Showing papers in "Frontiers in artificial intelligence in 2022"

Detection of COVID-19 Using Deep Learning Techniques and Cost Effectiveness Evaluation: A Survey

Abstract: Graphical-design-based symptomatic techniques in pandemics perform a quintessential purpose in screening hit causes that comparatively render better outcomes amongst the principal radioscopy mechanisms in recognizing and diagnosing COVID-19 cases. The deep learning paradigm has been applied vastly to investigate radiographic images such as Chest X-Rays (CXR) and CT scan images. These radiographic images are rich in information such as patterns and clusters like structures, which are evident in conformance and detection of COVID-19 like pandemics. This paper aims to comprehensively study and analyze detection methodology based on Deep learning techniques for COVID-19 diagnosis. Deep learning technology is a good, practical, and affordable modality that can be deemed a reliable technique for adequately diagnosing the COVID-19 virus. Furthermore, the research determines the potential to enhance image character through artificial intelligence and distinguishes the most inexpensive and most trustworthy imaging method to anticipate dreadful viruses. This paper further discusses the cost-effectiveness of the surveyed methods for detecting COVID-19, in contrast with the other methods. Several finance-related aspects of COVID-19 detection effectiveness of different methods used for COVID-19 detection have been discussed. Overall, this study presents an overview of COVID-19 detection using deep learning methods and their cost-effectiveness and financial implications from the perspective of insurance claim settlement.

Leveraging explanations in interactive machine learning: An overview

Abstract: Explanations have gained an increasing level of interest in the AI and Machine Learning (ML) communities in order to improve model transparency and allow users to form a mental model of a trained ML model. However, explanations can go beyond this one way communication as a mechanism to elicit user control, because once users understand, they can then provide feedback. The goal of this paper is to present an overview of research where explanations are combined with interactive capabilities as a mean to learn new models from scratch and to edit and debug existing ones. To this end, we draw a conceptual map of the state-of-the-art, grouping relevant approaches based on their intended purpose and on how they structure the interaction, highlighting similarities and differences between them. We also discuss open research issues and outline possible directions forward, with the hope of spurring further research on this blooming research topic.

Transfer Learning Approaches for Neuroimaging Analysis: A Scoping Review

Abstract: Deep learning algorithms have been moderately successful in diagnoses of diseases by analyzing medical images especially through neuroimaging that is rich in annotated data. Transfer learning methods have demonstrated strong performance in tackling annotated data. It utilizes and transfers knowledge learned from a source domain to target domain even when the dataset is small. There are multiple approaches to transfer learning that result in a range of performance estimates in diagnosis, detection, and classification of clinical problems. Therefore, in this paper, we reviewed transfer learning approaches, their design attributes, and their applications to neuroimaging problems. We reviewed two main literature databases and included the most relevant studies using predefined inclusion criteria. Among 50 reviewed studies, more than half of them are on transfer learning for Alzheimer's disease. Brain mapping and brain tumor detection were second and third most discussed research problems, respectively. The most common source dataset for transfer learning was ImageNet, which is not a neuroimaging dataset. This suggests that the majority of studies preferred pre-trained models instead of training their own model on a neuroimaging dataset. Although, about one third of studies designed their own architecture, most studies used existing Convolutional Neural Network architectures. Magnetic Resonance Imaging was the most common imaging modality. In almost all studies, transfer learning contributed to better performance in diagnosis, classification, segmentation of different neuroimaging diseases and problems, than methods without transfer learning. Among different transfer learning approaches, fine-tuning all convolutional and fully-connected layers approach and freezing convolutional layers and fine-tuning fully-connected layers approach demonstrated superior performance in terms of accuracy. These recent transfer learning approaches not only show great performance but also require less computational resources and time.

