Search

BackgroundDuring the spread of the novel coronavirus disease (COVID-19), internet hospitals in China were engaged with epidemic prevention and control, offering epidemic-related online services and medical support to the public. ObjectiveThe aim of this study is to explore the role of internet hospitals during the prevention and control of the COVID-19 outbreak in China. MethodsOnline epidemic-related consultations from multicenter internet hospitals in China during the COVID-19 epidemic were collected. The counselees were described and classified into seven type groups. Symptoms were recorded and compared with reported patients with COVID-19. Hypochondriacal suspicion and offline visit motivation were detected within each counselees’ group to evaluate the social panic of the epidemic along with the consequent medical-seeking behaviors. The counselees’ motivation and the doctors’ recommendation for an offline visit were compared. Risk factors affecting the counselees’ tendency of hypochondriacal suspicion and offline visit motivation were explored by logistic regression models. The epidemic prevention and control measures based on internet hospitals were listed, and the corresponding effects were discussed. ResultsA total of 4913 consultations were enrolled for analysis with the median age of the counselees at 28 years (IQR 22-33 years). There were 104 (2.12%) healthy counselees, 147 (2.99%) hypochondriacal counselees, 34 (0.69%) exposed counselees, 853 (17.36%) mildly suspicious counselees, 42 (0.85%) moderately suspicious counselees, 3550 (72.26%) highly suspicious counselees, and 183 (3.72%) severely suspicious counselees. A total of 94.20% (n=4628) of counselees had epidemic-related symptoms with a distribution similar to those of COVID-19. The hypochondriacal suspicion (n=2167, 44.11%) was common. The counselees’ motivation and the doctors’ recommendation for offline visits were inconsistent (P

Amidst the rapid advancement in the application of artificial intelligence learning, questions regarding the evaluation of students' learning status and how students without relevant learning foundation on this subject can be trained to familiarize themselves in the field of artificial intelligence are important research topics. This study employed the use of a self-built AI platform (Ladder) for students to systematically learn and apply AI learning model established by the partial least squares (PLS) method to investigate the influence between variables (learning attitudes, self-regulated learning, AI anxiety, individual impact, computational thinking abilities, cognitive styles). This study was particularly conducted in the Department of Computer Science and Information Engineering of a top national university in Southern Taiwan. The valid data were collected from 65 students (55 male students; 10 female students). Furthermore, this study demonstrated the relationship between cognitive style, self-regulated learning and computational thinking. For the first time, it explored the impact of AI anxiety and completed existing research on it. The results of this study show that interest in learning positively affects learning attitudes. In addition, learning attitudes have a positive influence on each individual's performance. Based on multiple theories and the artificial intelligence learning platform, the model proposed in this study effectively understood students' learning status.

Artificial intelligence (AI) education is becoming an advanced learning trend in programming education. However, AI subjects can be difficult to understand because they require high programming skills and complex knowledge. This makes it challenging to determine how different departments of students are affected by them. This study draws on research in programming education and STEM education to explore the different factors that affect students in AI learning. Therefore, the purpose of this study is to investigate the impact of AI learning platforms on information undergraduate and non-information undergraduate by using a research model. The course was implemented for 65 students in the information undergraduate group and 39 students in the non-information undergraduate group. The findings showed that the two groups had different learning effects under different variables. Students with different cognitive styles may use different skills to positively influence self-regulated learning. This study provides important evidence to understand the learning impact of artificial intelligence among university students from different disciplines.

Abstract Magnetic resonance imaging (MRI) examinations are a routine part of the cancer treatment process. In developing countries, disease diagnosis is often time-consuming and associated with serious prognostic problems. Moreover, MRI is characterized by high noise and low resolution. This creates difficulties in automatic segmentation of the lesion region, leading to a decrease in the segmentation performance of the model. This paper proposes a deep convolutional neural network osteosarcoma image segmentation system based on noise reduction and super-resolution reconstruction, which is the first time to introduce super-resolution methods in the task of osteosarcoma MRI image segmentation, effectively improving the Model generalization performance. We first refined the initial osteosarcoma dataset using a Differential Activation Filter, separating those image data that had little effect on model training. At the same time, we carry out rough initial denoising of the image. Then, an improved information multi-distillation network based on adaptive cropping is proposed to reconstruct the original image and improve the resolution of the image. Finally, a high-resolution network is used to segment the image, and the segmentation boundary is optimized to provide a reference for doctors. Experimental results show that this algorithm has a stronger segmentation effect and anti-noise ability than existing methods. Code: https://github.com/GFF1228/NSRDN.

ContextThe prevalence of unilateral primary aldosteronism (UPA) with cortisol co-secretion varies geographically.ObjectiveTo investigate the prevalence and clinical characteristics of UPA with cortisol co-secretion in a Chinese population.DesignRetrospective cohort study.MethodsWe recruited 580 patients with UPA who underwent cosyntropin stimulation test (CST) after the 1-mg dexamethasone suppression test (DST) and retrospectively analyzed the clinical characteristics and postoperative outcomes of UPA with and without cortisol co-secretion.ResultsUPA with cortisol co-secretion (1 mg DST>1.8 ug/dL) was identified in 65 of 580 (11.2%) patients. These patients were characterized by older age, longer duration of hypertension, higher concentration of plasma aldosterone and midnight cortisol, lower adrenocorticotropic hormone (ACTH) and dehydroepiandrosterone sulfate (DHEAS), larger tumor diameter, and more history of diabetes mellitus. Cortisol and aldosterone levels were higher and DHEAS level was lower in UPA with cortisol co-secretion at 0–120 min after CST. Among 342 UPA patients with KCNJ5 gene sequencing and follow-up results, the complete clinical success rate was lower in UPA with cortisol co-secretion (33.3% vs. 56.4%, P