Class Highlight Generative Adversarial Networks for Strip Steel Defect Classification

In order to solve the problem of dataset expansion in deep learning tasks such as image classification, this paper proposed an image generation model called Class Highlight Generative Adversarial Networks (CH-GANs). In order to highlight image categories, accelerate the convergence speed of the model and generate true-to-life images with clear categories, first, the image category labels were deconvoluted and integrated into the generator through [Formula: see text] convolution. Second, a novel discriminator that cannot only judge the authenticity of the image but also the image category was designed. Finally, in order to quickly and accurately classify strip steel defects, the lightweight image classification network GhostNet was appropriately improved by modifying the number of network layers and the number of network channels, adding SE modules, etc., and was trained on the dataset expanded by CH-GAN. In the comparative experiments, the average FID of CH-GAN is 7.59; the accuracy of the improved GhostNet is 95.67% with 0.19[Formula: see text]M parameters. The experimental results prove the effectiveness and superiority of the methods proposed in this paper in the generation and classification of strip steel defect images.

Language processing reveals conceptual combination in human infants

Across three eye-tracking experiments, we taught 12-month-olds (N = 60) two novel quantity labels denoting sets of one and two (e.g., “mize” for 1; “padu” for 2). We then showed that they could not only generalize these labels to sets of previously unseen objects, but also combine them with familiar category labels acquired prior to the lab visit (e.g., “ball”, “duck”). Their eye movements revealed adult-like compositional procedures that go beyond serial processing of constituent meanings. These findings indicate that certain combinatorial processes involved in extracting complex linguistic meaning are already available by the end of the first year of life and are ready to support language comprehension.

Continuous dynamics in behavior reveal perceptual nonlinearities aid speech categorization in noise

Spoken language comprehension requires listeners map continuous features of the speech signal to discrete category labels. Categories are however malleable to surrounding context; listeners’ percept can dynamically shift depending on the sequencing of adjacent stimuli resulting in a warping of the heard phonetic category (i.e., hysteresis). Here, we investigated whether such perceptual nonlinearities—which amplify categorical hearing—might further aid speech processing in noise-degraded listening scenarios. We measured continuous dynamics in perception and category judgments of an acoustic-phonetic vowel gradient via mouse tracking. Tokens were presented in serial vs. random orders to induce more/less perceptual warping while listeners categorized continua in clean and noise conditions. Listeners’ responses were faster and their mouse trajectories closer to the ultimate behavioral selection (marked visually on the screen) in serial vs. random order, suggesting increased perceptual attraction to category exemplars. Interestingly, order effects emerged earlier and persisted later in the trial time course when categorizing speech in noise. These data describe a new functional benefit of perceptual nonlinearities to speech perception yet undocumented: warping strengthens the behavioral attraction to relevant speech categories while simultaneously assisting perception in degraded acoustic environments.

The effects of age on prototype- and exemplar-based categorization

The need to learn new concepts and categories persists through the lifespan, yet little is known about how aging affects concept learning and generalization. Here, we trained young and older adults to classify typical and boundary category members and then tested category generalization to new stimuli. During training, older adults had increased difficulty compared to young adults learning category labels for boundary items, but not typical items. At test, categorization performance that included new items at all levels of typicality was comparable across age groups, but formal categorization models indicated that older adults relied to a greater degree on generalized (prototype) category representations than young adults. These findings align with the proposal that older adults are able to form category representations based on central tendency even when they have difficulty learning and remembering individual category members. More broadly, the results contribute to our understanding of multiple categorization strategies and the limited strategy flexibility in older adults. They also highlight how reliance on

Improving the Accuracy in Classification of Blood Pressure from Photoplethysmography Using Continuous Wavelet Transform and Deep Learning

Background. Continuous wavelet transform (CWT) based scalogram can be used for photoplethysmography (PPG) signal transformation to classify blood pressure (BP) with deep learning. We aimed to investigate the determinants that can improve the accuracy of BP classification based on PPG and deep learning and establish a better algorithm for the prediction. Methods. The dataset from PhysioNet was accessed to extract raw PPG signals for testing and its corresponding BPs as category labels. The BP category of normal or abnormal followed the criteria of the 2017 American College of Cardiology/American Heart Association (ACC/AHA) Hypertension Guidelines. The PPG signals were transformed into 224 ∗ 224 ∗ 3-pixel scalogram via different CWTs and segment units. All of them are fed into different convolutional neural networks (CNN) for training and validation. The receiver-operating characteristic and loss and accuracy curves were used to evaluate and compare the performance of different methods. Results. Both wavelet type and segment length could affect the accuracy, and Cgau1 wavelet and segment-300 revealed the best performance (accuracy 90%) without obvious overfitting. This method performed better than previously reported MATLAB Morse wavelet transformed scalogram on both of our proposed CNN and CNN-GoogLeNet. Conclusions. We have established a new algorithm with high accuracy to predict BP classification from PPG via matching of CWT type and segment length, which is a promising solution for rapid prediction of BP classification from real-time processing of PPG signal on a wearable device.