Lexical simplification benchmarks for English, Portuguese, and Spanish

Abstract: Even in highly-developed countries, as many as 15–30% of the population can only understand texts written using a basic vocabulary. Their understanding of everyday texts is limited, which prevents them from taking an active role in society and making informed decisions regarding healthcare, legal representation, or democratic choice. Lexical simplification is a natural language processing task that aims to make text understandable to everyone by replacing complex vocabulary and expressions with simpler ones, while preserving the original meaning. It has attracted considerable attention in the last 20 years, and fully automatic lexical simplification systems have been proposed for various languages. The main obstacle for the progress of the field is the absence of high-quality datasets for building and evaluating lexical simplification systems. In this study, we present a new benchmark dataset for lexical simplification in English, Spanish, and (Brazilian) Portuguese, and provide details about data selection and annotation procedures, to enable compilation of comparable datasets in other languages and domains. As the first multilingual lexical simplification dataset, where instances in all three languages were selected and annotated using comparable procedures, this is the first dataset that offers a direct comparison of lexical simplification systems for three languages. To showcase the usability of the dataset, we adapt two state-of-the-art lexical simplification systems with differing architectures (neural vs. non-neural) to all three languages (English, Spanish, and Brazilian Portuguese) and evaluate their performances on our new dataset. For a fairer comparison, we use several evaluation measures which capture varied aspects of the systems' efficacy, and discuss their strengths and weaknesses. We find that a state-of-the-art neural lexical simplification system outperforms a state-of-the-art non-neural lexical simplification system in all three languages, according to all evaluation measures. More importantly, we find that the state-of-the-art neural lexical simplification systems perform significantly better for English than for Spanish and Portuguese, thus posing a question if such an architecture can be used for successful lexical simplification in other languages, especially the low-resourced ones.

Inference-Optimized AI and High Performance Computing for Gravitational Wave Detection at Scale

Abstract: We introduce an ensemble of artificial intelligence models for gravitational wave detection that we trained in the Summit supercomputer using 32 nodes, equivalent to 192 NVIDIA V100 GPUs, within 2 h. Once fully trained, we optimized these models for accelerated inference using NVIDIA TensorRT. We deployed our inference-optimized AI ensemble in the ThetaGPU supercomputer at Argonne Leadership Computer Facility to conduct distributed inference. Using the entire ThetaGPU supercomputer, consisting of 20 nodes each of which has 8 NVIDIA A100 Tensor Core GPUs and 2 AMD Rome CPUs, our NVIDIA TensorRT-optimized AI ensemble processed an entire month of advanced LIGO data (including Hanford and Livingston data streams) within 50 s. Our inference-optimized AI ensemble retains the same sensitivity of traditional AI models, namely, it identifies all known binary black hole mergers previously identified in this advanced LIGO dataset and reports no misclassifications, while also providing a 3X inference speedup compared to traditional artificial intelligence models. We used time slides to quantify the performance of our AI ensemble to process up to 5 years worth of advanced LIGO data. In this synthetically enhanced dataset, our AI ensemble reports an average of one misclassification for every month of searched advanced LIGO data. We also present the receiver operating characteristic curve of our AI ensemble using this 5 year long advanced LIGO dataset. This approach provides the required tools to conduct accelerated, AI-driven gravitational wave detection at scale.

Nonlinear Noise Cleaning in Gravitational-Wave Detectors With Convolutional Neural Networks

Abstract: Currently, the sub-60 Hz sensitivity of gravitational-wave (GW) detectors like Advanced LIGO (aLIGO) is limited by the control noises from auxiliary degrees of freedom which nonlinearly couple to the main GW readout. One promising way to tackle this challenge is to perform nonlinear noise mitigation using convolutional neural networks (CNNs), which we examine in detail in this study. In many cases, the noise coupling is bilinear and can be viewed as a few fast channels' outputs modulated by some slow channels. We show that we can utilize this knowledge of the physical system and adopt an explicit “slow×fast” structure in the design of the CNN to enhance its performance of noise subtraction. We then examine the requirements in the signal-to-noise ratio (SNR) in both the target channel (i.e., the main GW readout) and in the auxiliary sensors in order to reduce the noise by at least a factor of a few. In the case of limited SNR in the target channel, we further demonstrate that the CNN can still reach a good performance if we use curriculum learning techniques, which in reality can be achieved by combining data from quiet times and those from periods with active noise injections.

Scaling and Disagreements: Bias, Noise, and Ambiguity

Abstract: Crowdsourced data are often rife with disagreement, either because of genuine item ambiguity, overlapping labels, subjectivity, or annotator error. Hence, a variety of methods have been developed for learning from data containing disagreement. One of the observations emerging from this work is that different methods appear to work best depending on characteristics of the dataset such as the level of noise. In this paper, we investigate the use of an approach developed to estimate noise, temperature scaling, in learning from data containing disagreements. We find that temperature scaling works with data in which the disagreements are the result of label overlap, but not with data in which the disagreements are due to annotator bias, as in, e.g., subjective tasks such as labeling an item as offensive or not. We also find that disagreements due to ambiguity do not fit perfectly either category.

