Search

Product returns will soon cost firms a trillion dollars annually; e-commerce appears more at risk for these costs relative to offline channels. Retailers thus need strategic insights into which factors increase customers’ return propensity and when the short-term costs of returns might be offset by future customer purchases. This article reports two studies that use transaction data from two large apparel e-tailers. Study 1 demonstrates that due to the differences between mobile and traditional online channels, consumers form a larger consideration set when they shop through mobile channels, and this distinctive search practice spills over onto return rates. That is, return rates are lower in mobile channels, and the use of the mobile channel also alters the effect of discount promotions. The findings in Study 2 suggest that future purchases by consumers, after a return, are contingent on category characteristics. In categories in which customers can easily learn from their previous return experience, product returns positively increase future spending, but the opposite is true in categories in which it is difficult to learn.

Does a bank’s dependence on different external funding sources shape its voluntary disclosure of information? We evaluate whether economic shocks that increase the supply of bank deposits alter the cost–benefit calculations of bank managers concerning voluntary information disclosure. We measure information disclosure using 10-K filings, 8-K filings, and earnings guidance. As for the funding shock, we use unanticipated technological innovations that triggered shale development and booms in bank deposits. Further analyses suggest that greater exposure to shale development reduced information disclosure by relaxing the incentives for managers to disclose information to attract funds from external capital markets. This paper was accepted by Kay Giesecke, finance.

Objective: Tele-health care service of alternative practice for chronic pain disease is worthwhile of developing, especially in the period of COVID- 19 pandemic. Targeting on myofascial trigger points, this study was performed to assess the possible short-term pain relief and functional improvement in patients by applying the device of BIOCERAMIC material enhanced by frequencies of tempo sound and visible light spectrum (BioS & L). Methods: Fourteen patients who participated in the procedure for the selection of trigger points for the BioS &L treatment, assessment of pain levels using a visual analog scale (VAS) analysis, and detection of abnormal resonance of 12 harmonic frequencies using a quantum resonance spectrometer (QRS). Results: Comparing the pre- and post-treatment of BioS &L on pain score of 12 HFs (V1-V12) as measured by VAS estimated by mixed model showed 91.7% (11/12) improvement with statistically significant results. The distribution of differences in the QRS score estimated by the mixed model among participants with pre-test QRS level ≥ 2 showed 83.3% (15/16) of HFs with statistically significant results. Conclusion: Treatment of BioS &L at trigger points providing pain relief is explained by the hypothesis of microvascular physiology and physics of wave propagation. This study provides a workshop with a concept of digitalization of complementarity and traditional medical service and tele-health care, which fulfills distant data connection and remote practice. In the period of epidemic spread, it helps to decrease close contact on both health care providers and patients.

Background: Previous epidemiological investigations examining the association between Kawasaki disease (KD) and cerebrovascular disease have had conflicting results. We analyzed the association between KD and cerebrovascular disease by conducting a population-based retrospective cohort study designed to investigate the hypothesis that KD could be a risk factor for subsequent cerebrovascular disease. Methods: From the National Health Insurance Research Database of Taiwan, the data of children (aged 0–18 years old) with KD (n=8467) were collected. Starting with the first year of study observation (referred to as the baseline year), data was collected for each child with KD, and 4 non-KD patients matched for sex, urbanization level of residence, and parental occupation were randomly selected to form the non-KD cohort (n=33 868) for our analysis. For the period from January 1, 2000, to December 31, 2012, we calculated the follow-up person-years for each patient, which is the time from the index date to the diagnosis of cerebrovascular disease, death, or the end of 2012. Furthermore, we compared the incidence, the incidence rate ratio, and the 95% CI of cerebrovascular disease between the KD and non-KD cohorts. Results: The overall cerebrovascular disease incidence rate was found to be 3.19-fold higher, which is significantly higher, in the KD cohort than in the non-KD cohort (14.73 versus 4.62 per 100 000 person-years), and the overall risk of cerebrovascular disease remained higher in the KD cohort (adjusted hazard ratio, 3.16 [95% CI, 1.46–6.85]). Furthermore, children aged Conclusions: This nationwide retrospective cohort study shows that KD may increase the risk of subsequent cerebrovascular disease, especially in those with KD aged

Machine learning has been widely used for solving challenging problems in diverse areas. However, to the best of our knowledge, seldom literature has discussed in-depth how machine learning approaches can be used effectively to “hunt” (identify) threats, especially advanced persistent threats (APTs) , in a monitored environment. In this study, we share our past experiences in building machine learning-based threat-hunting models. Several challenges must be considered when a security team attempts to build such models. These challenges include (1) weak signal, (2) imbalanced data sets, (3) lack of high-quality labels, and (4) no storyline. In this study, we propose Fuchikoma and APTEmu to demonstrate how we tackle the above-mentioned challenges. The former is a proof of concept system for demonstrating the ideas behind autonomous threat-hunting. It is a machine learning-based anomaly detection and threat hunting system which leverages natural language processing (NLP) and graph algorithms. The latter is an APT emulator, which emulates the behavior of a well-known APT called APT3, which is the target used in the first round of MITRE ATT&CK Evaluations. APTEmu generates attacks on Windows machines in a virtualized environment, and the captured system events can be further used to train and enhance Fuchikoma’s capabilities. We illustrate the steps and experiments we used to build the models, discuss each model’s effectiveness and limitations of each model, and propose countermeasures and solutions to improve the models. Our evaluation results show that machine learning algorithms can effectively assist threat hunting processes and significantly reduce security analysts’ efforts. Fuchikoma correctly identifies malicious commands and achieves high performance in terms of over 80% True Positive Rate and True Negative Rate and over 60% F3. We believe our proposed approaches provide valuable experiences in the area and shed light on automated threat-hunting research.

