Search

Abstract Automated coronary angiography assessment requires precise vessel segmentation, a task complicated by uneven contrast filling and background noise. Our research introduces an ensemble U-Net model, SE-RegUNet, designed to accurately segment coronary vessels using 100 labeled angiographies from angiographic images. SE-RegUNet incorporates RegNet encoders and squeeze-and-excitation blocks to enhance feature extraction. A dual-phase image preprocessing strategy further improves the model's performance, employing unsharp masking and contrast-limited adaptive histogram equalization. Following fivefold cross-validation and Ranger21 optimization, the SE-RegUNet 4GF model emerged as the most effective, evidenced by performance metrics such as a Dice score of 0.72 and an accuracy of 0.97. Its potential for real-world application is highlighted by its ability to process images at 41.6 frames per second. External validation on the DCA1 dataset demonstrated the model's consistent robustness, achieving a Dice score of 0.76 and an accuracy of 0.97. The SE-RegUNet 4GF model's precision in segmenting blood vessels in coronary angiographies showcases its remarkable efficiency and accuracy. However, further development and clinical testing are necessary before it can be routinely implemented in medical practice.

Using a deep learning algorithm in the development of a computer-aided system for colon polyp detection is effective in reducing the miss rate. This study aimed to develop a system for colon polyp detection and classification. We used a data augmentation technique and conditional GAN to generate polyp images for YOLO training to improve the polyp detection ability. After testing the model five times, a model with 300 GANs (GAN 300) achieved the highest average precision (AP) of 54.60% for SSA and 75.41% for TA. These results were better than those of the data augmentation method, which showed AP of 53.56% for SSA and 72.55% for TA. The AP, mAP, and IoU for the 300 GAN model for the HP were 80.97%, 70.07%, and 57.24%, and the data increased in comparison with the data augmentation technique by 76.98%, 67.70%, and 55.26%, respectively. We also used Gaussian blurring to simulate the blurred images during colonoscopy and then applied DeblurGAN-v2 to deblur the images. Further, we trained the dataset using YOLO to classify polyps. After using DeblurGAN-v2, the mAP increased from 25.64% to 30.74%. This method effectively improved the accuracy of polyp detection and classification.

Background: As COVID-19 has become a pandemic emerging infectious disease it is important to examine whether there was a spatiotemporal clustering phenomenon in the globe during the rapid spread after the first outbreak reported from southern China. Materials and methods: The open data on the number of COVID-19 cases reported at daily basis form the globe were used to assess the evolution of outbreaks with international air link on the same latitude and also including Taiwan. The dynamic Susceptible-Infected-Recovered model was used to evaluate continental transmission from December 2019 to March 2020 before the declaration of COVID-19 pandemic with basic reproductive number and effective reproductive number before and after containment measurements. Results: For the initial COVID-19 outbreak in China, the estimated reproductive number was reduced from 2.84 during the overwhelming outbreaks in early January to 0.43 after the strict lockdown policy. It is very surprising to find there were three countries (including South Korea, Iran, and Italy) and the Washington state of the USA on the 38° North Latitude involved with large-scale community-acquired outbreaks since the first imported COVID-19 cases from China. The propagation of continental transmission was augmented from hotspot to hotspot with higher reproductive number immediately before the declaration of pandemic. By contrast, there was not any large community-acquired outbreak in Taiwan. Conclusion: The propagated spatiotemporal transmission from China to other hotspots may explain the emerging pandemic that can only be exempted by timely border control and preparedness of containment measurements according to Taiwan experience.

