Along with the development of Fintech, many scholars have studied how information technology is applied to financial services with a focus on extended methods for application. Few scholars have studied the influence mechanism behind the adoption of Fintech services. This paper proposes an improved technology acceptance model (TAM) that incorporates user innovativeness, government support, brand image, and perceived risk as determinants of trust to investigate how users adopt Fintech services. We designed a questionnaire, sent it to active customers of the Hefei Science and Technology Rural Commercial Bank, and obtained 387 eligible responses. We analyzed the data with a structural equation model (SEM) to test the hypotheses, including the relationships of all latent variables. The results reveal that users’ trust in Fintech services has a very significant influence on users’ attitudes for adoption. In addition, perceived ease of use and perceived risk does not affect users’ attitudes toward the adoption regarding Fintech services. This study contributes to the literature of the adoption of Fintech services by providing a more comprehensive view of the determinants of users’ attitudes by combining trust of Fintech services with TAM.
Recently, a variety of artificial intelligence (AI)-driven smart healthcare services are rapidly emerging in the medical market, such as intelligent image analysis, surgical navigation systems, and aided diagnosis. However, one of the major challenges is practitioners’ hesitation and unwillingness to employ these new technologies in medical practice. This study focuses on identifying the influential factors of adoption intention of smart healthcare services for both clinicians and non-clinicians from the perspective of technology transfer. Through collecting 484 questionnaire data from doctors in Anhui, China, we find support to show that perceived usefulness (PU), attitude, and the experience of using mHealth are key factors that influence both clinicians and non-clinician’s adoption intention. Meanwhile, it is confirmed that subjective norm has a positive effect on only clinicians’ behavioural intention (BI) while perceived risk (PR) has a negative impact on only non-clinicians’ attitude. Among all the constructs, the experience of using mHealth has the strongest positive effect on doctors’ adoption intention on smart healthcare services, a positive impact on the PU and perceived ease of use, and a negative impact on the PR. This study provides an improved understanding of doctors’ BI of smart healthcare services, and practice guidance for product development and marketing strategies.
Virtual reality technology has been applied in minimally invasive surgery to provide doctors with a more immersive surgical experience. The key to build the virtual reality model is estimation of depth and motion . In this study, a fully unsupervised learning method for depth and motion estimation with the continuous monocular endoscopic video is proposed. After highlighted region detection, EndoMotionNet and EndoDepthNet are designed to estimate ego-motion and depth simultaneously. The timing information between consecutive frames are considered with a long short-term memory layer by EndoMotionNet to enhance the accuracy of ego-motion estimation. The estimated depth of the previous frame is used to estimate the depth of the next frame in EndoDepthNet with a multi-mode fusion mechanism. The custom loss function is designed to improve the accuracy of the unsupervised learning method. Experiments verify that the proposed method can effectively improve the accuracy of depth and motion estimation.
25 July 2023 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Facial video-based remote physiological measurement aims to estimate remote photoplethysmography (rPPG) signals from human facial videos and then measure multiple vital signs (e.g. heart rate, respiration frequency) from rPPG signals. Recent approaches achieve it by training deep neural networks, which normally require abundant facial videos and synchronously recorded photoplethysmography (PPG) signals for supervision. However, the collection of these annotated corpora is not easy in practice. In this paper, we introduce a novel frequency-inspired self-supervised framework that learns to estimate rPPG signals from facial videos without the need of ground truth PPG signals. Given a video sample, we first augment it into multiple positive/negative samples which contain similar/dissimilar signal frequencies to the original one. Specifically, positive samples are generated using spatial augmentation; negative samples are generated via a learnable frequency augmentation module, which performs non-linear signal frequency transformation on the input without excessively changing its visual appearance. Next, we introduce a local rPPG expert aggregation module to estimate rPPG signals from augmented samples. It encodes complementary pulsation information from different face regions and aggregates them into one rPPG prediction. Finally, we propose a series of frequency-inspired losses, i.e. frequency contrastive loss, frequency ratio consistency loss, and cross-video frequency agreement loss, for the optimization of estimated rPPG signals from multiple augmented video samples. We conduct rPPG-based heart rate, heart rate variability, and respiration frequency estimation on five standard benchmarks. The experimental results demonstrate that our method improves the state of the art by a large margin. Our codes will be available at https://github.com/yuezijie/Video-based-Remote-Physiological-Measurement-via-Self-supervised-Learning .