SrFBiS2 is a quaternary n-type semiconductor with rock-salt-type BiS2 and fluorite-type SrF layers alternately stacked along the c axis. The tunability of the crystal and electronic structures as well as the intrinsically low thermal conductivity make this compound a promising parent material for thermoelectric applications. In the current work, we show that alloying of Se and S in SrFBiS2 reduces the optical band gap with the second conduction band serving as an electron-transport medium, simultaneously increasing the electron concentration and effective mass. In addition, the raw material Bi2Se3 is shown to act as liquid adjuvant during the annealing process, favoring preferred-orientation grain growth and forming strengthen microstructural texturing in bulk samples after hot-pressed sintering. Highly ordered lamellar grains are stacked perpendicular to the pressure direction, leading to enhanced mobility along this direction. The synthetic effect results in a maximum power factor of 5.58 μW cm-1 K-2 at 523 K for SrFBiSSe and a peak zT = 0.34 at 773 K, enhancements of 180% compared with those of pristine SrFBiS2.

Liver function test is the first step to diagnose various liver diseases by measuring certain proteins and liver enzymes from the blood sample. After getting the required data of the clinical parameters from the blood test it is possible to calculate Child-Pugh (CTP), AST to PLT ratio (APRI) and Model for end-stage liver disease (MELD) clinical scores that help the doctors about the severity of the disease progression. Volatile organic compounds (VOCs) found in-breath and monitoring their concentration may be a prevailing method for disease diagnosis. In this work, Isoprene, Limonene, and Dimethyl sulphide (DMS) are considered as a potential breath biomarker related to liver disease. A dataset is designed, that includes the biomarkers concentration analysed from the breath sample before and after study subjects performed an exercise. Four regression methods are performed to predict the clinical scores using breath biomarkers data as features set by the machine learning techniques. Regression methods on the dataset for prediction are evaluated by mean absolute error and root mean square error. A significant difference observed for isoprene concentration (p

Recommender system has recently received a lot of attention in the information service community. In many application scenarios, such as Internet of Things (IoTs) environments, item multimodal auxiliary information (such as text and image) can be obtained to expand their feature representation and to increase user satisfaction with recommendations. Motivated by this fact, this paper introduces a novel two-stage embedding model (TSEM), which adequately leverage item multimodal auxiliary information to substantially improve recommendation performance. Specifically, it encompasses two sequential stages: graph convolutional embedding (GCE) and multimodal joint fuzzy embedding (MJFE). In the former, we first generate a bipartite graph for user-item interactions, and then utilize it to construct user and item backbone features via a spatial-based graph convolutional network (SGCN). While in the latter, by employing item multimodal auxiliary information, we integrate multi-task deep learning, deterministic Softmax, and fuzzy Softmax into a convolutional neural network (CNN)-based learning framework, which is optimized to obtain user backbone features and item semantic-enhanced fuzzy (SEF) features accurately. After TSEM converges, user backbone features and item SEF features can be utilized to calculate user preferences on items via Euclidean distance . Extensive experiments over two real-world datasets show that the proposed TSEM model significantly outperforms the state-of-the-art baselines in terms of various evaluation metrics .

The aim of this paper is to estimate the six-degree-of-freedom (6DOF) poses of objects from a single RGB image in which the target objects are partially occluded. Most recent studies have formulated methods for predicting the projected two-dimensional (2D) locations of three-dimensional keypoints through a deep neural network and then used a PnP algorithm to compute the 6DOF poses. Several researchers have pointed out the uncertainty of the predicted locations and modelled it according to predefined rules or functions, but the performance of such approaches may still be degraded if occlusion is present. To address this problem, we formulated 2D keypoint locations as probabilistic distributions in our novel loss function and developed a confidence-based pose estimation network. This network not only predicts the 2D keypoint locations from each visible patch of a target object but also provides the corresponding confidence values in an unsupervised fashion. Through the proper fusion of the most reliable local predictions, the proposed method can improve the accuracy of pose estimation when target objects are partially occluded. Experiments demonstrated that our method outperforms state-of-the-art methods on a main occlusion data set used for estimating 6D object poses. Moreover, this framework is efficient and feasible for realtime multimedia applications.

Rapid damage assessment after hurricane disasters is crucial for initiating effective emergency response actions. Unsupervised domain adaptation (UDA), which improves the classification accuracy of the target domain with leveraging abundant labeled data of the source domain, exhibits the potential to solve the issue of lacking labeled data in the task of damage building assessment. However, the application of UDA to building damage detection remains a challenge due to the complexity of post-hurricane imageries. The image characteristics of the source domain differ from those of the target domain, and the similarity of interclass samples is high, thereby degrading the transfer performance of UDA. We propose Duplex Alignment Networks with enhanced feature discrimination, which consists of a pair of generative adversarial networks (GANs) and a dedicated classifier, to address this issue. Our approach can be highlighted in two aspects: 1) a pair of GANs are used to eliminate the discrepancy between the source and the target domains by the alignments at the feature and pixel levels and 2) a dedicated classifier is integrated into the second discriminator to improve the discrimination of the extracted features of the target domain, which can alleviate performance deterioration caused by the high similarity of interclass samples. The proposed approach is validated by two challenging transfer tasks by using the data sets of Hurricanes Sandy, Maria, and Irma. Average accuracy rates of 85.1% (Hurricane Sandy → Maria) and 98.3% (Hurricane Sandy → Irma) are achieved, thereby leading to superior performance compared with the state-of-the-art methods.

