Advances in natural language processing

Natural language processing and machine learning to assist radiation oncology incident learning

PURPOSE

To develop a Natural Language Processing (NLP) and Machine Learning (ML) pipeline that can be integrated into an Incident Learning System (ILS) to assist radiation oncology incident learning by semi-automating incident classification. Our goal was to develop ML models that can generate label recommendations, arranged according to their likelihoods, for three data elements in Canadian NSIR-RT taxonomy.

METHODS

Over 6000 incident reports were gathered from the Canadian national ILS as well as our local ILS database. Incident descriptions from these reports were processed using various NLP techniques. The processed data with the expert-generated labels were used to train and evaluate over 500 multi-output ML algorithms. The top three models were identified and tuned for each of three different taxonomy data elements, namely: (1) process step where the incident occurred, (2) problem type of the incident and (3) the contributing factors of the incident. The best-performing model after tuning was identified for each data element and tested on unseen data.

RESULTS

The MultiOutputRegressor extended Linear SVR models performed best on the three data elements. On testing, our models ranked the most appropriate label 1.48 ± 0.03, 1.73 ± 0.05 and 2.66 ± 0.08 for process-step, problem-type and contributing factors respectively.

CONCLUSIONS

We developed NLP-ML models that can perform incident classification. These models will be integrated into our ILS to generate a drop-down menu. This semi-automated feature has the potential to improve the usability, accuracy and efficiency of our radiation oncology ILS.

Machine Learning in Modeling of Mouse Behavior

Mouse behavior is a primary outcome in evaluations of therapeutic efficacy. Exhaustive, continuous, multiparametric behavioral phenotyping is a valuable tool for understanding the pathophysiological status of mouse brain diseases. Automated home cage behavior analysis produces highly granulated data both in terms of number of features and sampling frequency. Previously, we demonstrated several ways to reduce feature dimensionality. In this study, we propose novel approaches for analyzing 33-Hz data generated by CleverSys software. We hypothesized that behavioral patterns within short time windows are reflective of physiological state, and that computer modeling of mouse behavioral routines can serve as a predictive tool in classification tasks. To remove bias due to researcher decisions, our data flow is indifferent to the quality, value, and importance of any given feature in isolation. To classify day and night behavior, as an example application, we developed a data preprocessing flow and utilized logistic regression (LG), support vector machines (SVM), random forest (RF), and one-dimensional convolutional neural networks paired with long short-term memory deep neural networks (1DConvBiLSTM). We determined that a 5-min video clip is sufficient to classify mouse behavior with high accuracy. LG, SVM, and RF performed similarly, predicting mouse behavior with 85% accuracy, and combining the three algorithms in an ensemble procedure increased accuracy to 90%. The best performance was achieved by combining the 1DConv and BiLSTM algorithms yielding 96% accuracy. Our findings demonstrate that computer modeling of the home-cage ethome can clearly define mouse physiological state. Furthermore, we showed that continuous behavioral data can be analyzed using approaches similar to natural language processing. These data provide proof of concept for future research in diagnostics of complex pathophysiological changes that are accompanied by changes in behavioral profile.

Development of machine learning and natural language processing algorithms for preoperative prediction and automated identification of intraoperative vascular injury in anterior lumbar spine surgery

BACKGROUND

Intraoperative vascular injury (VI) may be an unavoidable complication of anterior lumbar spine surgery; however, vascular injury has implications for quality and safety reporting as this intraoperative complication may result in serious bleeding, thrombosis, and postoperative stricture.

PURPOSE

The purpose of this study was to (1) develop machine learning algorithms for preoperative prediction of VI and (2) develop natural language processing (NLP) algorithms for automated surveillance of intraoperative VI from free-text operative notes.

PATIENT SAMPLE

Adult patients, 18 years or age or older, undergoing anterior lumbar spine surgery at two academic and three community medical centers were included in this analysis.

OUTCOME MEASURES

The primary outcome was unintended VI during anterior lumbar spine surgery.

METHODS

Manual review of free-text operative notes was used to identify patients who had unintended VI. The available population was split into training and testing cohorts. Five machine learning algorithms were developed for preoperative prediction of VI. An NLP algorithm was trained for automated detection of intraoperative VI from free-text operative notes. Performance of the NLP algorithm was compared to current procedural terminology and international classification of diseases codes.

RESULTS

In all, 1035 patients underwent anterior lumbar spine surgery and the rate of intraoperative VI was 7.2% (n=75). Variables used for preoperative prediction of VI were age, male sex, body mass index, diabetes, L4-L5 exposure, and surgery for infection (discitis, osteomyelitis). The best performing machine learning algorithm achieved c-statistic of 0.73 for preoperative prediction of VI (https://sorg-apps.shinyapps.io/lumbar_vascular_injury/). For automated detection of intraoperative VI from free-text notes, the NLP algorithm achieved c-statistic of 0.92. The NLP algorithm identified 18 of the 21 patients (sensitivity 0.86) who had a VI whereas current procedural terminologyand international classification of diseases codes identified 6 of the 21 (sensitivity 0.29) patients. At this threshold, the NLP algorithm had a specificity of 0.93, negative predictive value of 0.99, positive predictive value of 0.51, and F1-score of 0.64.

CONCLUSION

Relying on administrative procedural and diagnosis codes may underestimate the rate of unintended intraoperative VI in anterior lumbar spine surgery. External and prospective validation of the algorithms presented here may improve quality and safety reporting.

Applying Text Mining, Clustering Analysis, and Latent Dirichlet Allocation Techniques for Topic Classification of Environmental Education Journals

Facing the big data wave, this study applied artificial intelligence to cite knowledge and find a feasible process to play a crucial role in supplying innovative value in environmental education. Intelligence agents of artificial intelligence and natural language processing (NLP) are two key areas leading the trend in artificial intelligence; this research adopted NLP to analyze the research topics of environmental education research journals in the Web of Science (WoS) database during 2011–2020 and interpret the categories and characteristics of abstracts for environmental education papers. The corpus data were selected from abstracts and keywords of research journal papers, which were analyzed with text mining, cluster analysis, latent Dirichlet allocation (LDA), and co-word analysis methods. The decisions regarding the classification of feature words were determined and reviewed by domain experts, and the associated TF-IDF weights were calculated for the following cluster analysis, which involved a combination of hierarchical clustering and K-means analysis. The hierarchical clustering and LDA decided the number of required categories as seven, and the K-means cluster analysis classified the overall documents into seven categories. This study utilized co-word analysis to check the suitability of the K-means classification, analyzed the terms with high TF-IDF wights for distinct K-means groups, and examined the terms for different topics with the LDA technique. A comparison of the results demonstrated that most categories that were recognized with K-means and LDA methods were the same and shared similar words; however, two categories had slight differences. The involvement of field experts assisted with the consistency and correctness of the classified topics and documents.