Segmentation of Lung Nodules on CT Images Using a Nested Three-Dimensional Fully Connected Convolutional Network

Abstract: In computer-aided diagnosis systems for lung cancer, segmentation of lung nodules is important for analyzing image features of lung nodules on computed tomography (CT) images and distinguishing malignant nodules from benign ones. However, it is difficult to accurately and robustly segment lung nodules attached to the chest wall or with ground-glass opacities using conventional image processing methods. Therefore, this study aimed to develop a method for robust and accurate three-dimensional (3D) segmentation of lung nodule regions using deep learning. In this study, a nested 3D fully connected convolutional network with residual unit structures was proposed, and designed a new loss function. Compared with annotated images obtained under the guidance of a radiologist, the Dice similarity coefficient (DS) and intersection over union (IoU) were 0.845 ± 0.008 and 0.738 ± 0.011, respectively, for 332 lung nodules (lung adenocarcinoma) obtained from 332 patients. On the other hand, for 3D U-Net and 3D SegNet, the DS was 0.822 ± 0.009 and 0.786 ± 0.011, respectively, and the IoU was 0.711 ± 0.011 and 0.660 ± 0.012, respectively. These results indicate that the proposed method is significantly superior to well-known deep learning models. Moreover, we compared the results obtained from the proposed method with those obtained from conventional image processing methods, watersheds, and graph cuts. The DS and IoU results for the watershed method were 0.628 ± 0.027 and 0.494 ± 0.025, respectively, and those for the graph cut method were 0.566 ± 0.025 and 0.414 ± 0.021, respectively. These results indicate that the proposed method is significantly superior to conventional image processing methods. The proposed method may be useful for accurate and robust segmentation of lung nodules to assist radiologists in the diagnosis of lung nodules such as lung adenocarcinoma on CT images.

Neural Network Training With Asymmetric Crosspoint Elements

Abstract: Analog crossbar arrays comprising programmable non-volatile resistors are under intense investigation for acceleration of deep neural network training. However, the ubiquitous asymmetric conductance modulation of practical resistive devices critically degrades the classification performance of networks trained with conventional algorithms. Here we first describe the fundamental reasons behind this incompatibility. Then, we explain the theoretical underpinnings of a novel fully-parallel training algorithm that is compatible with asymmetric crosspoint elements. By establishing a powerful analogy with classical mechanics, we explain how device asymmetry can be exploited as a useful feature for analog deep learning processors. Instead of conventionally tuning weights in the direction of the error function gradient, network parameters can be programmed to successfully minimize the total energy (Hamiltonian) of the system that incorporates the effects of device asymmetry. Our technique enables immediate realization of analog deep learning accelerators based on readily available device technologies.

Supporting Artificial Social Intelligence With Theory of Mind

Abstract: In this paper, we discuss the development of artificial theory of mind as foundational to an agent's ability to collaborate with human team members. Agents imbued with artificial social intelligence will require various capabilities to gather the social data needed to inform an artificial theory of mind of their human counterparts. We draw from social signals theorizing and discuss a framework to guide consideration of core features of artificial social intelligence. We discuss how human social intelligence, and the development of theory of mind, can contribute to the development of artificial social intelligence by forming a foundation on which to help agents model, interpret and predict the behaviors and mental states of humans to support human-agent interaction. Artificial social intelligence will need the processing capabilities to perceive, interpret, and generate combinations of social cues to operate within a human-agent team. Artificial Theory of Mind affords a structure by which a socially intelligent agent could be imbued with the ability to model their human counterparts and engage in effective human-agent interaction. Further, modeling Artificial Theory of Mind can be used by an ASI to support transparent communication with humans, improving trust in agents, so that they may better predict future system behavior based on their understanding of and support trust in artificial socially intelligent agents.