The effect of spatial distance between objects on categorization level

We show that spatial distance between two objects influences how people categorize these objects. We report three (two pre-registered) experiments that show that when objects are presented close together (proximal), they are more likely to be categorized in a superordinate category than when they are presented further apart (distant). In Experiments 1A and 1B, participants provided spontaneous category labels in an open response format. In Experiment 2, we asked participants to indicate their preference for either of two category labels. We found that when objects were close together, they were categorized more often into superordinate categories than when objects were far apart (Experiments 1A and 2). Our findings demonstrate that the categorization of objects is, in part, determined by where they are in relation to other objects.

EXPRESS: Yes! I love my mother as much as myself!!! Self and mother-association effects in an Indian sample.

Information associated with the self is preferentially processed compared to others. However, cultural differences appear to exist in the way information is processed about those close to us like our mothers. In eastern compared to western cultures, information about mother seems to be processed as well as our self. However, it is not clear whether this lack of difference is due to familiarity or would extend to processing arbitrary perceptual information associated with different categorical labels. The current study employs a perceptual association paradigm in which category labels like self, mother and none are associated with arbitrary shapes to study self vs mother processing in an Indian sample. We hypothesized that there would be no difference between self and mother processing given the familial and collectivistic tendencies in India. Participants performed a matching task between shape and a pre-assigned category label, with self, mother, and none as categories in Experiment 1A and self, friend, and none as categories in Experiment 1B. Analysis of RT, accuracies and signal detection theoretic measures showed that information about mother is processed as well as self in Experiment 1A, but this effect is not present with friend in Experiment 1B. Moreover, participants’ processing for the self-associated information gets attenuated depending upon the other close person category used in the task (friend vs mother) indicating that self-information processing is dynamically dependent on the categorical contexts in which such processing takes place. Our findings have implications for understanding the processing of self-associated information across cultures and contexts.

Safety Culture: Identifying a Healthcare Organization’s Approach to Safety Event Review and Response Through the Analysis of Event Recommendations

A nonpunitive approach to safety event reporting and analysis is an important dimension of healthcare organization safety culture. A system-based safety event review process, one focused on understanding and improving the conditions in which individuals do their work, generally leads to more effective and sustainable safety solutions. On the contrary, the more typical person-based approach, that blames individuals for errors, often results in unsustainable and ineffective safety solutions, but these solutions can be faster and less resource intensive to implement. We sought to determine the frequency of system-based and person-based approaches to adverse event reviews through analysis of the recommendation text provided by a healthcare organization in response to an event report. Human factors and clinical safety science experts developed a taxonomy to describe the content of the recommendation text, reviewed 8,546 event report recommendations, and assigned one or more taxonomy category labels to each recommendation. The taxonomy categories aligned with a system-based approach, aligned with a person-based approach, did not provide an indicator of the approach, or indicated the review/analysis was pending. A total of 9,848 category labels were assigned to the 8,546 event report recommendations. The most frequently used category labels did not provide an indicator of the approach to event review (4,145 of 9,848 category labels, 42.1%), followed by a person-based approach (2,327, 23.6%), review/analysis pending (1,862 ,18.9%), and a system-based approach (1,514, 15.4%). Analyzing the data at the level of each recommendation, 23.2% (1,979 of 8,546) had at least one person-based and no system-based category, 13.3% (1,133) had at least one system-based and no person-based category, and 3% (254) had at least one person-based and one system-based category. There was variability in the event review approach based on the general event type assigned to the safety event (e.g., medication, transfusion, etc.) as well as harm severity. Results suggest improvements in applying system-based approaches are needed, especially for certain general event type categories. Recommendations for improving safety event reviews are provided.

Semi-Supervised Noisy Label Learning for Chinese Medical Named Entity Recognition

This paper describes our approach for the Chinese Medical named entity recognition(MER) task organized by the 2020 China conference on knowledge graph and semantic computing(CCKS) competition. In this task, we need to identify the entity boundary and category labels of six entities from Chinese electronic medical record(EMR). We construct a hybrid system composed of a semi-supervised noisy label learning model based on adversarial training and a rule postprocessing module. The core idea of the hybrid system is to reduce the impact of data noise by optimizing the model results. Besides, we use post-processing rules to correct three cases of redundant labeling, missing labeling, and wrong labeling in the model prediction results. Our method proposed in this paper achieved strict criteria of 0.9156 and relax criteria of 0.9660 on the final test set, ranking first.