Ok Google: Using virtual assistants for data collection in psychological and behavioral research

Because of the increasing popularity of voice-controlled virtual assistants, such as Amazon’s Alexa and Google Assistant, they should be considered a new medium for psychological and behavioral research. We developed Survey Mate, an extension of Google Assistant, and conducted two studies to analyze the reliability and validity of data collected through this medium. In the first study, we assessed validated procrastination and shyness scales as well as social desirability indicators for both the virtual assistant and an online questionnaire. The results revealed comparable internal consistency and construct and criterion validity. In the second study, five social psychological experiments, which have been successfully replicated by the Many Labs projects, were successfully reproduced using a virtual assistant for data collection. Comparable effects were observed for users of both smartphones and smart speakers. Our findings point to the applicability of virtual assistants in data collection independent of the device used. While we identify some limitations, including data privacy concerns and a tendency toward more socially desirable responses, we found that virtual assistants could allow the recruitment of participants who are hard to reach with established data collection techniques, such as people with visual impairment, dyslexia, or lower education. This new medium could also be suitable for recruiting samples from non-Western countries because of its wide availability and easily adaptable language settings. It could also support an increase in the generalizability of theories in the future.

Improving broad-coverage medical entity linking with semantic type prediction and large-scale datasets

OBJECTIVES

Biomedical natural language processing tools are increasingly being applied for broad-coverage information extraction-extracting medical information of all types in a scientific document or a clinical note. In such broad-coverage settings, linking mentions of medical concepts to standardized vocabularies requires choosing the best candidate concepts from large inventories covering dozens of types. This study presents a novel semantic type prediction module for biomedical NLP pipelines and two automatically-constructed, large-scale datasets with broad coverage of semantic types.

METHODS

We experiment with five off-the-shelf biomedical NLP toolkits on four benchmark datasets for medical information extraction from scientific literature and clinical notes. All toolkits adopt a staged approach of mention detection followed by two stages of medical entity linking: (1) generating a list of candidate concepts, and (2) picking the best concept among them. We introduce a semantic type prediction module to alleviate the problem of overgeneration of candidate concepts by filtering out irrelevant candidate concepts based on the predicted semantic type of a mention. We present MedType, a fully modular semantic type prediction model which we integrate into the existing NLP toolkits. To address the dearth of broad-coverage training data for medical information extraction, we further present WikiMed and PubMedDS, two large-scale datasets for medical entity linking.

RESULTS

Semantic type filtering improves medical entity linking performance across all toolkits and datasets, often by several percentage points of F-1. Further, pretraining MedType on our novel datasets achieves state-of-the-art performance for semantic type prediction in biomedical text.

CONCLUSIONS

Semantic type prediction is a key part of building accurate NLP pipelines for broad-coverage information extraction from biomedical text. We make our source code and novel datasets publicly available to foster reproducible research.

An AI-assisted chatbot for radiation safety education in radiotherapy

Purpose. We created a virtual assistant chatbot that will serve as a tool for radiation safety training for clinical staff, including radiation oncologist, radiotherapist and medical physicist, in cancer treatment. The Bot can also be used to test their knowledge on radiation safety. Methods. The Bot was constructed using IBM’s Watson Assistant functionalities on the IBM cloud. A layered structure approach was used in the workflow of the Bot to interact with the user. Through answering various questions concerning radiation safety in radiotherapy, the users can learn the essential information to gain knowledge, when working in a cancer centre/hospital. Results. The user interface of the Bot was a front-end window operating on Internet, which could easily be accessed by any Internet-of-things such as smartphone, tablet or laptop. The Bot could communicate with the user for radiation safety Q&A. If the Bot could not identify what the user needed, the Bot would provide a list of options as a guidance. Using the natural language processing in communication, knowledge transfer from the Bot to user could be carried out. Conclusion. It is concluded that the radiation safety chatbot worked as intended, utilizing all the tools provided by the IBM Watson Assistant. The Bot could provide radiation safety information to the radiation staff effectively, and be used in staff training in radiotherapy.

ChemPix: automated recognition of hand-drawn hydrocarbon structures using deep learning

Inputting molecules into chemistry software, such as quantum chemistry packages, currently requires domain expertise, expensive software and/or cumbersome procedures. Leveraging recent breakthroughs in machine learning, we develop ChemPix: an offline, hand-drawn hydrocarbon structure recognition tool designed to remove these barriers. A neural image captioning approach consisting of a convolutional neural network (CNN) encoder and a long short-term memory (LSTM) decoder learned a mapping from photographs of hand-drawn hydrocarbon structures to machine-readable SMILES representations. We generated a large auxiliary training dataset, based on RDKit molecular images, by combining image augmentation, image degradation and background addition. Additionally, a small dataset of ∼600 hand-drawn hydrocarbon chemical structures was crowd-sourced using a phone web application. These datasets were used to train the image-to-SMILES neural network with the goal of maximizing the hand-drawn hydrocarbon recognition accuracy. By forming a committee of the trained neural networks where each network casts one vote for the predicted molecule, we achieved a nearly 10 percentage point improvement of the molecule recognition accuracy and were able to assign a confidence value for the prediction based on the number of agreeing votes. The ensemble model achieved an accuracy of 76% on hand-drawn hydrocarbons, increasing to 86% if the top 3 predictions were considered.

A Novel Hybrid Methodology of Measuring Sentence Similarity

The problem of measuring sentence similarity is an essential issue in the natural language processing area. It is necessary to measure the similarity between sentences accurately. Sentence similarity measuring is the task of finding semantic symmetry between two sentences, regardless of word order and context of the words. There are many approaches to measuring sentence similarity. Deep learning methodology shows a state-of-the-art performance in many natural language processing fields and is used a lot in sentence similarity measurement methods. However, in the natural language processing field, considering the structure of the sentence or the word structure that makes up the sentence is also important. In this study, we propose a methodology combined with both deep learning methodology and a method considering lexical relationships. Our evaluation metric is the Pearson correlation coefficient and Spearman correlation coefficient. As a result, the proposed method outperforms the current approaches on a KorSTS standard benchmark Korean dataset. Moreover, it performs a maximum of a 65% increase than only using deep learning methodology. Experiments show that our proposed method generally results in better performance than those with only a deep learning model.

Automated rating of patient and physician emotion in primary care visits

OBJECTIVE

Train machine learning models that automatically predict emotional valence of patient and physician in primary care visits.