Stock Price Forecasting by a Deep Convolutional Generative Adversarial Network

Abstract: Stock market prices are known to be very volatile and noisy, and their accurate forecasting is a challenging problem. Traditionally, both linear and non-linear methods (such as ARIMA and LSTM) have been proposed and successfully applied to stock market prediction, but there is room to develop models that further reduce the forecast error. In this paper, we introduce a Deep Convolutional Generative Adversarial Network (DCGAN) architecture to deal with the problem of forecasting the closing price of stocks. To test the empirical performance of our proposed model we use the FTSE MIB (Financial Times Stock Exchange Milano Indice di Borsa), the benchmark stock market index for the Italian national stock exchange. By conducting both single-step and multi-step forecasting, we observe that our proposed model performs better than standard widely used tools, suggesting that Deep Learning (and in particular GANs) is a promising field for financial time series forecasting.

Knowledge, attitude, and practice of artificial intelligence among doctors and medical students in Syria: A cross-sectional online survey

Abstract: Artificial intelligence has been prevalent recently as its use in the medical field is noticed to be increased. However, middle east countries like Syria are deficient in multiple AI implementation methods in the field of medicine. So, holding these AI implementation methods in the medical field is necessary, which may be incredibly beneficial for making diagnosis more accessible and help in the treatment. This paper intends to determine AI's knowledge, attitude, and practice among doctors and medical students in Syria. A questionnaire conducted an online cross-sectional study on the google form website consisting of demographic data, knowledge, and perception of AI. There were 1,494 responses from both doctors and medical students. We included Syrian medical students and doctors who are currently residing in Syria. Of the 1,494 participants, 255 (16.9%) are doctors, while the other 1,252 (83.1%) are undergraduate medical students. About 1,055 (70%) participants have previous knowledge about AI. However, only 357 (23.7%) participants know about its application in the medical field. Most have shown positive attitudes toward its necessity in the medical field; 689 (45.7%) individuals strongly agree, and 628 (41.7%) agree. The undergraduate students had 3.327 times more adequate knowledge of AI than students in the first year. In contrast, the undergraduate 6th-year students had 2.868 times the attitude toward AI higher than students in the first year. The residents and assistant professors had 2.371 and 4.422 times the practice of AI higher than students, respectively. Although most physicians and medical students do not sufficiently understand AI and its significance in the medical field, they have favorable views regarding using AI in the medical field. Syrian medical authorities and international organizations should suggest including artificial intelligence in the medical field, particularly when training residents and fellowship physicians.

Automated quantification of penile curvature using artificial intelligence

Abstract: Objective To develop and validate an artificial intelligence (AI)-based algorithm for capturing automated measurements of Penile curvature (PC) based on 2-dimensional images. Materials and methods Nine 3D-printed penile models with differing curvature angles (ranging from 18 to 88°) were used to compile a 900-image dataset featuring multiple camera positions, inclination angles, and background/lighting conditions. The proposed framework of PC angle estimation consisted of three stages: automatic penile area localization, shaft segmentation, and curvature angle estimation. The penile model images were captured using a smartphone camera and used to train and test a Yolov5 model that automatically cropped the penile area from each image. Next, an Unet-based segmentation model was trained, validated, and tested to segment the penile shaft, before a custom Hough-Transform-based angle estimation technique was used to evaluate degree of PC. Results The proposed framework displayed robust performance in cropping the penile area [mean average precision (mAP) 99.4%] and segmenting the shaft [Dice Similarity Coefficient (DSC) 98.4%]. Curvature angle estimation technique generally demonstrated excellent performance, with a mean absolute error (MAE) of just 8.5 when compared with ground truth curvature angles. Conclusions Considering current intra- and inter-surgeon variability of PC assessments, the framework reported here could significantly improve precision of PC measurements by surgeons and hypospadiology researchers.

Theory of Mind and Preference Learning at the Interface of Cognitive Science, Neuroscience, and AI: A Review

Abstract: Theory of Mind (ToM)—the ability of the human mind to attribute mental states to others—is a key component of human cognition. In order to understand other people's mental states or viewpoint and to have successful interactions with others within social and occupational environments, this form of social cognition is essential. The same capability of inferring human mental states is a prerequisite for artificial intelligence (AI) to be integrated into society, for example in healthcare and the motoring industry. Autonomous cars will need to be able to infer the mental states of human drivers and pedestrians to predict their behavior. In the literature, there has been an increasing understanding of ToM, specifically with increasing cognitive science studies in children and in individuals with Autism Spectrum Disorder. Similarly, with neuroimaging studies there is now a better understanding of the neural mechanisms that underlie ToM. In addition, new AI algorithms for inferring human mental states have been proposed with more complex applications and better generalisability. In this review, we synthesize the existing understanding of ToM in cognitive and neurosciences and the AI computational models that have been proposed. We focus on preference learning as an area of particular interest and the most recent neurocognitive and computational ToM models. We also discuss the limitations of existing models and hint at potential approaches to allow ToM models to fully express the complexity of the human mind in all its aspects, including values and preferences.