Background: As Coronavirus disease 2019 (COVID-19) pandemic led to the unprecedent large-scale repeated surges of epidemics worldwide since the end of 2019, data-driven analysis to look into the duration and case load of each episode of outbreak worldwide has been motivated. Methods: Using open data repository with daily infected, recovered and death cases in the period between March 2020 and April 2021, a descriptive analysis was performed. The susceptible-exposed-infected-recovery model was used to estimate the effective productive number (Rt). The duration taken from Rt > 1 to Rt

Background: Vaccine is supposed to be the most effective means to prevent COVID-19 as it may not only save lives but also reduce productivity loss due to resuming pre-pandemic activities. Providing the results of economic evaluation for mass vaccination is of paramount importance for all stakeholders worldwide. Methods: We developed a Markov decision tree for the economic evaluation of mass vaccination against COVID-19. The effectiveness of reducing outcomes after the administration of three COVID-19 vaccines (BNT162b2 (Pfizer-BioNTech), mRNA-1273 (Moderna), and AZD1222 (Oxford-AstraZeneca)) were modelled with empirical parameters obtained from literatures. The direct cost of vaccine and COVID-19 related medical cost, the indirect cost of productivity loss due to vaccine jabs and hospitalization, and the productivity loss were accumulated given different vaccination scenarios. We reported the incremental cost-utility ratio and benefit/cost (B/C) ratio of three vaccines compared to no vaccination with a probabilistic approach. Results: Moderna and Pfizer vaccines won the greatest effectiveness among the three vaccines under consideration. After taking both direct and indirect costs into account, all of the three vaccines dominated no vaccination strategy. The results of B/C ratio show that one dollar invested in vaccine would have USD $13, USD $23, and USD $28 in return for Moderna, Pfizer, and AstraZeneca, respectively when health and education loss are considered. The corresponding figures taking value of the statistical life into account were USD $176, USD $300, and USD $443. Conclusion: Mass vaccination against COVID-19 with three current available vaccines is cost-saving for gaining more lives and less cost incurred.

Background: In most countries, large cerebral artery occlusion is identified as the leading cause of disability. In 2015, five large-scale clinical trials confirmed the benefit of intra-arterial thrombectomy. However, thrombectomy is a highly technical and facility-dependent procedure. Primary stroke centers need to transfer patients to comprehensive stroke centers to perform thrombectomy. The time-lapse during interhospital transfer would decrease the chance of the patient's proper recovery. Communication barriers also contribute to this delay.Aims: We used a smartphone application to overcome communication barriers between hospitals. We aimed to shorten the door-to-puncture time of interhospital transfer patients.Methods: We began using a smartphone application, “LINE,” to facilitate interhospital communication on May 01, 2018. We carried out retrospective data analyses for all the transfer patients (n = 351), with the primary outcome being the door-to-puncture time in our comprehensive stroke center (China Medical University Hospital). We compared the three periods: May 01 to Dec 31, 2017 (before the use of the smartphone application); May 01 to Dec 31, 2018 (the 1st year of using the smartphone application); and May 01 to Dec 31, 2019 (the 2nd year of using the smartphone application). We also compared the transfer data with non-transfer thrombectomies in the same period.Results: We compared 2017, 2018, and 2019 data. The total number of transfer patients increased over the years: 63, 113, 175, respectively. The mean door-to-puncture time decreased significantly, going from 109, through 102, to 92 min. Meanwhile, the mean door-to-puncture time in non-transfer patients were 140.3, 122.1, and 129.3 min. The main reason of time saving was the change of the way of communication, from point-to-point interhospital communication to hub-to-spoke interhospital communication.Conclusions: We used this smartphone application to enhance interhospital communication, changed from the point-to-point to hub-to-spoke method. It made us overcome the communication barrier and build up interhospital connection, thus shortening the door-to-puncture time. Our experience demonstrated the importance of close communication and teamwork in hyperacute stroke care, especially in interhospital transfer for thrombectomy.