METHODS

Using transcripts from 353 primary care office visits with 350 patients and 84 physicians (Cook, 2002 [1], Tai-Seale et al., 2015 [2]), we developed two machine learning models (a recurrent neural network with a hierarchical structure and a logistic regression classifier) to recognize the emotional valence (positive, negative, neutral) (Posner et al., 2005 [3]) of each utterance. We examined the agreement of human-generated ratings of emotional valence with machine learning model ratings of emotion.

RESULTS

The agreement of emotion ratings from the recurrent neural network model with human ratings was comparable to that of human-human inter-rater agreement. The weighted-average of the correlation coefficients for the recurrent neural network model with human raters was 0.60, and the human rater agreement was also 0.60.

CONCLUSIONS

The recurrent neural network model predicted the emotional valence of patients and physicians in primary care visits with similar reliability as human raters.

PRACTICE IMPLICATIONS

As the first machine learning-based evaluation of emotion recognition in primary care visit conversations, our work provides valuable baselines for future applications that might help monitor patient emotional signals, supporting physicians in empathic communication, or examining the role of emotion in patient-centered care.

The straw man effect: Partisan misrepresentation in natural language

Political discourse often seems divided not just by different preferences, but by entirely different representations of the debate. Are partisans able to accurately describe their opponents’ position, or do they instead generate unrepresentative “straw man” arguments? In this research we examined an (incentivized) political imitation game by asking partisans on both sides of the U.S. health care debate to describe the most common arguments for and against ObamaCare. We used natural language-processing algorithms to benchmark the biases and blind spots of our participants. Overall, partisans showed a limited ability to simulate their opponents’ perspective, or to distinguish genuine from imitation arguments. In general, imitations were less extreme than their genuine counterparts. Individual difference analyses suggest that political sophistication only improves the representations of one’s own side but not of an opponent’s side, exacerbating the straw man effect. Our findings suggest that false beliefs about partisan opponents may be pervasive.

Content analysis-based documentation and exploration of research articles

Purpose

With the wealth of information available on the World Wide Web, it is difficult for anyone from a general user to the researcher to easily fulfill their information need. The main challenge is to categorize the documents systematically and also take into account more valuable data such as semantic information. The purpose of this paper is to develop a concept-based search system that leverages the external knowledge resources as the background knowledge for getting the accurate and efficient meaningful search results.

Design/methodology/approach

The paper introduces the approach which is based on formal concept analysis (FCA) with the semantic information to support the document management in information retrieval (IR). To describe the semantic information of the documents, the system uses the popular knowledge resources WordNet and Wikipedia. By using FCA, the system creates the concept lattice as the concept hierarchy of the document and proposes the navigation algorithm for retrieving the hierarchy based on the user query.

Findings

The semantic information of the document is based on the two external popular knowledge resources; the authors find that it will be more efficient to deal with the semantic mismatch problems of user need.

Originality/value

The navigation algorithm proposed in this research is applied to the scientific articles of the National Science Foundation (NSF). The proposed system can enhance the integration and exploration of the scientific articles for the advancement of the Scientific and Engineering Research Community.

Machine learning for sperm selection

Infertility rates and the number of couples seeking fertility care have increased worldwide over the past few decades. Over 2.5 million cycles of assisted reproductive technologies are being performed globally every year, but the success rate has remained at ~33%. Machine learning, an automated method of data analysis based on patterns and inference, is increasingly being deployed within the health-care sector to improve diagnostics and therapeutics. This technique is already aiding embryo selection in some fertility clinics, and has also been applied in research laboratories to improve sperm analysis and selection. Tremendous opportunities exist for machine learning to advance male fertility treatments. The fundamental challenge of sperm selection - selecting the most promising candidate from 10⁸ gametes - presents a challenge that is uniquely well-suited to the high-throughput capabilities of machine learning algorithms paired with modern data processing capabilities.

Recognition of the Typical Distress in Concrete Pavement Based on GPR and 1D-CNN

Ground-penetrating radar (GPR) signal recognition depends much on manual feature extraction. However, the complexity of radar detection signals leads to conventional intelligent algorithms lacking sufficient flexibility in concrete pavement detection. Focused on these problems, we proposed an adaptive one-dimensional convolution neural network (1D-CNN) algorithm for interpreting GPR data. Firstly, the training dataset and testing dataset were constructed from the detection signals on pavement samples of different types of distress; secondly, the raw signals are were directly inputted into the 1D-CNN model, and the raw signal features of the radar wave are extracted using the adaptive deep learning network; finally, the output used the Soft-Max classifier to provide the classification result of the concrete pavement distress. Through simulation experiments and actual field testing, the results show that the proposed method has high accuracy and excellent generalization performance compared to the conventional method. It also has practical applications.

The Application and Development of Deep Learning in Radiotherapy: A Systematic Review

With the massive use of computers, the growth and explosion of data has greatly promoted the development of artificial intelligence (AI). The rise of deep learning (DL) algorithms, such as convolutional neural networks (CNN), has provided radiation oncologists with many promising tools that can simplify the complex radiotherapy process in the clinical work of radiation oncology, improve the accuracy and objectivity of diagnosis, and reduce the workload, thus enabling clinicians to spend more time on advanced decision-making tasks. As the development of DL gets closer to clinical practice, radiation oncologists will need to be more familiar with its principles to properly evaluate and use this powerful tool. In this paper, we explain the development and basic concepts of AI and discuss its application in radiation oncology based on different task categories of DL algorithms. This work clarifies the possibility of further development of DL in radiation oncology.

Robot Grasping System and Grasp Stability Prediction Based on Flexible Tactile Sensor Array

As an essential perceptual device, the tactile sensor can efficiently improve robot intelligence by providing contact force perception to develop algorithms based on contact force feedback. However, current tactile grasping technology lacks high-performance sensors and high-precision grasping prediction models, which limits its broad application. Herein, an intelligent robot grasping system that combines a highly sensitive tactile sensor array was constructed. A dataset that can reflect the grasping contact force of various objects was set up by multiple grasping operation feedback from a tactile sensor array. The stability state of each grasping operation was also recorded. On this basis, grasp stability prediction models with good performance in grasp state judgment were proposed. By feeding training data into different machine learning algorithms and comparing the judgment results, the best grasp prediction model for different scenes can be obtained. The model was validated to be efficient, and the judgment accuracy was over 98% in grasp stability prediction with limited training data. Further, experiments prove that the real-time contact force input based on the feedback of the tactile sensor array can periodically control robots to realize stable grasping according to the real-time grasping state of the prediction model.