On the Construction of Group Equivariant Non-Expansive Operators via Permutants and Symmetric Functions

Abstract: Group Equivariant Operators (GEOs) are a fundamental tool in the research on neural networks, since they make available a new kind of geometric knowledge engineering for deep learning, which can exploit symmetries in artificial intelligence and reduce the number of parameters required in the learning process. In this paper we introduce a new method to build non-linear GEOs and non-linear Group Equivariant Non-Expansive Operators (GENEOs), based on the concepts of symmetric function and permutant. This method is particularly interesting because of the good theoretical properties of GENEOs and the ease of use of permutants to build equivariant operators, compared to the direct use of the equivariance groups we are interested in. In our paper, we prove that the technique we propose works for any symmetric function, and benefits from the approximability of continuous symmetric functions by symmetric polynomials. A possible use in Topological Data Analysis of the GENEOs obtained by this new method is illustrated.

Generative Models of Brain Dynamics

Abstract: This review article gives a high-level overview of the approaches across different scales of organization and levels of abstraction. The studies covered in this paper include fundamental models in computational neuroscience, nonlinear dynamics, data-driven methods, as well as emergent practices. While not all of these models span the intersection of neuroscience, AI, and system dynamics, all of them do or can work in tandem as generative models, which, as we argue, provide superior properties for the analysis of neuroscientific data. We discuss the limitations and unique dynamical traits of brain data and the complementary need for hypothesis- and data-driven modeling. By way of conclusion, we present several hybrid generative models from recent literature in scientific machine learning, which can be efficiently deployed to yield interpretable models of neural dynamics.

The Issue of Proxies and Choice Architectures. Why EU Law Matters for Recommender Systems

Abstract: Recommendations are meant to increase sales or ad revenue, as these are the first priority of those who pay for them. As recommender systems match their recommendations with inferred preferences, we should not be surprised if the algorithm optimizes for lucrative preferences and thus co-produces the preferences they mine. This relates to the well-known problems of feedback loops, filter bubbles, and echo chambers. In this article, I discuss the implications of the fact that computing systems necessarily work with proxies when inferring recommendations and raise a number of questions about whether recommender systems actually do what they are claimed to do, while also analysing the often-perverse economic incentive structures that have a major impact on relevant design decisions. Finally, I will explain how the choice architectures for data controllers and providers of AI systems as foreseen in the EU's General Data Protection Regulation (GDPR), the proposed EU Digital Services Act (DSA) and the proposed EU AI Act will help to break through various vicious circles, by constraining how people may be targeted (GDPR, DSA) and by requiring documented evidence of the robustness, resilience, reliability, and the responsible design and deployment of high-risk recommender systems (AI Act).

An Overview of Alphafold's Breakthrough

Abstract: This paper presents a short summary of the protein folding problem, what it is and why it is significant. Introduces the CASP competition and how accuracy is measured. Looks at different approaches for solving the problem followed by a review of the current breakthroughs in the field introduced by AlphaFold 1 and AlphaFold 2.

IRC-Safe Graph Autoencoder for Unsupervised Anomaly Detection

Abstract: Anomaly detection through employing machine learning techniques has emerged as a novel powerful tool in the search for new physics beyond the Standard Model. Historically similar to the development of jet observables, theoretical consistency has not always assumed a central role in the fast development of algorithms and neural network architectures. In this work, we construct an infrared and collinear safe autoencoder based on graph neural networks by employing energy-weighted message passing. We demonstrate that whilst this approach has theoretically favorable properties, it also exhibits formidable sensitivity to non-QCD structures.

AI Technologies, Privacy, and Security

Abstract: Privacy remains one of the most recurrent concerns that people have about AI technologies. The meaning of the concept of “privacy” has proven to be fairly elusive. Accordingly, the concerns people have about privacy are often vague and ill-formed, which makes it correspondingly difficult to address these concerns, and to explain the ways in which AI technologies do or do not pose threats to people's interests. In this article, we draw attention to some important distinctions that are frequently overlooked, and spell out their implications for concerns about the threats that AI-related technology poses for privacy. We argue that, when people express concerns about privacy in relation to AI technologies, they are usually referring to security interests rather than interests in privacy per se. Nevertheless, we argue that focusing primarily on security interests misses the importance that interests in privacy per se have through their contribution to autonomy and the development of our identities. Improving insight about these issues can make it easier for the developers of AI technologies to provide explanations for users about what interests are and are not at stake through the use of AI systems.