Whether a second ketamine infusion in the first week improves the antidepressant, antisuicidal, and anti-inflammatory effects of the first low-dose ketamine infusion remains unclear.A total of 78 patients with medication-resistant depression were allocated to receive two ketamine infusions (n = 30; days 1 and 4), a single ketamine infusion (n = 24; only day 1), or normal saline placebo infusion (n = 24; only day 1). The Montgomery-Asberg Depression Scale (MADRS) and 17-item Hamilton Rating Scale for Depression (HDRS) were administered before and at 40 min, 240 min, day 2, day 4, day 5, and day 7 after infusion. Serum concentrations of interleukin (IL)-2 and tumor necrosis factor (TNF)-α were assessed.Two ketamine infusions improved the overall depressive symptoms (p 0.001) and melancholic symptoms (p 0.001) than a single ketamine or placebo infusion. The antisuicidal effect did not differ between the ketamine treatment groups. Two ketamine infusions increased TNF-α levels compared with a single ketamine or placebo infusion (p = 0.015). A single ketamine infusion improved the TNF-α-to-IL-2 ratio, an index of average anti-inflammatory effect, than two ketamine infusions or a single placebo infusion (p = 0.027).Repeated low-dose ketamine infusions improved the antidepressant effect, but not the antisuicidal effect, compared with a single infusion. However, repeated ketamine infusions may exert a lesser anti-inflammatory effect than a single infusion.

Purpose: To report the surgical outcomes of antegrade balloon dacryocystoplasty combined with pushed-type monocanalicular intubation as a primary surgical treatment in patients with congenital nasolacrimal duct obstruction. Methods: A retrospective cohort study was conducted at National Cheng Kung University Hospital. The medical records of all patients with congenital nasolacrimal duct obstruction who underwent antegrade balloon dacryocystoplasty followed by pushed-type monocanalicular intubation as the primary surgical treatment from January 2018 to July 2020 were included. The monocanalicular tube was removed 1 to 2 weeks after intubation. Surgical success was defined as resolved epiphora 1 month after the tube was removed. Results: A total of 62 eyes of 48 patients were involved in this study. The mean age of the total population was 24.1 months (range: 12 to 66 months). The mean duration of the tube indwelling in the nasolacrimal duct was 9.5 days (range: 4 to 15 days). A total of 60 of the 62 eyes (96.77%) reported surgical success. Early tube loss occurred in 2 eyes (3.23%); however, epiphora was not reported afterward. No complications other than tube loss were recorded. There were no recurrences in patients who had undergone successful surgery observed up to July 2020. Conclusions: Antegrade balloon dacryocystoplasty with short-term pushed-type monocanalicular intubation as a primary surgical treatment for congenital nasolacrimal duct obstruction may have high potential with a high success rate and a low complication rate. [ J Pediatr Ophthalmol Strabismus . 2021;58(6):365–369.]

BACKGROUND Acute appendicitis is one of the most common acute abdominal emergencies around the world, which is always associated with infection. Infection with Salmonella typhi, an enteric pathogen, is a rare cause of acute appendicitis. We here report a patient with acute appendicitis associated with Samonella typhi infection, accompanied with spleen and kidney infarction, providing a rare example for a common surgical emergency. CASE SUMMARY A 25-year-old Pakistani man presented to the hospital with a 3-d history of fevers, vomiting, and abdominal pain. Computed tomography (CT) revealed a thickened intestinal wall of the ileocecal junction with multiple enlarged lymph nodes nearby. He was diagnosed with acute appendicitis and received laparoscopic appendectomy, which showed mild inflammation of the appendix. After the surgery, the patient presented again with a high fever (> 39 °C) and diarrhea. A CT angiography scan indicated spleen and kidney infarction. According to the blood culture, the diagnosis was finally clear to be Samonella typhi infection. The pyrexia and enteric symptoms were relieved after the application of intravenous levofloxacin. CONCLUSION This case, characterized by the combination of Salmonella typhi infection, acute appendicitis, and renal and splenic infraction, serves as a rare example for a common surgical emergency.

Background Microelectrode recording (MER)-guided deep brain stimulation (DBS) remains the standard electrophysiological procedure to place the DBS lead at the optimal target. When single-track MER or test stimulation yields suboptimal results, trajectory adjustments are needed. Intraoperative computed tomography (iCT) can be useful to visualize the microelectrode and verify possible adjustments. The aim of this study was to evaluate the effect of iCT in MER during frameless stereotactic DBS for Parkinson disease (PD). Methods We retrospectively collected 28 PD patients, of whom 19 received iCT and 9 did not, and measured intracranial volume, cerebral volume, cerebrospinal fluid volume, and pneumocephalus volume. Euclidean distance was assessed according to merged preoperative brain CT and magnetic resonance imaging and postoperative brain CT. Results Fifty-six hemispheres in the 28 patients were analyzed for MER tracks. The patients who received iCT had a significantly lower mean number of MER tracks (1.6 vs. 2.6, P = 0.013) and lower mean Euclidean distance (2.2 mm vs. 2.7 mm, P = 0.033) compared with those who did not receive iCT. Although there was a trend of a decrease in pneumocephalus using intraoperative imaging, there was no significant difference in surgical time. Conclusions iCT can reduce the number of MER tracks and increase surgical accuracy. Further studies are warranted to investigate whether iCT can reduce surgical complications and improve surgical outcomes.

Many studies have revealed that statin therapy reduced mortality in cancer patients, especially in breast cancer, but the effect for second cancer was unclear. We, therefore, performed a comparable cohort study to determine the risk of second cancer in breast cancer patients with statin therapy. Using claims data from Taiwan’s National Health Insurance Program, this study enrolled newly diagnosed breast cancer patients from 2000 to 2007 with and without statin therapy as the statin (n = 1222) and nonstatin (n = 4888) cohorts, respectively. The nonstatin cohort was propensity score matched by cohort entry year, age, and randomly selected comorbidities. These two cohorts were followed up until the diagnosis of second cancer, death, or the end of 2011. Cox proportional hazard models were used to estimate the hazard ratios. The statin cohort had a lower incidence rate than the nonstatin cohort for second cancer (7.37 vs. 8.36 per 1000 person-years), although the difference was not significant (adjusted hazard ratio [aHR] 0.90, 95% confidence interval [CI] 0.65–1.26). Compared with the nonstatin cohort, the second cancer risk was significantly higher for patients taking pravastatin (aHR 2.71, 95% CI 1.19–6.19) but lower for those receiving multiple statin treatment (aHR 0.45, 95% CI 0.25–0.81) and combined lipophilic and hydrophilic type of statin (aHR 0.42, 95% CI 0.20–0.89). The risk was lower for patients receiving a cumulative defined daily dose (cDDD) of > 430 (aHR 0.41, 95% CI 0.19–0.86). This study showed that there is little association between statin use and second cancer risk in breast cancer patients.