From Classical Machine Learning to Deep Neural Networks: A Simplified Scientometric Review

There are promising prospects on the way to widespread use of AI, as well as problems that need to be overcome to adapt AI&ML technologies in industries. The paper systematizes the AI sections and calculates the dynamics of changes in the number of scientific articles in machine learning sections according to Google Scholar. The method of data acquisition and calculation of dynamic indicators of changes in publication activity is described: growth rate (D1) and acceleration of growth (D2) of scientific publications. Analysis of publication activity, in particular, showed a high interest in modern transformer models, the development of datasets for some industries, and a sharp increase in interest in methods of explainable machine learning. Relatively small research domains are receiving increasing attention, as evidenced by the negative correlation between the number of articles and D1 and D2 scores. The results show that, despite the limitations of the method, it is possible to (1) identify fast-growing areas of research regardless of the number of articles, and (2) predict publication activity in the short term with satisfactory accuracy for practice (the average prediction error for the year ahead is 6%, with a standard deviation of 7%). This paper presents results for more than 400 search queries related to classified research areas and the application of machine learning models to industries. The proposed method evaluates the dynamics of growth and the decline of scientific domains associated with certain key terms. It does not require access to large bibliometric archives and allows to relatively quickly obtain quantitative estimates of dynamic indicators.

Diagnostic Informatics-The Role of Digital Health in Diagnostic Stewardship and the Achievement of Excellence, Safety, and Value

Diagnostic investigations (pathology laboratory and medical imaging) aim to: increase certainty of the presence or absence of disease by supporting the process of differential diagnosis; support clinical management; and monitor a patient's trajectory (e. g., disease progression or response to treatment). Digital health can be defined as the collection, storage, retrieval, transmission, and utilization of data, information, and knowledge to support healthcare. Digital health has become an essential component of the diagnostic process, helping to facilitate the accuracy and timeliness of information transfer and enhance the effectiveness of decision-making processes. Digital health is also important to diagnostic stewardship, which involves coordinated guidance and interventions to ensure the appropriate utilization of diagnostic tests for therapeutic decision-making. Diagnostic stewardship and informatics are thus important in efforts to establish shared decision-making. This is because they contribute to the establishment of shared information platforms (enabling patients to read, comment on, and share in decisions about their care) based on timely and meaningful communication. This paper will outline key diagnostic informatics and stewardship initiatives across three interrelated fields: (1) diagnostic error and the establishment of outcomes-based diagnostic research; (2) the safety and effectiveness of test result management and follow-up; and (3) digitally enhanced decision support systems.

‘Alexa, I feel for you!’ Observers’ Empathetic Reactions towards a Conversational Agent

Conversational agents and smart speakers have grown in popularity offering a variety of options for use, which are available through intuitive speech operation. In contrast to the standard dyad of a single user and a device, voice-controlled operations can be observed by further attendees resulting in new, more social usage scenarios. Referring to the concept of ‘media equation’ and to research on the idea of ‘computers as social actors,’ which describes the potential of technology to trigger emotional reactions in users, this paper asks for the capacity of smart speakers to elicit empathy in observers of interactions. In a 2 × 2 online experiment, 140 participants watched a video of a man talking to an Amazon Echo either rudely or neutrally (factor 1), addressing it as ‘Alexa’ or ‘Computer’ (factor 2). Controlling for participants’ trait empathy, the rude treatment results in participants’ significantly higher ratings of empathy with the device, compared to the neutral treatment. The form of address had no significant effect. Results were independent of the participants’ gender and usage experience indicating a rather universal effect, which confirms the basic idea of the media equation. Implications for users, developers and researchers were discussed in the light of (future) omnipresent voice-based technology interaction scenarios.

Addressing Biodisaster X Threats With Artificial Intelligence and 6G Technologies: Literature Review and Critical Insights

BACKGROUND

With advances in science and technology, biotechnology is becoming more accessible to people of all demographics. These advances inevitably hold the promise to improve personal and population well-being and welfare substantially. It is paradoxical that while greater access to biotechnology on a population level has many advantages, it may also increase the likelihood and frequency of biodisasters due to accidental or malicious use. Similar to "Disease X" (describing unknown naturally emerging pathogenic diseases with a pandemic potential), we term this unknown risk from biotechnologies "Biodisaster X." To date, no studies have examined the potential role of information technologies in preventing and mitigating Biodisaster X.

OBJECTIVE

This study aimed to explore (1) what Biodisaster X might entail and (2) solutions that use artificial intelligence (AI) and emerging 6G technologies to help monitor and manage Biodisaster X threats.

METHODS

A review of the literature on applying AI and 6G technologies for monitoring and managing biodisasters was conducted on PubMed, using articles published from database inception through to November 16, 2020.

RESULTS

Our findings show that Biodisaster X has the potential to upend lives and livelihoods and destroy economies, essentially posing a looming risk for civilizations worldwide. To shed light on Biodisaster X threats, we detailed effective AI and 6G-enabled strategies, ranging from natural language processing to deep learning-based image analysis to address issues ranging from early Biodisaster X detection (eg, identification of suspicious behaviors), remote design and development of pharmaceuticals (eg, treatment development), and public health interventions (eg, reactive shelter-at-home mandate enforcement), as well as disaster recovery (eg, sentiment analysis of social media posts to shed light on the public's feelings and readiness for recovery building).

CONCLUSIONS

Biodisaster X is a looming but avoidable catastrophe. Considering the potential human and economic consequences Biodisaster X could cause, actions that can effectively monitor and manage Biodisaster X threats must be taken promptly and proactively. Rather than solely depending on overstretched professional attention of health experts and government officials, it is perhaps more cost-effective and practical to deploy technology-based solutions to prevent and control Biodisaster X threats. This study discusses what Biodisaster X could entail and emphasizes the importance of monitoring and managing Biodisaster X threats by AI techniques and 6G technologies. Future studies could explore how the convergence of AI and 6G systems may further advance the preparedness for high-impact, less likely events beyond Biodisaster X.

A Thousand Words Express a Common Idea? Understanding International Tourists’ Reviews of Mt. Huangshan, China, through a Deep Learning Approach

Tourists’ experiential perceptions and specific behaviors are of importance to facilitate geographers’ and planners’ understanding of landscape surroundings. In addition, the potentially significant role of online user generated content (UGC) in tourism landscape research has only received limited attention, especially in the era of artificial intelligence. The motivation of the present study is to understand international tourists’ online reviews of Mt. Huangshan in China. Through a state-of-the-art natural language processing network (BERT) analyzing posted reviews across international tourists, our results facilitate relevant landscape development and design decisions. Second, the proposed analytic method can be an exemplified model to inspire relevant landscape planners and decision-makers to conduct future researches. Through the clustering results, several key topics are revealed, including international tourists’ perceptual image of Mt. Huangshan, tour route planning, and negative experience of staying.