Keep Calm and Do Not Carry-Forward: Toward Sensor-Data Driven AI Agent to Enhance Human Learning

Abstract: The integration of Multimodal Data (MMD) and embodied learning systems (such as Motion Based Educational Games, MBEG), can help learning researchers to better understand the synergy between students' interactions and their learning experiences. Unfolding the dynamics behind this important synergy can lead to the design of intelligent agents which leverage students' movements and support their learning. However, real-time use of student-generated MMD derived from their interactions with embodied learning systems (MBEG in our case) is challenging and remains under-explored due to its complexity (e.g., handle sensor-data and enable an AI agent to use them). To bridge this gap, we conducted an in-situ study where 40 children, aged 9–12, played MBEG on maths and language development. We automatically, unobtrusively, and continuously monitored students' experiences using eye-tracking glasses, physiological wristbands, and Kinect, during game-play. This allowed us to understand the different cognitive and physiological dimensions of students' progress (right/wrong responses) during the three different stages of the MBEG problem-solving processes, namely the “see-solve-move-respond” (S2MR) cycle. We introduce the novel Carry Forward Effect (CFE); a phenomenon occurring in such games, whereby students propagate, or “carry forward,” the cognitive and physiological effects derived from their MMD, to subsequent phases in the see-solve-move-respond cycle. By identifying moments when the Carry Forward Effect is congruent (or not) to students' learning performance, we uncover opportunities for feedback delivery to encourage or subdue the impact of the CFE. Our results demonstrate the importance of wristband and eye-tracking data as key indicators for prioritizing adaptive feedback to support students in MBEG and emphasize the significance of using MMD to support students' performance in real-time educational settings.

High Resolution, Annual Maps of Field Boundaries for Smallholder-Dominated Croplands at National Scales

Abstract: Mapping the characteristics of Africa’s smallholder-dominated croplands, including the sizes and numbers of fields, can provide critical insights into food security and a range of other socioeconomic and environmental concerns. However, accurately mapping these systems is difficult because there is 1) a spatial and temporal mismatch between satellite sensors and smallholder fields, and 2) a lack of high-quality labels needed to train and assess machine learning classifiers. We developed an approach designed to address these two problems, and used it to map Ghana’s croplands. To overcome the spatio-temporal mismatch, we converted daily, high resolution imagery into two cloud-free composites (the primary growing season and subsequent dry season) covering the 2018 agricultural year, providing a seasonal contrast that helps to improve classification accuracy. To address the problem of label availability, we created a platform that rigorously assesses and minimizes label error, and used it to iteratively train a Random Forests classifier with active learning, which identifies the most informative training sample based on prediction uncertainty. Minimizing label errors improved model F1 scores by up to 25%. Active learning increased F1 scores by an average of 9.1% between first and last training iterations, and 2.3% more than models trained with randomly selected labels. We used the resulting 3.7 m map of cropland probabilities within a segmentation algorithm to delineate crop field boundaries. Using an independent map reference sample (n = 1,207), we found that the cropland probability and field boundary maps had respective overall accuracies of 88 and 86.7%, user’s accuracies for the cropland class of 61.2 and 78.9%, and producer’s accuracies of 67.3 and 58.2%. An unbiased area estimate calculated from the map reference sample indicates that cropland covers 17.1% (15.4–18.9%) of Ghana. Using the most accurate validation labels to correct for biases in the segmented field boundaries map, we estimated that the average size and total number of field in Ghana are 1.73 ha and 1,662,281, respectively. Our results demonstrate an adaptable and transferable approach for developing annual, country-scale maps of crop field boundaries, with several features that effectively mitigate the errors inherent in remote sensing of smallholder-dominated agriculture.