In this paper, the effects of two key process parameters of the selective laser melting process, namely laser power and scanning speed, on the single-track morphologies and the bead characteristics, especially the depth-to-width D/W and height-to-width H/W ratios, were investigated using both experimental and Machine Learning (ML) approaches. A total of 840 single tracks were fabricated with several combinations of laser power and scanning speed levels. Surface morphologies of the single tracks and bead profiles were thoroughly investigated, providing a track-type map and the evolutions of the bead characteristics as a function of laser power and scanning speed. The results indicate neither severe balling nor keyholing effect for all combinations of laser power and scanning speed. Besides, simple relationships cannot accurately describe the evolutions of the D/W and H/W ratios as a function of laser power and scanning speed. Two Machine Learning-based regression models, Random Forest and Artificial Neural Network, were chosen to estimate the D/W and H/W ratios using laser power and scanning speed. The Bayesian optimization algorithm was employed to optimize the model hyperparameter selection. Both Machine Learning-based models appear to be able to predict reasonably well the two aspect ratios, D/W and H/W, with an overall R2 value reaching about 90%, evaluated on the cross-validation dataset after a few seconds of training time, respectively.

China’s rapid economic development has caused considerable environmental problems in waste generation and treatment. One important reason for this is China’s manufacturing exports to other countrie...

The rapid development of the Internet of Things has led to demand for high-speed data transformation. Serving this purpose is the Tactile Internet, which facilitates data transfer in extra-low latency. In particular, a Tactile Internet based on software-defined networking (SDN) has been broadly deployed because of the proven benefits of SDN in flexible and programmable network management. However, the vulnerabilities of SDN also threaten the security of the Tactile Internet. Specifically, an SDN controller relies on the network status (provided by the underlying switches) to make network decisions, e.g., calculating a routing path to deliver data in the Tactile Internet. Hence, the attackers can compromise the switches to jeopardize the SDN and further attack Tactile Internet systems. For example, an attacker can compromise switches to launch distributed denial-of-service attacks to overwhelm the SDN controller, which will disrupt all the applications in the Tactile Internet. In pursuit of a more secure Tactile Internet, the problem of abnormal SDN switches in the Tactile Internet is analyzed in this article, including the cause of abnormal switches and their influences on different network layers. Then we propose an approach that leverages the messages sent by all switches to identify abnormal switches, which adopts a linear structure to store historical messages at a relatively low cost. By mapping each flow message to the flow establishment model, our method can effectively identify malicious SDN switches in the Tactile Internet and thus enhance its security.

Background and Purpose: No data exists on whether proportional recovery (PR) is associated with health-related quality of life (HRQOL) domains. We evaluated whether PR was associated with domain-specific HRQOL scores at 3 months after ischemic stroke. Methods: This prospective cohort study enrolled patients with ischemic stroke between January 2017 and June 2018. Impaired strength was assessed using the Fugl-Meyer Upper Extremity (range, 0–66 points) and Motricity Index (range, 0–100 points) during index hospitalization and 3 months. Both measures are well-validated and reliable in patients with stroke to assesses motor functioning. PR (defined as 70% of difference between initial score and maximum possible recovery) was calculated from the initial measurements. HRQOL was measured using Neuro-QOL domains: upper extremity, depression, and cognition domains. PR was evaluated with HRQOL domains using binomial logistic regression. Results: Final analysis included 84 patients (mean age 67.8±16.4 years; 44% male; 51.2% White). For both Fugl-Meyer Upper Extremity and Motricity Index, the PR threshold was met for 48.8% of patients. Failure to meet Motricity Index PR was only associated with increased odds of HRQOL depression impairment (adjusted odds ratio, 11.8 [95% CI, 1.23–112.7]). Failure to meet Fugl-Meyer Upper Extremity PR threshold was not associated with HRQOL impairment after adjustment. Conclusions: Our findings suggest that reaching the PR threshold provides poor discrimination of HRQOL. Despite not meeting expected PR thresholds, patients can still maintain un-impaired HRQOL, suggesting other factors play a role in preserved HRQOL.

Ultrasound (US) technology, with major advances and new developments, has become an essential and first-line imaging modality for clinical diagnosis and interventional treatment. US imaging has evolved from one-dimensional, two-dimensional to three-dimensional display, and from static to real-time imaging, as well as from structural to functional imaging. Based on its portability and advanced digital imaging technique, US was first adopted by emergency medicine in the 1980s and gradually gained popularity among other specialists for clinical diagnosis and interventional treatment. Point-of-Care Ultrasound (POCUS) was then proposed as a new concept and developed for new uses, which greatly extended clinical US applications. Nowadays, artificial intelligence (AI), cloud computing, 5G network, robotics, and remote technologies are starting to be integrated into US equipment. US systems have gradually evolved to an intelligent terminal platform with powerful imaging and communication tools. In addition, specialized US machines tend to be more suitable and important to meet increasing demands and requirements by various clinical specialties and departments. In this article, we review current US technology and POCUS as new concepts and its future trends, as well as related technological developments and clinical applications.