Purpose: Approximately 5%–10% of men who receive prostate cancer radiotherapy will suffer from radiation cystitis. Bladder filling before the administration of radiotherapy results in lower radiation exposure to the bladder. BladderScan, an ultrasound-based bladder volume scanner, has the potential to evaluate bladder volume during radiotherapy; thus, a prospective pilot study was initiated. Methods: Eleven men receiving tomotherapy for localized prostate cancer were enrolled. The validity of BladderScan was evaluated by comparing the measurements from BladderScan with the calculated volume from megavoltage computed tomography (MVCT). With a crossover design to compare different methods in bladder filling, the radiotherapy was divided into 2 sequences. Conventional method: the patient was asked to drink water after voiding urine. The amount of water and the duration of waiting were the same as in the setting of the simulation. BladderScan feedback method: the bladder filling procedure depended on the BladderScan measurements. Results: There were 314 sets of data from 11 patients. The correlation coefficient between V BS and V CT was 0.87, where V BS is the mean volume of 3 measurements by BladderScan and V CT is the bladder volume derived from MVCT. The BladderScan feedback method resulted in a significant larger bladder volume than the conventional method, with a mean difference of 36.9 mL. When the failure was defined as V CT

The development of artificial synapses is inspired by the energy‐efficient recognition ability of the central nervous system of living organisms and is proposed for use as the basic building units of next‐generation neuromorphic computing networks. Herein, the perovskite/MoO3/Ag synaptic device exhibits very uniform electrical characteristics, a low operation voltage, and linear and repeatable analog switching. Important synaptic functions such as paired‐pulse facilitation (PPF) and short‐ and long‐term potentiation (STP/LTP) are all successfully and stably implemented in the device. The diffusion of Ag+ and I− ions is believed to contribute to the synaptic behavior of the device. Furthermore, the uniform and stable volatile characteristics are utilized in a proof‐of‐concept reservoir‐computing‐based simulation program using all the experimental parameters; the results strongly suggest that the neuromorphic computing that utilizes such components can simultaneously reduce the circuitry complexity and increase its recognition accuracy with noisy inputs. The findings of this work promise to significantly facilitate the utilization of highly homogeneous vacuum‐deposited perovskite/metal oxide structures in high‐performance neuromorphic computing systems.

Music can generate a positive effect in runners’ performance and motivation. However, the practical implementation of music intervention during exercise is mostly absent from the literature. Therefore, this paper designs a playback sequence system for joggers by considering music emotion and physiological signals. This playback sequence is implemented by a music selection module that combines artificial intelligence techniques with physiological data and emotional music. In order to make the system operate for a long time, this paper improves the model and selection music module to achieve lower energy consumption. The proposed model obtains fewer FLOPs and parameters by using logarithm scaled Mel-spectrogram as input features. The accuracy, computational complexity, trainable parameters, and inference time are evaluated on the Bi-modal, 4Q emotion, and Soundtrack datasets. The experimental results show that the proposed model is better than that of Sarkar et al. and achieves competitive performance on Bi-modal (84.91%), 4Q emotion (92.04%), and Soundtrack (87.24%) datasets. More specifically, the proposed model reduces the computational complexity and inference time while maintaining the classification accuracy, compared to other models. Moreover, the size of the proposed model for network training is small, which can be applied to mobiles and other devices with limited computing resources. This study designed the overall playback sequence system by considering the relationship between music emotion and physiological situation during exercise. The playback sequence system can be adopted directly during exercise to improve users’ exercise efficiency.

Swing movements and muscle strength are essential for mastering badminton techniques. Traditionally, students learn badminton through their instructor's physical demonstration, verbal instructions, and small group activities. To enhance students' learning experience and assist badminton instructors more effectively, this study proposes an alternative teaching approach that utilizes a Badminton Teaching-Assisted System (BTAS) to teach swing movements. The system uses a gesture control armband to collect electromyography and gyroscope signals from professional players, which are then used to build a machine learning model based on the back propagation neural network. The model can be used as a scoring tool to assess the correctness of the student's swing movements and muscle strength. A total of 87 students participated in an eleven-week experiment focusing on two fundamental badminton skills: The backhand push and backhand lift. The results demonstrated a significant improvement in the correctness of the muscle strength during swing movements for the experimental group, compared to the control group. The BTAS can help to enhance the effectiveness of badminton instruction, thereby improving students' swing techniques. Future research could expand upon this study to develop more precise artificial intelligence-based sports assistance systems, which would facilitate personalized teaching and learning in physical education.