Deep ensemble multitask classification of emergency medical call incidents combining multimodal data improves emergency medical dispatch

The objective of this work was to develop a predictive model to aid non-clinical dispatchers to classify emergency medical call incidents by their life-threatening level (yes/no), admissible response delay (undelayable, minutes, hours, days) and emergency system jurisdiction (emergency system/primary care) in real time. We used a total of 1 244 624 independent incidents from the Valencian emergency medical dispatch service in Spain, compiled in retrospective from 2009 to 2012, including clinical features, demographics, circumstantial factors and free text dispatcher observations. Based on them, we designed and developed DeepEMC², a deep ensemble multitask model integrating four subnetworks: three specialized to context, clinical and text data, respectively, and another to ensemble the former. The four subnetworks are composed in turn by multi-layer perceptron modules, bidirectional long short-term memory units and a bidirectional encoding representations from transformers module. DeepEMC² showed a macro F1-score of 0.759 in life-threatening classification, 0.576 in admissible response delay and 0.757 in emergency system jurisdiction. These results show a substantial performance increase of 12.5 %, 17.5 % and 5.1 %, respectively, with respect to the current in-house triage protocol of the Valencian emergency medical dispatch service. Besides, DeepEMC² significantly outperformed a set of baseline machine learning models, including naive bayes, logistic regression, random forest and gradient boosting (α = 0.05). Hence, DeepEMC² is able to: 1) capture information present in emergency medical calls not considered by the existing triage protocol, and 2) model complex data dependencies not feasible by the tested baseline models. Likewise, our results suggest that most of this unconsidered information is present in the free text dispatcher observations. To our knowledge, this study describes the first deep learning model undertaking emergency medical call incidents classification. Its adoption in medical dispatch centers would potentially improve emergency dispatch processes, resulting in a positive impact in patient wellbeing and health services sustainability.

A Visually Interpretable Deep Learning Framework for Histopathological Image-Based Skin Cancer Diagnosis

Owing to the high incidence rate and the severe impact of skin cancer, the precise diagnosis of malignant skin tumors is a significant goal, especially considering treatment is normally effective if the tumor is detected early. Limited published histopathological image sets and the lack of an intuitive correspondence between the features of lesion areas and a certain type of skin cancer pose a challenge to the establishment of high-quality and interpretable computer-aided diagnostic (CAD) systems. To solve this problem, a light-weight attention mechanism-based deep learning framework, namely, DRANet, is proposed to differentiate 11 types of skin diseases based on a real histopathological image set collected by us during the last 10 years. The CAD system can output not only the name of a certain disease but also a visualized diagnostic report showing possible areas related to the disease. The experimental results demonstrate that the DRANet obtains significantly better performance than baseline models (i.e., InceptionV3, ResNet50, VGG16, and VGG19) with comparable parameter size and competitive accuracy with fewer model parameters. Visualized results produced by the hidden layers of the DRANet actually highlight part of the class-specific regions of diagnostic points and are valuable for decision making in the diagnosis of skin diseases.

Identifying research gaps: A review of virtual patient education and self-management

BACKGROUND

Avatars in Virtual Reality (VR) can not only represent humans, but also embody intelligent software agents that communicate with humans, thus enabling a new paradigm of human-machine interaction.

OBJECTIVE

The research agenda proposed in this paper by an interdisciplinary team is motivated by the premise that a conversation with a smart agent avatar in VR means more than giving a face and body to a chatbot. Using the concrete communication task of patient education, this research agenda is rather intended to explore which patterns and practices must be constructed visually, verbally, para- and nonverbally between humans and embodied machines in a counselling context so that humans can integrate counselling by an embodied VR smart agent into their thinking and acting in one way or another.

METHODS

The scientific literature in different bibliographical databases was reviewed. A qualitative narrative approach was applied for analysis.

RESULTS

A research agenda is proposed which investigates how recurring consultations of patients with healthcare professionals are currently conducted and how they could be conducted with an embodied smart agent in immersive VR.

CONCLUSIONS

Interdisciplinary teams consisting of linguists, computer scientists, visual designers and health care professionals are required which need to go beyond a technology-centric solution design approach. Linguists' insights from discourse analysis drive the explorative experiments to identify test and discover what capabilities and attributes the smart agent in VR must have, in order to communicate effectively with a human being.

Affective Voice Interaction and Artificial Intelligence: A Research Study on the Acoustic Features of Gender and the Emotional States of the PAD Model

New types of artificial intelligence products are gradually transferring to voice interaction modes with the demand for intelligent products expanding from communication to recognizing users' emotions and instantaneous feedback. At present, affective acoustic models are constructed through deep learning and abstracted into a mathematical model, making computers learn from data and equipping them with prediction abilities. Although this method can result in accurate predictions, it has a limitation in that it lacks explanatory capability; there is an urgent need for an empirical study of the connection between acoustic features and psychology as the theoretical basis for the adjustment of model parameters. Accordingly, this study focuses on exploring the differences between seven major “acoustic features” and their physical characteristics during voice interaction with the recognition and expression of “gender” and “emotional states of the pleasure-arousal-dominance (PAD) model.” In this study, 31 females and 31 males aged between 21 and 60 were invited using the stratified random sampling method for the audio recording of different emotions. Subsequently, parameter values of acoustic features were extracted using Praat voice software. Finally, parameter values were analyzed using a Two-way ANOVA, mixed-design analysis in SPSS software. Results show that gender and emotional states of the PAD model vary among seven major acoustic features. Moreover, their difference values and rankings also vary. The research conclusions lay a theoretical foundation for AI emotional voice interaction and solve deep learning's current dilemma in emotional recognition and parameter optimization of the emotional synthesis model due to the lack of explanatory power.

Extracting knowledge networks from plant scientific literature: potato tuber flesh color as an exemplary trait

BACKGROUND

Scientific literature carries a wealth of information crucial for research, but only a fraction of it is present as structured information in databases and therefore can be analyzed using traditional data analysis tools. Natural language processing (NLP) is often and successfully employed to support humans by distilling relevant information from large corpora of free text and structuring it in a way that lends itself to further computational analyses. For this pilot, we developed a pipeline that uses NLP on biological literature to produce knowledge networks. We focused on the flesh color of potato, a well-studied trait with known associations, and we investigated whether these knowledge networks can assist us in formulating new hypotheses on the underlying biological processes.