By enhancing vagal activity, auricle transcutaneous electric nerve stimulation (TENS) is developed as a non-invasive therapy for heart failure. Nevertheless, though shoulder TENS used for treating adhesive capsulitis could affect vagal tone, its potential impact on heart functions remains unclear. In this study, electrocardiogram (ECG) and heart rate (HR) of subjects in response to sham, right-sided, or left-sided shoulder TENS (TENS-S, TENS-R, and TENS-L, respectively; 5 min) were recorded and analyzed. During the stimulation period, TENS-R constantly and TENS-L transiently decreased the HR of subjects; both TENS-R and TENS-L increased powers of the low- and high-frequency spectra. While TENS-R exhibiting no effect, TENS-L increased the ratio of low/high-frequency power spectrum indicating TENS-R decreased the HR through potentiating cardiac vagal tone. Collectively, these results suggest TENS could be an early and non-invasive therapy for heart failure patients before considering implant devices or devices are not feasible; moreover, therapists/physicians need to carefully monitor the potential adverse events during treatment for patient safety.Trial registration: The study protocol was registered in ClinicalTrials.gov (NCT03982472; 11/06/2019).

With the development of Internet of Things (IoT), data are increasingly appearing at the edge of a network. Processing tasks at the network edge can effectively solve the problems of personal privacy leakage and server overloading. As a result, it has attracted a great deal of attention. A number of efficient convolutional neural network (CNN) models are proposed to do so. However, since they require much computing and memory resources, none of them can be deployed to such typical edge computing devices as Raspberry Pi 3B+ and 4B+ to meet the real-time requirements of user tasks. Considering that a traditional machine learning method can precisely locate an object with a highly acceptable calculation load, this work reviews state-of-the-art literature and then proposes a CNN with reduced input size for an object detection system that can be deployed in edge computing devices. It splits an object detection task into object positioning and classification. In particular, this work proposes a CNN model with 44 $$\times$$ 44-pixel inputs instead of much more inputs, e.g., 224 $$\times$$ 224-pixel in many existing methods, for edge computing devices with slow memory access and limited computing resources. Its overall performance has been verified via a facial expression detection system realized in Raspberry Pi 3B+ and 4B+. The work makes accurate object detection at the edge possible.

PurposeThis research attempts to evaluate the effects of personal innovativeness and the perceived value of disclosure on the hierarchical nature of privacy concerns under the contingency of self-control when using proximity Bluetooth-beacon technology (PBBT) service in proximity marketing.Design/methodology/approachThe field study takes place in areas where PBBT service is installed in Taipei, Taiwan. A quota sampling approach is used, with 401 qualified respondents participating. The data are analyzed using the partial least square method.FindingsThe results confirm the importance of personal innovativeness and perceived value of disclosure as an important determinant to influence privacy concerns about data collection. It is also found that self-control plays a negative moderating role in these two relationships. Moreover, data collection is found to be a fundamental concern leading to other privacy concern facets.Originality/valueThis research represents a pioneer work in proximity marketing regarding how privacy concerns are influenced and how privacy concerns facets are causal-related when using a PBBT platform. More detailed, conditional insight is given as the research is studied under the contingency of self-control. A set of applicable guidelines with empirical evidence is thus provided.

Higher education institutions in the United States resorted to remote instruction after the disruption caused by the coronavirus (COVID-19) pandemic. The disjointed nature of this transition of managing the academic crisis needs to be critically engaged. By examining the experiences of global communication faculty at a midsize Midwestern university, this study attempts to reflect on the circumstances, challenges, and some unexpected outcomes of the phenomenon. Based on testimonies generated from the authors’ experiences and interviews conducted with instructors, the study outlines lessons learned from the adoption of media technology and offers insights on reimagining global communication pedagogy in the post-pandemic period.

Reversible data hiding in encrypted images (RDHEI) is a technique that allows for secret data to be embedded in images while preserving the privacy of the image content when the image owner and the data hider are different entities. This paper proposes a new reversible data hiding scheme in encrypted images, in which the data hider does not need a data hiding key in the data embedding process. We first reserve embedding room before image encryption based on the congruence relations of pixel values within image blocks. Then, the pre-processed original image is encrypted using a multi-secret sharing based image encryption algorithm that satisfies additive homomorphism. Using the homomorphism of the image encryption, we embed the secret data into the reserved embedding room directly in the encrypted images without a data hiding key. The receiver can extract the secret data by checking the congruence relation between pixel values in a decrypted image block and then recover the original image. We analyze the security of the proposed scheme and embedding capacity with different block sizes and different parameters of transformation. The experimental results and analysis show that the proposed scheme achieves a high embedding capacity with a suitable block size especially in smooth images.

It is well understood that when firms receive favorable treatment from the government because of their political connections and not necessarily their economic merits, they may operate inefficiently while enjoying market advantages over their unconnected peers. However, just how firms respond to the sustained removal of their political connections has not been carefully studied. This article evaluates an unanticipated reform in China that removed government-related personnel from independent directorships of publicly listed companies. Our evidence indicates that treated firms experienced a temporary increase in their cost of debt, but invested more in R&D, imported more machinery, and became more productive and transparent. These adjustments counterbalanced the negative value effect from the financial markets when the regulation was first announced (JEL G38, P26, K20).