Social media or social networking sites have been used to support online learning with good interactive features. If an existing system can retain current users and attract new users, it can provide greater benefits and influence in the field of online learning. However, most previous studies focus on learner participation intention, and rarely note their switching intention. Therefore, this study attempts to analyze and discuss the learner switching intention from the perspective of Push-Pull-Mooring Model. A research framework was proposed for examining the switching intention of learners in social network-based learning platforms. A total of 371 valid samples were used to examine the research framework using the partial least squares approach. The results show that switching intention is significantly affected by the push effects (social interaction and service quality), mooring effects (switching costs and prior switching experience) and pull effects (attractiveness of new services and social effect). To retain and attract users, platform providers can raise switching costs through improving service quality and system capabilities and educators can develop appropriate approaches to enhance social interaction and social effects.

The aim of this study was to develop a badminton teaching system using the wearable technology for improving badminton teaching and learning. This system can provide the similarity scores automatically by comparing a student's movement and strength while playing badminton with a well-trained expert model. A quasi-experiment was conducted for eight weeks to evaluate the effectiveness of the developed system. A paired-sample t-test was analyzed to evaluate whether there is a significant difference between the pretest and posttest for the experimental group (EG) and the control group (CG) respectively. Also, an independent sample t test was utilized to compare the posttests between EG and CG. The results show that the badminton teaching system developed in this study could enhance the learning effectiveness of novice badminton learners. The results not only show that applying wearable technology, the Myo armband, can benefit learners' badminton training, but also can demonstrate the importance of providing instant and adaptive feedback in motor skill learning.

This research demonstrates the design of a Joyful Classroom Learning System (JCLS) with flexible, mobile and joyful features. The theoretical foundations of this research include the experiential learning theory, constructivist learning theory and joyful learning. The developed JCLS consists of the robot learning companion (RLC), sensing input device, mobile computation unit, mobile display device, wireless local network and operating software. The aim of this research is to design and evaluate the JCLS, which is implemented by using robot and RFID technologies. The developed JCLS system has been applied in real world for supporting children to learn mathematical multiplication. Both pilot experiment and formal experiment were conducted and the results showed that the JCLS can provide learners with more opportunities for hands-on exercises and deepening their impressions about the learning contents. Having many opportunities for hands-on exercises, learners can have more thinking time for knowledge construction. Using robot to design RLC can simultaneously increase learners' motivations and offer a more joyful perception to learners during the learning process. On the other hand, the JCLS can support instructors to immediately acquire the learning statuses of every learner for adjusting his/her in-class instructional strategy and giving after-school assistances. (Contains 7 tables and 8 figures.)

English vocabulary is the foundation of English learning. According to information processing theory, recitation helps learners remember and understand English terminology. However, this type of exercise may lead to boredom, anxiety, difficulty concentrating, and other negative effects. This study proposes the integration of competitive gaming with personalized assistance as a means of reducing anxiety and enhancing the degree of immersion, as well as raising the overall effectiveness of the lesson. The board game Monopoly was used as the basis for the design of a vocabulary-focused competitive game. Personalized assistance was provided to help learners study and review new English vocabulary words in an attempt to improve their memorization. A total of 120 students were recruited to serve as participants in the study. Each was randomly assigned to one of four different groups to take part in select English vocabulary activities. The results show that the integration of a competitive gaming scenario with a personalized assistance strategy helped students improve their vocabulary acquisition.

Clinical resources in nursing schools are always insufficient for satisfying the practice requirements of each student at the same time during a formal course session. Although several studies have applied information and communication technology to develop computer-based learning tools for addressing this problem, most of these developments lack appropriate interactive and diagnostic services to assist students in bridging theory-practice gaps and realizing individual flaws in their nursing skills. Therefore, this study aims to develop an interactive clinical learning and diagnosis system that provides a web-based interactive learning environment and personalized diagnostic service for nursing education. An experiment was conducted in a foundations of nursing course at a nursing college in Taiwan. The experimental results show that the proposed approach is helpful to students in improving their learning motivation, performance, and attitude. Moreover, it was also found that most students showed positive perceptions toward the usage of the proposed system.