid Feature Fusion (AHFF) module. The MFF module integrates deep semantic information and shallow spatial information to ensure that the features at each scale in the middle layer are fully utilized. The AHFF module adaptively fuses the hybrid features at different scales to generate two feature descriptors with different focuses which can improve the expressiveness of the model. AFFD-Net is evaluated on three public databases including DRIVE, STARE, and CHASE_DB1, and the sensitivity values obtained are 84.19%, 84.58%, and 82.62%, respectively. It has higher sensitivity and better generalization ability than other state-of-the-art methods. Compared with classical networks including U-Net, U-Net++, and U-Net3+, AFFD-Net has fewer parameters and higher segmentation accuracy. Our proposed segmentation model exhibits superior performance across a range of metrics, indicating its promising potential for practical applications.

Biomedical imaging plays a critical role in medical diagnostics and research, yet image noise remains a significant challenge that hinders accurate analysis. To address this issue, we propose BDNet, a real-time biomedical image denoising network optimized for enhancing gradient and high-frequency information while effectively suppressing noise. The network adopts a lightweight U-Net-inspired encoder–decoder architecture, incorporating a Convolutional Block Attention Module at the bottleneck to refine spatial and channel-wise feature extraction. A novel gradient-based loss function—combining Sobel operator-derived gradient loss with L1, L2, and LSSIM losses—ensures faithful preservation of fine structural details. Extensive experiments on the Fluorescence Microscopy Denoising (FMD) dataset demonstrate that BDNet achieves state-of-the-art performance across multiple metrics, including PSNR, RMSE, SSIM, and LPIPS, outperforming both convolutional and Transformer-based models in accuracy and efficiency. With its superior denoising capability and real-time inference speed, BDNet provides an effective and practical solution for improving biomedical image quality, particularly in fluorescence microscopy applications.

Structural magnetic resonance imaging (sMRI) plays a pivotal role in the diagnosis of Alzheimer's disease (AD), owing to its high-resolution whole-brain imaging capability. Recent studies have shifted from relying on prior knowledge-based extraction of sMRI features to employing deep learning approaches for automatic feature extraction. However, the deep learning-based approaches may overlook the interrelationships among features from various dimensions, which focus solely on inter-slice information or spatial structural features, thereby limiting the comprehensiveness of brain lesion characterization. To address this limitation, we propose a multi-dimensional cooperative analysis (MCA) framework designed to integrate multi-dimensional feature information effectively. The framework comprises two stages. In the first stage, multi-frequency aware (MFA) downsampling utilizes a novel multi-frequency aware fusion module based on a three-dimensional Sym2 wavelet transform to extract multi-scale features. In the second stage, the multi-view slice learner (MVSL) and the spatial structural learner (SSL), which respectively employ a multi-view encoder and a graph hybrid attention convolutional block, collaborate to capture AD lesion features from diverse perspectives. Extensive experimental validation demonstrates that the MCA framework achieves an accuracy of 95.12% and a sensitivity of 94.07% on the ADNI dataset, outperforming recent state-of-the-art methods. It also exhibits robust generalization on the AIBL and OASIS datasets. SHAP-based analysis provides interpretability to the MCA framework by revealing the differential contributions of various brain regions in AD diagnosis.