Objective Observational studies utilising diffusion tractography have suggested a common mechanism for tremor alleviation in deep brain stimulation for essential tremor: the decussating portion of the dentato-rubro-thalamic tract. We hypothesised that directional stimulation of the dentato-rubro-thalamic tract would result in greater tremor improvement compared to sham programming, as well as comparable improvement as more tedious standard-of-care programming. Methods A prospective, blinded crossover trial was performed to assess the feasibility, safety and outcomes of programming based solely on dentato-rubro-thalamic tract anatomy. Using magnetic resonance imaging diffusion-tractography, the dentato-rubro-thalamic tract was identified and a connectivity-based treatment setting was derived by modelling a volume of tissue activated using directional current steering oriented towards the dentato-rubro-thalamic tract centre. A sham setting was created at approximately 180° opposite the connectivity-based treatment. Standard-of-care programming at 3 months was compared to connectivity-based treatment and sham settings that were blinded to the programmer. The primary outcome measure was percentage improvement in the Fahn–Tolosa–Marín tremor rating score compared to the preoperative baseline. Results Among the six patients, tremor rating scores differed significantly among the three experimental conditions ( P=0.030). The mean tremor rating score improvement was greater with the connectivity-based treatment settings (64.6% ± 14.3%) than with sham (44.8% ± 18.6%; P=0.031) and standard-of-care programming (50.7% ± 19.2%; P=0.062). The distance between the centre of the dentato-rubro-thalamic tract and the volume of tissue activated inversely correlated with the percentage improvement in the tremor rating score (R2=0.24; P=0.04). No significant adverse events were encountered. Conclusions Using a blinded, crossover trial design, we have shown the technical feasibility, safety and potential efficacy of connectivity-based stimulation settings in deep brain stimulation for treatment of essential tremor.

Background/purpose To investigate nerve excitability changes in patients with fibromyalgia and the correlation with clinical severity. Methods We enrolled 20 subjects with fibromyalgia and 22 sex and age-matched healthy subjects to receive nerve excitability test and nerve conduction study to evaluate the peripheral axonal function. Results In the fibromyalgia cohort, the sensory axonal excitability test revealed increased superexcitability (%) (P = 0.029) compared to healthy control. Correlational study showed a negative correlation between increased subexcitability (%) (r = −0.534, P = 0.022) with fibromyalgia impact questionnaire (FIQ) score. Computer modeling confirmed that the sensory axon excitability pattern we observed in fibromyalgia cohort was best explained by increased Barrett–Barrett conductance, which was thought to be attributed to paranodal fast K+ channel dysfunction. Conclusion The present study revealed that paranodal sensory K+ conductance was altered in patients with fibromyalgia. The altered conductance indicated dysfunction of paranodal fast K+ channels, which is known to be associated with the generation of pain.

Contemporary cloud storage and social media networks have matured and; as a result, the privacy protection of image content in these applications have attracted more attention. Reversible data hiding in encrypted i mages (RDHEI) is an efficient technique to embed additional data such as timestamps, watermarks, and copyright information into encrypted images to protect the image content. This paper proposes a reversible data hiding scheme in encrypted images based on a block-based adaptive MSB e ncoding technique (BAME-RDHEI). The content owner first uses a block-based image encryption method , including block permutation , pixel permutation in each block, and the block-based bitwise exclusive-or operation, to encrypt an image. This specific image encryption method preserves the relevance of the MSB bit planes in each block, and the block-based adaptive MSB encoding technique can be applied to classify and encode eight block types according to the number of MSB bit planes where all the values are all ‘1’ or ‘0’. After embedding the indicators that are generated by Huffman coding into the first MSB bit plane of each block, the remaining values of these MSB bit planes can be vacated for data embedding. The receiver can extract additional data, decrypted the image or recover the original image according to the different keys. The experimental results demonstrate that our proposed scheme provides enhanced security, significantly improves the embedding rate, and obtains higher visual quality for the decrypted image compared to other state-of-the-art schemes.

Background Retrospective studies have suggested that patients with dementia have higher prevalence of atopic dermatitis (AD) than those without dementia. However, the temporal association of AD with subsequent dementia remains unknown. Objective To assess the temporal association of AD with subsequent dementia. Methods We included data of patients with AD aged 45 years and older (n = 1059) and 1:10 age, sex, residence, income, and dementia-related comorbidity–matched controls (n = 10,590) from the Taiwan National Health Insurance Research Database and reviewed their subsequent dementia development from the enrollment date to the end of 2013. Results After adjustments for dementia-related comorbidities, patients with AD were found to be more likely to develop any dementia (hazard ratio [HR], 2.02; 95% confidence interval [CI], 1.24-3.29), particularly Alzheimer's disease (HR, 3.74; 95% CI, 1.17-11.97), during the follow-up period than those in the control group. Moderate-to-severe AD was associated with a high subsequent dementia risk (HR, 4.64; 95% CI, 2.58-8.33). Sensitivity analyses with the exclusion of the first 3 (HR, 2.20; 95% CI, 1.28-3.80) or 5 (HR, 2.05; 95% CI, 1.08-3.89) years of observation revealed consistent findings. Conclusion AD may be an independent risk factor for new-onset dementia. Clinicians may monitor the trajectory of neurocognitive function among elderly patients with AD. Additional studies elucidating the pathomechanisms between AD and subsequent dementia are warranted.

Background and Purpose: Patients with obsessive-compulsive disorder (OCD) tend to be comorbid with stroke-related risk factors, including obesity, hypertension, and diabetes. However, the temporal association between OCD and subsequent stroke risk is unclear. Methods: Using data collected between 2001 and 2010 by Taiwan’s National Health Insurance Research Database, 28 064 adult patients with OCD (International Classification of Diseases, Ninth Revision, Clinical Modification [ ICD-9-CM ] code: 300.3) and 28 064 age-, sex-, and comorbidity-matched controls were included in this study. Patients who developed ischemic ( ICD-9-CM codes: 433, 434, and 435) and hemorrhagic ( ICD-9-CM codes: 430, 431, and 432) stroke during follow-up (from enrollment to end of 2011) were identified. Moreover, medications used for treating OCD were assessed. Results: Patients with OCD (hazard ratio [HR], 3.02 [95% CI, 1.91–4.77]), especially middle-aged (HR, 2.66 [95% CI, 1.34–5.29]) and elderly adults (HR, 3.46 [95% CI, 1.70–7.05]), had an elevated risk of developing ischemic stroke during the follow-up period compared with non-OCD controls. The cumulative HR of hemorrhagic stroke did not differ (HR, 0.87 [95% CI, 0.42–1.80]) between the OCD and non-OCD groups. In patients with OCD, both short- (HR, 1.69 [95% CI, 0.74–3.88]; HR, 0.31 [95% CI, 0.05–1.95]) and long-term use (HR, 1.37 [95% CI, 0.60–3.16]; HR, 0.90 [95% CI, 0.22–3.76]) of OCD medications were not correlated with ischemic and hemorrhagic stroke compared with nonuse. Conclusions: Clinicians should closely monitor cerebrovascular disease and related risks in patients with OCD. The pathomechanism of OCD with an increased risk of ischemic stroke warrants further investigation.

