Monitoring continuous BP signal is an important issue, because blood pressure (BP) varies over days, minutes, or even seconds for short-term cases. Most of photoplethysmography (PPG)-based BP estimation methods are susceptible to noise and only provides systolic blood pressure (SBP) and diastolic blood pressure (DBP) prediction. Here, instead of estimating a discrete value, we focus on different perspectives to estimate the whole waveform of BP. We propose a novel deep learning model to learn how to perform signal-to-signal translation from PPG to arterial blood pressure (ABP). Furthermore, using a raw PPG signal only as the input, the output of the proposed model is a continuous ABP signal. Based on the translated ABP signal, we extract the SBP and DBP values accordingly to ease the comparative evaluation. Our prediction results achieve average absolute error under 5 mmHg, with 70% confidence for SBP and 95% confidence for DBP without complex feature engineering. These results fulfill the standard from Association for the Advancement of Medical Instrumentation (AAMI) and the British Hypertension Society (BHS) with grade A. From the results, we believe that our model is applicable and potentially boosts the accuracy of an effective signal-to-signal continuous blood pressure estimation.

Due to the growing public awareness of cardiovascular disease (CVD), blood pressure (BP) estimation models have been developed based on physiological parameters extracted from both electrocardiograms (ECGs) and photoplethysmograms (PPGs). Still, in order to enhance the usability as well as reduce the sensor cost, researchers endeavor to establish a generalized BP estimation model using only PPG signals. In this paper, we propose a deep neural network model capable of extracting 32 features exclusively from PPG signals for BP estimation. The effectiveness and accuracy of our proposed model was evaluated by the root mean square error (RMSE), mean absolute error (MAE), the Association for the Advancement of Medical Instrumentation (AAMI) standard and the British Hypertension Society (BHS) standard. Experimental results showed that the RMSEs in systolic blood pressure (SBP) and diastolic blood pressure (DBP) are 4.643 mmHg and 3.307 mmHg, respectively, across 9000 subjects, with 80.63% of absolute errors among estimated SBP records lower than 5 mmHg and 90.19% of absolute errors among estimated DBP records lower than 5 mmHg. We demonstrated that our proposed model has remarkably high accuracy on the largest BP database found in the literature, which shows its effectiveness compared to some prior works.

This study introduced an app, called Smart UEnglish, for helping EFL conversation practices in authentic contexts. These conversation practices were categorized into ‘designed talk’ and ‘free talk’, based on the content of an English textbook and authentic ambient environment that includes such things as transportation, weather and scenic descriptions, respectively. The mechanisms in Smart UEnglish were designed in consideration of flexible, sustainable and adaptive conversations to improve English-speaking and lexical resources. Statistical analysis results show that the learning achievement of the experimental group is significantly higher than that of the control group. Moreover, both ‘designed talk’ and ‘free talk’ have significant positive correlations with learning achievement. Further analysis with stepwise multiple regression found that ‘designed talk’ plays the most important role and can significantly influence learning achievement. This is because ‘designed talk’ can motivate them to practice what they learned in class during their daily conversations and consolidate the understanding and applying of English vocabularies and sentence patterns through ‘designed talk’. According to interview records, participants also perceived that using Smart UEnglish for conversation practice in authentic contexts can stimulate their interests and ideas to think and practice more useful English conversation. Hence, Smart UEnglish can effectively help English-speaking and conversation in authentic contexts anytime and anywhere.

As an interdisciplinary research between watermarking and cryptography, privacy-aware reversible watermarking permits a party to entrust the task of embedding watermarks to a cloud service provider without compromising information privacy. The early development of schemes were primarily based upon traditional symmetric-key cryptosystems, which involve an extra implementation cost of key exchange. Although recent research attentions were drawn to schemes compatible with asymmetric-key cryptosystems, there were notable limitations in the practical aspects. In particular, the host signal must either be enciphered in a redundant way or be pre-processed prior to encryption, which would largely limit the storage efficiency and scheme universality. To relax the restrictions, we propose a novel research paradigm and devise different schemes compatible with different homomorphic cryptosystems. In the proposed schemes, the encoding function is recognised as an operation of adding noise, whereas the decoding function is perceived as a corresponding denoising process. Both online and offline contentadaptive predictors are developed to assist watermark decoding for various operational requirements. A three-way trade-off between the capacity, fidelity and reversibility is analysed mathematically and empirically. It is shown that the proposed schemes achieve the state-the-art performance.

Deep neural networks have revolutionised the research landscape of steganography. However, their potential has not been explored in invertible steganography, a special class of methods that permits the recovery of distorted objects due to steganographic perturbations to their pristine condition. In this paper, we revisit the regular-singular (RS) method and show that this elegant but obsolete invertible steganographic method can be reinvigorated and brought forwards to modern generation via neuralisation. Towards developing a renewed RS method, we introduce adversarial learning to capture the regularity of natural images automatically in contrast to handcrafted discrimination functions based on heuristic image prior. Specifically, we train generative adversarial networks (GANs) to predict bit-planes that have been used to carry hidden information. We then form a synthetic image and use it as a reference to provide guidance on data embedding and image recovery. Experimental results showed a significant improvement over the prior implementation of the RS method based on large-scale statistical evaluations.

Iris segmentation plays an important and significant role in the iris recognition system. The prerequisite for accurate iris recognition is the correctness of iris segmentation. However, the efficiency and robustness of traditional iris segmentation methods are severely challenged in a non-cooperative environment because of unfavorable factors, for instance, occlusion, blur, low resolution, off-axis, motion, and specular reflections. All of the above factors seriously reduce the accuracy of iris segmentation. In this paper, we present a novel iris segmentation algorithm that localizes the outer and inner boundaries of the iris image. We propose a neural network model called “Interleaved Residual U-Net” (IRUNet) for semantic segmentation and iris mask synthesis. The K-means clustering is applied to select saliency points set in order to recover the outer boundary of the iris, whereas the inner border is recovered by selecting another set of saliency points on the inner side of the mask. Experimental results demonstrate that the proposed iris segmentation algorithm can achieve the mean IOU value of 98.9% and 97.7% for inner and outer boundary estimation, respectively, which outperforms the existing approaches on the challenging CASIA-Iris-Thousand database.