RESULTS

We trained an NLP model based on a manually annotated corpus of 34 full-text potato articles, to recognize relevant biological entities and relationships between them in text (genes, proteins, metabolites and traits). This model detected the number of biological entities with a precision of 97.65% and a recall of 88.91% on the training set. We conducted a time series analysis on 4023 PubMed abstract of plant genetics-based articles which focus on 4 major Solanaceous crops (tomato, potato, eggplant and capsicum), to determine that the networks contained both previously known and contemporaneously unknown leads to subsequently discovered biological phenomena relating to flesh color. A novel time-based analysis of these networks indicates a connection between our trait and a candidate gene (zeaxanthin epoxidase) already two years prior to explicit statements of that connection in the literature.

CONCLUSIONS

Our time-based analysis indicates that network-assisted hypothesis generation shows promise for knowledge discovery, data integration and hypothesis generation in scientific research.

Improving the Retrieval of Arabic Web Search Results Using Enhanced k-Means Clustering Algorithm

Traditional information retrieval systems return a ranked list of results to a user’s query. This list is often long, and the user cannot explore all the results retrieved. It is also ineffective for a highly ambiguous language such as Arabic. The modern writing style of Arabic excludes the diacritical marking, without which Arabic words become ambiguous. For a search query, the user has to skim over the document to infer if the word has the same meaning they are after, which is a time-consuming task. It is hoped that clustering the retrieved documents will collate documents into clear and meaningful groups. In this paper, we use an enhanced k-means clustering algorithm, which yields a faster clustering time than the regular k-means. The algorithm uses the distance calculated from previous iterations to minimize the number of distance calculations. We propose a system to cluster Arabic search results using the enhanced k-means algorithm, labeling each cluster with the most frequent word in the cluster. This system will help Arabic web users identify each cluster’s topic and go directly to the required cluster. Experimentally, the enhanced k-means algorithm reduced the execution time by 60% for the stemmed dataset and 47% for the non-stemmed dataset when compared to the regular k-means, while slightly improving the purity.

Artificial Intelligence in Plastic Surgery: Applications and Challenges

New developments in artificial intelligence (AI) offer opportunities to enhance plastic surgery practice, research, and education. In this article, we review relevant AI tools and applications, including machine learning, reinforcement learning, and natural language processing. Our own Markov decision process for keloid treatment illustrates how these models are developed and can be used to enhance decision-making in clinical practice. Finally, we discuss challenges of implementing AI and knowledge gaps that must be addressed to successfully apply AI in plastic surgery. Level of Evidence V This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

The growing amplification of social media: measuring temporal and social contagion dynamics for over 150 languages on Twitter for 2009-2020

Working from a dataset of 118 billion messages running from the start of 2009 to the end of 2019, we identify and explore the relative daily use of over 150 languages on Twitter. We find that eight languages comprise 80% of all tweets, with English, Japanese, Spanish, Arabic, and Portuguese being the most dominant. To quantify social spreading in each language over time, we compute the 'contagion ratio': The balance of retweets to organic messages. We find that for the most common languages on Twitter there is a growing tendency, though not universal, to retweet rather than share new content. By the end of 2019, the contagion ratios for half of the top 30 languages, including English and Spanish, had reached above 1-the naive contagion threshold. In 2019, the top 5 languages with the highest average daily ratios were, in order, Thai (7.3), Hindi, Tamil, Urdu, and Catalan, while the bottom 5 were Russian, Swedish, Esperanto, Cebuano, and Finnish (0.26). Further, we show that over time, the contagion ratios for most common languages are growing more strongly than those of rare languages.

How do you feel? Using natural language processing to automatically rate emotion in psychotherapy

Emotional distress is a common reason for seeking psychotherapy, and sharing emotional material is central to the process of psychotherapy. However, systematic research examining patterns of emotional exchange that occur during psychotherapy sessions is often limited in scale. Traditional methods for identifying emotion in psychotherapy rely on labor-intensive observer ratings, client or therapist ratings obtained before or after sessions, or involve manually extracting ratings of emotion from session transcripts using dictionaries of positive and negative words that do not take the context of a sentence into account. However, recent advances in technology in the area of machine learning algorithms, in particular natural language processing, have made it possible for mental health researchers to identify sentiment, or emotion, in therapist-client interactions on a large scale that would be unattainable with more traditional methods. As an attempt to extend prior findings from Tanana et al. (2016), we compared their previous sentiment model with a common dictionary-based psychotherapy model, LIWC, and a new NLP model, BERT. We used the human ratings from a database of 97,497 utterances from psychotherapy to train the BERT model. Our findings revealed that the unigram sentiment model (kappa = 0.31) outperformed LIWC (kappa = 0.25), and ultimately BERT outperformed both models (kappa = 0.48).

Natural language processing for urban research: A systematic review

Natural language processing (NLP) has shown potential as a promising tool to exploit under-utilized urban data sources. This paper presents a systematic review of urban studies published in peer-reviewed journals and conference proceedings that adopted NLP. The review suggests that the application of NLP in studying cities is still in its infancy. Current applications fell into five areas: urban governance and management, public health, land use and functional zones, mobility, and urban design. NLP demonstrates the advantages of improving the usability of urban big data sources, expanding study scales, and reducing research costs. On the other hand, to take advantage of NLP, urban researchers face challenges of raising good research questions, overcoming data incompleteness, inaccessibility, and non-representativeness, immature NLP techniques, and computational skill requirements. This review is among the first efforts intended to provide an overview of existing applications and challenges for advancing urban research through the adoption of NLP.

KazNewsDataset: Single Country Overall Digital Mass Media Publication Corpus

Mass media is one of the most important elements influencing the information environment of society. The mass media is not only a source of information about what is happening but is often the authority that shapes the information agenda, the boundaries, and forms of discussion on socially relevant topics. A multifaceted and, where possible, quantitative assessment of mass media performance is crucial for understanding their objectivity, tone, thematic focus and, quality. The paper presents a corpus of Kazakhstan media, which contains over 4 million publications from 36 primary sources (which has at least 500 publications). The corpus also includes more than 2 million texts of Russian media for comparative analysis of publication activity of the countries, also about 4000 sections of state policy documents. The paper briefly describes the natural language processing and multiple-criteria decision-making methods, which are the algorithmic basis of the text and mass media evaluation method, and describes the results of several research cases, such as identification of propaganda, assessment of the tone of publications, calculation of the level of socially relevant negativity, comparative analysis of publication activity in the field of renewable energy. Experiments confirm the general possibility of evaluating the socially significant news, identifying texts with propagandistic content, evaluating the sentiment of publications using the topic model of the text corpus since the area under receiver operating characteristics curve (ROC AUC) values of 0.81, 0.73 and 0.93 were achieved on abovementioned tasks. The described cases do not exhaust the possibilities of thematic, tonal, dynamic, etc., analysis of the considered corpus of texts. The corpus will be interesting to researchers considering both multiple publications and mass media analysis, including comparative analysis and identification of common patterns inherent in the media of different countries.