The telegraph was introduced to China in the late 19th century, a time when China also saw the rise of modern banks. Based on this historical context, this paper documents the importance of information technology in banking development. We construct a data set on the distributions of telegraph stations and banks across 287 prefectures between 1881 and 1936. The results show that the telegraph significantly expanded banks’ branch networks in terms of both number and geographic scope. The effect of the telegraph remains robust when we instrument it using proximity to the early military telegraph trunk.

Purpose Exploration of the effect of chronic recurrent seizures in focal epilepsy on brain volumes has produced many conflicting reports. To determine differences in brain structure in temporal lobe epilepsy (TLE) and extratemporal epilepsy (using frontal lobe epilepsy (FLE) a surrogate) further, we performed a retrospective analysis of a large cohort of patients with seizure-onset zone proven by intracranial monitoring. Methods A total of 120 TLE patients, 86 FLE patients, and 54 healthy controls were enrolled in this study. An analysis of variance of voxel-based morphometry (VBM) was used to seek morphometric brain differences among TLE patients, FLE patients, and healthy controls. Additionally, a vertex-based surface analysis was utilized to analyze the hippocampus and thalamus. Significant side-specific differences in hippocampal gray matter volume were present between the left TLE (LTLE), right TLE RTLE (RTLE), and control groups ( pResults Vertex analyses revealed significant volume reduction in inferior parts of the left hippocampus in the LTLE group and lateral parts of the right hippocampus in the RTLE group compared to controls ( pConclusion Our results highlight the variation in morphometric lateralized changes in the brain between different epilepsy onset zones, providing critical insight into the natural history of people with drug-resistant focal epilepsies.

BACKGROUND Follicular lymphoma (FL) is more common in lymph nodes, while primary extranodal lymphomas are rare. Urinary tract lymphoid neoplasms are extremely rare, accounting for less than 5% of all extranodal lymphomas. Only one case of FL from the renal pelvis has previously been reported. CASE SUMMARY A 70-year-old male patient with a history of esophageal cancer visited our hospital for follow-up examination. Abdominal computed tomography revealed a malignant mass in the right renal pelvis. The whole-body positron emission tomography/computed tomography showed a significant increase in fluorodeoxyglucose uptake of this soft tissue mass and no abnormal fluorodeoxyglucose uptake in the esophageal wall. The patient underwent radical resection of a malignant tumor in the right kidney, which was confirmed by postoperative pathology to be FL. The patient received no radiation or chemotherapy after surgery, and no recurrence of lymphoma or other malignant tumors was found at the 1-year follow-up. CONCLUSION Extranodal FL is more common in the skin and gastrointestinal tract but rarely occurs in the urinary tract. This is the second report of primary renal FL. Localized extranodal FL is expected to have a favorable prognosis and can be cured by local resection.

The contemporary adoption of cloud storage and social media networks has increased the need for privacy protection of image content. A key area of research utilizes reversible data hiding in encrypted images (RDHEI) to embed additional data into an encrypted image while protecting image content from disclosure. This paper proposes an RDHEI scheme based on prediction error and block classification (PB-RDHEI). Different from the conventional schemes, we combine a prediction error matrix and the bit planes block classification technique to create a new level for an encrypted image. For each block, some MSB bit planes can be used for additional data hiding to efficiently improve upon the available reserved room. Receivers can separately extract the complete additional data or recover the original image. The experimental results show that our proposed scheme significantly improves the embedding rate and can recover the original image losslessly.

Due to the popularity of smartphones, laptops and wireless communication technologies, compressed images have become widely used for such applications, because a smaller storage space is needed. Protecting the security of information transmission over the Internet while reducing file size (also called the size of the final coding stream) has become a critical issue. Three related works have proposed using the reversible data hiding (RDH) method on AMBTC-compressed images to reduce the final stego-coding stream. However, these approaches still have room for improvement. To further optimize the performance of the related works, in this paper, we present an efficient dual prediction–based reversible data hiding and reduced code method for AMBTC-compressed images that uses three significant techniques: the dual prediction strategy (DP), the reversible standardized parameters (RSP) and the improved error division technique (IED). We can thus embed more secret data while effectively reducing the length of the final coding stream. Experimental results show that the proposed method has both the lowest true bitrate and the highest embedding efficiency in broader usage scenarios when compared to other related methods.

A pixel-based fragile image watermarking method based on s absolute moment block truncation coding (AMBTC) is proposed in this paper. To enhance the readability of the tampered image, we apply pixel concept to design pixel-based tampering detection and content recovery mechanisms. In the proposed scheme, recovery information is derived from the AMBTC compression codes of the original image, and authentication codes are generated by a pseudo random generator (PRG). Both data are then discretely embedded into the two least significant bits (LSBs) of pixels in the original image with turtle shell-based data hiding method. Therefore, each non-overlapped pixel pair in a watermarked image has 2 bits of recovery information and 2 bits of authentication code. When the recipient suspects that the received image may have been tampered with, the hidden authentication code can be used to locate tampered pixels, and then the hidden recovery information can be extracted and then used to restore the tampered pixels.