Cognitive analysis of metabolomics data for systems biology

Cognitive computing is revolutionizing the way big data are processed and integrated, with artificial intelligence (AI) natural language processing (NLP) platforms helping researchers to efficiently search and digest the vast scientific literature. Most available platforms have been developed for biomedical researchers, but new NLP tools are emerging for biologists in other fields and an important example is metabolomics. NLP provides literature-based contextualization of metabolic features that decreases the time and expert-level subject knowledge required during the prioritization, identification and interpretation steps in the metabolomics data analysis pipeline. Here, we describe and demonstrate four workflows that combine metabolomics data with NLP-based literature searches of scientific databases to aid in the analysis of metabolomics data and their biological interpretation. The four procedures can be used in isolation or consecutively, depending on the research questions. The first, used for initial metabolite annotation and prioritization, creates a list of metabolites that would be interesting for follow-up. The second workflow finds literature evidence of the activity of metabolites and metabolic pathways in governing the biological condition on a systems biology level. The third is used to identify candidate biomarkers, and the fourth looks for metabolic conditions or drug-repurposing targets that the two diseases have in common. The protocol can take 1-4 h or more to complete, depending on the processing time of the various software used.

Classification of the Disposition of Patients Hospitalized with COVID-19: Reading Discharge Summaries Using Natural Language Processing

Background

Medical notes are a rich source of patient data; however, the nature of unstructured text has largely precluded the use of these data for large retrospective analyses. Transforming clinical text into structured data can enable large-scale research studies with electronic health records (EHR) data. Natural language processing (NLP) can be used for text information retrieval, reducing the need for labor-intensive chart review. Here we present an application of NLP to large-scale analysis of medical records at 2 large hospitals for patients hospitalized with COVID-19.

Objective

Our study goal was to develop an NLP pipeline to classify the discharge disposition (home, inpatient rehabilitation, skilled nursing inpatient facility [SNIF], and death) of patients hospitalized with COVID-19 based on hospital discharge summary notes.

Methods

Text mining and feature engineering were applied to unstructured text from hospital discharge summaries. The study included patients with COVID-19 discharged from 2 hospitals in the Boston, Massachusetts area (Massachusetts General Hospital and Brigham and Women’s Hospital) between March 10, 2020, and June 30, 2020. The data were divided into a training set (70%) and hold-out test set (30%). Discharge summaries were represented as bags-of-words consisting of single words (unigrams), bigrams, and trigrams. The number of features was reduced during training by excluding n-grams that occurred in fewer than 10% of discharge summaries, and further reduced using least absolute shrinkage and selection operator (LASSO) regularization while training a multiclass logistic regression model. Model performance was evaluated using the hold-out test set.

Results

The study cohort included 1737 adult patients (median age 61 [SD 18] years; 55% men; 45% White and 16% Black; 14% nonsurvivors and 61% discharged home). The model selected 179 from a vocabulary of 1056 engineered features, consisting of combinations of unigrams, bigrams, and trigrams. The top features contributing most to the classification by the model (for each outcome) were the following: “appointments specialty,” “home health,” and “home care” (home); “intubate” and “ARDS” (inpatient rehabilitation); “service” (SNIF); “brief assessment” and “covid” (death). The model achieved a micro-average area under the receiver operating characteristic curve value of 0.98 (95% CI 0.97-0.98) and average precision of 0.81 (95% CI 0.75-0.84) in the testing set for prediction of discharge disposition.

Conclusions

A supervised learning–based NLP approach is able to classify the discharge disposition of patients hospitalized with COVID-19. This approach has the potential to accelerate and increase the scale of research on patients’ discharge disposition that is possible with EHR data.

Natural Language Processing and Machine Learning for Detection of Respiratory Illness by Chest CT Imaging and Tracking of COVID-19 Pandemic in the US

BACKGROUND

Coronavirus disease 2019 (COVID-19) has spread quickly throughout the United States (US) causing significant disruption in healthcare and society. Tools to identify hot spots are important for public health planning. The goal of our study was to determine if natural language processing (NLP) algorithm assessment of thoracic computed tomography (CT) imaging reports correlated with the incidence of official COVID-19 cases in the US.

METHODS

Using de-identified HIPAA compliant patient data from our common imaging platform interconnected with over 2,100 facilities covering all 50 states, we developed three NLP algorithms to track positive CT imaging features of respiratory illness typical in SARS-CoV-2 viral infection. We compared our findings against the number of official COVID-19 daily, weekly and state-wide.

RESULTS

The NLP algorithms were applied to 450,114 patient chest CT comprehensive reports gathered from January 1st to October 3rd, 2020. The best performing NLP model exhibited strong correlation with daily official COVID-19 cases (r2=0.82, p<0.005). The NLP models demonstrated an early rise in cases followed by the increase of official cases, suggesting the possibility of an early predictive marker, with strong correlation to official cases on a weekly basis (r2=0.91, p<0.005). There was also substantial correlation between the NLP and official COVID-19 incidence by state (r2=0.92, p<0.005).

CONCLUSION

Using big data, we developed a novel machine-learning based NLP algorithm that can track imaging findings of respiratory illness detected on chest CT imaging reports with strong correlation with the progression of the COVID-19 pandemic in the US.

Motivation and reward mechanisms in health behavior change processes

With increasing prevalence of lifestyle-related chronic diseases worldwide, understanding health behavior change and the development of successful interventions to support lifestyle modification is gaining increasing interest among politicians, scientists, therapists and patients alike. A number of health behavior change theories have been developed aiming at explaining health behavior change and understanding the domains that make change more likely. Until now, only few studies have taken into account automatic, implicit or non-cognitive aspects of behavior, including emotion and positive affect. Recent progress in the neuroscience of motivation and reward systems can provide further insights into the relevance of such domains. In this integrative review, we present a description of the possible motivation and reward systems (approach/wanting = pleasure; aversion/avoiding = relief; assertion/non-wanting = quiescence) involved in behavior change. Therefore, based on established theories encompassing both initiation and maintenance of behavior change, we create a flexible seven-stage behavior change process with three engagement phases (non-engagement, motivational engagement, executive engagement) and relate the motivation and reward systems to each of these stages. We propose that either appetitive (preferably) or aversive motivational salience is activated during motivational engagement, that learning leads to continued behavior and that assertive salience prevails when the new behavior has become habitual. We discuss under which circumstances these mechanisms and reward-motivation pathways are likely to occur and address potential shortcomings of our proposed theoretical framework. We highlight implications for future interventions aiming at lifestyle modification.