Drug–target interaction prediction is important for drug development and drug repurposing. Many computational methods have been proposed for drug–target interaction prediction due to their potential to the time and cost reduction. In this review, we introduce the molecular docking and machine learning-based methods, which have been widely applied to drug–target interaction prediction. Particularly, machine learning-based methods are divided into different types according to the data processing form and task type. For each type of method, we provide a specific description and propose some solutions to improve its capability. The knowledge of heterogeneous network and learning to rank are also summarized in this review. As far as we know, this is the first comprehensive review that summarizes the knowledge of heterogeneous network and learning to rank in the drug–target interaction prediction. Moreover, we propose three aspects that can be explored in depth for future research.

Objective: In stroke patients, extensive interventions for incidental thyroid nodules can be burdensome and costly. It appears that the risk of malignancy has not been reported in angiographically detected nodules and outcomes have not yet been described in patients, receiving acute stroke work-up. Materials and Methods: Incidental thyroid nodules were found on neck computerized tomography angiography or magnetic resonance angiography performed during inpatient stroke workup (January 2017 to September 2019). These patient cases were reviewed based on sonography reports, diagnosis, and follow-up care. Results: Of the 13 563 patients, 192 had incidental thyroid nodules (prevalence 1.4%). Twenty-six died from comorbidities and 22 received thyroid sonography. Twelve nodules from 10 patients had sonographic characteristics for biopsy: 10 benign, 1 indeterminate, and 1 papillary thyroid cancer (risk of malignancy: 8%). The cancer patient underwent hemithyroidectomy and is disease-free. Follow-up of the remaining patients showed no worsening or suspicious nodules. The American College of Radiology (ACR) guidelines would have prevented 8 unnecessary sonograms and 1 biopsy without missing malignancy. Conclusion: Although a small risk of malignancy was noted, 95% of patients undergoing additional diagnostic thyroid testing had clinically insignificant results. The ACR guidelines can prevent unnecessary interventions. Given the 14% mortality rate in the study cohort, it is proposed that a clinical evaluation of patients is important before undergoing further diagnostics, as comorbidities may be worse than a thyroid cancer.

Due to the rapid development of the Internet of Things, cloud computing, big data, and software-defined networks technologies, vehicular ad hoc networks have evolved into software-defined Internet of Vehicles (IoV). In software-defined IoV, messages transmitted by vehicles are transferred using wireless communication technology, so attackers can easily attack specific vehicles and Road-Side Units (RSUs) within the wireless communication range. When any vehicle or RSU is under attack or control, the entire network may collapse and produce incorrect computing results, which in the worst case, can lead to car accidents. Therefore, building a highly reliable and fault-tolerant software-defined IoV is of high importance. In this paper, we visit the consensus problem in software-defined IoV. Software-defined IoV is built with infrastructures, that is, RSUs. RSUs have a backhaul capacity and a powerful computing capacity. Thus, in software-defined IoV, we will focus on how to use RSUs to assist vehicles in reaching a consensus.

Smart wristband has become a dominant device in the wearable ecosystem, providing versatile functions such as fitness tracking, mobile payment, and transport ticketing. However, the small form-factor, low-profile hardware interfaces and computational resources limit their capabilities in security checking. Many wristband devices have recently witnessed alarming vulnerabilities, e.g., personal data leakage and payment fraud, due to the lack of authentication and access control. To fill this gap, we propose a secure text pin input system, namely Taprint, which extends a virtual number pad on the back of a user's hand. Taprint builds on the key observation that the hand “landmarks”, especially finger knuckles, bear unique vibration characteristics when being tapped by the user herself. It thus uses the tapping vibrometry as biometrics to authenticate the user, while distinguishing the tapping locations. Taprint reuses the inertial measurement unit in the wristband, “overclocks” its sampling rate with the cubic spline interpolation to extrapolate fine-grained features, and further refines the features to enhance the uniqueness and reliability. Extensive experiments on 128 users demonstrate that Taprint achieves a high accuracy (96 percent) of keystrokes recognition. It can authenticate users, even through a single-tap, at extremely low error rate (2.2 percent), and under various practical usage disturbances.

This study explores how institutional and individual investors respond to analyst recommendations. Using a unique account-level trading dataset taken from the Shanghai Stock Exchange, we obtain direct evidence to show that (1) active institutional investors are significantly net buyers (net sellers) on “strong buy” and “buy” (“hold” and “sell”) recommendations; (2) active institutional investors condition their trades based on the buy-side pressure of analysts; (3) institutional investors earn abnormal returns by incorporating analysts’ buy-side pressure into their trading reactions to analyst recommendations; and (4) individual investors, in contrast, exhibit abnormal trade reactions opposite to those of active institutional investors. Our results are robust to alternative measures and different specifications. This study provides evidence that active institutional investors are more sophisticated processors of information and provides support for regulators’ concerns about the sub-optimal investment decisions made by individual investors who are unaware of the potential conflicts of interest analysts may face.

An envelope tracking supply converter was presented for the RF power amplifiers of sub-6 GHz 5G NR mobile devices. The hybrid supply converter consisted of a linear, fast switching, and slow switching regulators connected in parallel. The two-phase hysteretic control was proposed for use in the fast switching regulator to track the fast-varying envelope of 40-MHz 5G NR signal. The 0.25-μm CMOS supply converter achieved a maximum efficiency of 79.1%. The power-added efficiency improvement of the RF power amplifier with the proposed supply converter was 3.1% when tested with a 40-MHz 256-QAM 5G NR signal.

We investigate whether and to what extent shareholder litigation shapes corporate innovation by examining the staggered adoption of universal demand laws in 23 states from 1989 to 2005. These laws impose obstacles against shareholders filing derivative lawsuits, thereby significantly reducing managers’ litigation risk. Using a difference-in-differences design and a matched sample, we find that, following the passage of the laws, firms invested more in research and development, produced more patents in new technological classes and more patents based on new knowledge, generated more patents with significant impacts, and achieved higher patent value. Our findings suggest that the external pressure imposed by shareholder litigation discourages managers from engaging in explorative innovation activities. This paper was accepted by David Simchi-Levi, finance.