A class for itself? On the worldviews of the new tech elite

The emergence of a new tech elite in Silicon Valley and beyond raises questions about the economic reach, political influence, and social importance of this group. How do these inordinately influential people think about the world and about our common future? In this paper, we test a) whether members of the tech elite share a common, meritocratic view of the world, b) whether they have a "mission" for the future, and c) how they view democracy as a political system. Our data set consists of information about the 100 richest people in the tech world, according to Forbes, and rests on their published pronouncements on Twitter, as well as on their statements on the websites of their philanthropic endeavors. Automated "bag-of-words" text and sentiment analyses reveal that the tech elite has a more meritocratic view of the world than the general US Twitter-using population. The tech elite also frequently promise to "make the world a better place," but they do not differ from other extremely wealthy people in this respect. However, their relationship to democracy is contradictory. Based on these results, we conclude that the tech elite may be thought of as a "class for itself" in Marx's sense-a social group that shares particular views of the world, which in this case means meritocratic, missionary, and inconsistent democratic ideology.

Applying Artificial Intelligence in Physical Education and Future Perspectives

Artificial intelligence (AI) is gradually influencing every aspect of everyday life, including education. AI can also provide special support to learners through academic sustainability or discontinuation predictions. While AI research remains in its early stages, we must examine how it evolves and exerts its potential over time. By utilizing AI in physical education (PE), we can increase its potential use in sports applications, and enact changes upon the nature of PE, its visualization, and repeatability. Based on the concept of AI and related research areas, this study explores its principles and use in PE, and presents a focused, in-depth analysis of the areas of PE technology where AI could be applied—customized PE classes, knowledge provision, learner evaluation, and learner counseling methods. Our findings highlight the expertise required for future PE teachers in applying AI. Regarding practice implications, this study addresses the topic of AI innovations affecting all life domains, including PE; it highlights AI applications’ relevance to PE technology, based on existing research; it proposes that the implications of AI for PE may apply to other educational domains; and finally, it contributes to existing literature and also shares future research prospects regarding AI applications in education and sports.

Development and Optimization of Clinical Informatics Infrastructure to Support Bioinformatics at an Oncology Center

Translational bioinformatics for therapeutic discovery requires the infrastructure of clinical informatics. In this chapter, we describe the clinical informatics components needed for successful implementation of translational research at a cancer center. This chapter is meant to be an introduction to those clinical informatics concepts that are needed for translational research. For a detailed account of clinical informatics, the authors will guide the reader to comprehensive resources. We provide examples of workflows from Moffitt Cancer Center led by Drs. Perkins and Markowitz. This perspective represents an interesting collaboration as Dr. Perkins is the Chief Medical Information Officer and Dr. Markowitz is a translational researcher in Melanoma with an active informatics component to his laboratory to study the mechanisms of resistance to checkpoint blockade and an active member of the clinical informatics team.

Identifying gaps in using artificial intelligence to support students with intellectual disabilities from education and health perspectives

Purpose

Students with developmental/intellectual disabilities (ID/DD) often have serious health issues that require additional medical care and supervision. Serious health issues also mean increased absence and additional lags in academic achievement and development of adaptive and social skills. The incorporation of artificial intelligence in the education of a child with ID/DD could ameliorate the educational, adaptive and social skill gaps that occur as a direct result of persistent health problems.

Design/methodology/approach

The literature regarding the use of artificial intelligence in education for students with ID/DD was collected systematically from international online databases based on specific inclusion and exclusion criteria. The collected articles were analyzed deductively, looking for the different gaps in the domain. Based on the literature, an artificial intelligence–based architecture is proposed and sketched.

Findings

The findings show that there are many gaps in supporting students with ID/DD through the utilization of artificial intelligence. Given that the majority of students with ID/DD often have serious and chronic and comorbid health conditions, the potential use of health information in artificial intelligence is even more critical. Therefore, there is a clear need to develop a system that facilitates communication and access to health information for students with ID/DD, one that provides information to caregivers and education providers, limits errors, and, therefore, improves these individuals' education and quality of life.

Practical implications

This review highlights the gap in the current literature regarding using artificial intelligence in supporting the education of students with ID/DD. There is an urgent need for an intelligent system in collaboration with the updated health information to improve the quality of services submitted for people with intellectual disabilities and as a result improving their quality of life.

Originality/value

This study contributes to the literature by highlighting the gaps in incorporating artificial intelligence and its service to individuals with ID/DD. The research additionally proposes a solution based on the confounding variables of students’ health and individual characteristics. This solution will provide an automated information flow as a functional diagnostic and intervention tool for teachers, caregivers and parents. It could potentially improve the educational and practical outcomes for individuals with ID/DD and, ultimately, their quality of life.

Deep learning algorithms for rotating machinery intelligent diagnosis: An open source benchmark study

Rotating machinery intelligent diagnosis based on deep learning (DL) has gone through tremendous progress, which can help reduce costly breakdowns. However, different datasets and hyper-parameters are recommended to be used, and few open source codes are publicly available, resulting in unfair comparisons and ineffective improvement. To address these issues, we perform a comprehensive evaluation of four models, including multi-layer perception (MLP), auto-encoder (AE), convolutional neural network (CNN), and recurrent neural network (RNN), with seven datasets to provide a benchmark study. We first gather nine publicly available datasets and give a comprehensive benchmark study of DL-based models with two data split strategies, five input formats, three normalization methods, and four augmentation methods. Second, we integrate the whole evaluation codes into a code library and release it to the public for better comparisons. Third, we use specific-designed cases to point out the existing issues, including class imbalance, generalization ability, interpretability, few-shot learning, and model selection. Finally, we release a unified code framework for comparing and testing models fairly and quickly, emphasize the importance of open source codes, provide the baseline accuracy (a lower bound), and discuss existing issues in this field. The code library is available at: https://github.com/ZhaoZhibin/DL-based-Intelligent-Diagnosis-Benchmark.

Reducing Cognitive Load and Improving Warfighter Problem Solving With Intelligent Virtual Assistants

Recent times have seen increasing interest in conversational assistants (e.g., Amazon Alexa) designed to help users in their daily tasks. In military settings, it is critical to design assistants that are, simultaneously, helpful and able to minimize the user's cognitive load. Here, we show that embodiment plays a key role in achieving that goal. We present an experiment where participants engaged in an augmented reality version of the relatively well-known desert survival task. Participants were paired with a voice assistant, an embodied assistant, or no assistant. The assistants made suggestions verbally throughout the task, whereas the embodied assistant further used gestures and emotion to communicate with the user. Our results indicate that both assistant conditions led to higher performance over the no assistant condition, but the embodied assistant achieved this with less cognitive burden on the decision maker than the voice assistant, which is a novel contribution. We discuss implications for the design of intelligent collaborative systems for the warfighter.