TB-Net: A Tailored, Self-Attention Deep Convolutional Neural Network Design for Detection of Tuberculosis Cases From Chest X-Ray Images

Abstract: Tuberculosis (TB) remains a global health problem, and is the leading cause of death from an infectious disease. A crucial step in the treatment of tuberculosis is screening high risk populations and the early detection of the disease, with chest x-ray (CXR) imaging being the most widely-used imaging modality. As such, there has been significant recent interest in artificial intelligence-based TB screening solutions for use in resource-limited scenarios where there is a lack of trained healthcare workers with expertise in CXR interpretation. Motivated by this pressing need and the recent recommendation by the World Health Organization (WHO) for the use of computer-aided diagnosis of TB in place of a human reader, we introduce TB-Net, a self-attention deep convolutional neural network tailored for TB case screening. We used CXR data from a multi-national patient cohort to train and test our models. A machine-driven design exploration approach leveraging generative synthesis was used to build a highly customized deep neural network architecture with attention condensers. We conducted an explainability-driven performance validation process to validate TB-Net's decision-making behavior. Experiments on CXR data from a multi-national patient cohort showed that the proposed TB-Net is able to achieve accuracy/sensitivity/specificity of 99.86/100.0/99.71%. Radiologist validation was conducted on select cases by two board-certified radiologists with over 10 and 19 years of experience, respectively, and showed consistency between radiologist interpretation and critical factors leveraged by TB-Net for TB case detection for the case where radiologists identified anomalies. The proposed TB-Net not only achieves high tuberculosis case detection performance in terms of sensitivity and specificity, but also leverages clinically relevant critical factors in its decision making process. While not a production-ready solution, we hope that the open-source release of TB-Net as part of the COVID-Net initiative will support researchers, clinicians, and citizen data scientists in advancing this field in the fight against this global public health crisis.

Artificial Intelligence and Employment: New Cross-Country Evidence

Abstract: Recent years have seen impressive advances in artificial intelligence (AI) and this has stoked renewed concern about the impact of technological progress on the labor market, including on worker displacement. This paper looks at the possible links between AI and employment in a cross-country context. It adapts the AI occupational impact measure developed by Felten, Raj and Seamans—an indicator measuring the degree to which occupations rely on abilities in which AI has made the most progress—and extends it to 23 OECD countries. Overall, there appears to be no clear relationship between AI exposure and employment growth. However, in occupations where computer use is high, greater exposure to AI is linked to higher employment growth. The paper also finds suggestive evidence of a negative relationship between AI exposure and growth in average hours worked among occupations where computer use is low. One possible explanation is that partial automation by AI increases productivity directly as well as by shifting the task composition of occupations toward higher value-added tasks. This increase in labor productivity and output counteracts the direct displacement effect of automation through AI for workers with good digital skills, who may find it easier to use AI effectively and shift to non-automatable, higher-value added tasks within their occupations. The opposite could be true for workers with poor digital skills, who may not be able to interact efficiently with AI and thus reap all potential benefits of the technology1.

The interactive reading task: Transformer-based automatic item generation

Abstract: Automatic item generation (AIG) has the potential to greatly expand the number of items for educational assessments, while simultaneously allowing for a more construct-driven approach to item development. However, the traditional item modeling approach in AIG is limited in scope to content areas that are relatively easy to model (such as math problems), and depends on highly skilled content experts to create each model. In this paper we describe the interactive reading task, a transformer-based deep language modeling approach for creating reading comprehension assessments. This approach allows a fully automated process for the creation of source passages together with a wide range of comprehension questions about the passages. The format of the questions allows automatic scoring of responses with high fidelity (e.g., selected response questions). We present the results of a large-scale pilot of the interactive reading task, with hundreds of passages and thousands of questions. These passages were administered as part of the practice test of the Duolingo English Test. Human review of the materials and psychometric analyses of test taker results demonstrate the feasibility of this approach for automatic creation of complex educational assessments.

Improving Robotic Hand Prosthesis Control With Eye Tracking and Computer Vision: A Multimodal Approach Based on the Visuomotor Behavior of Grasping

Abstract: The complexity and dexterity of the human hand make the development of natural and robust control of hand prostheses challenging. Although a large number of control approaches were developed and investigated in the last decades, limited robustness in real-life conditions often prevented their application in clinical settings and in commercial products. In this paper, we investigate a multimodal approach that exploits the use of eye-hand coordination to improve the control of myoelectric hand prostheses. The analyzed data are from the publicly available MeganePro Dataset 1, that includes multimodal data from transradial amputees and able-bodied subjects while grasping numerous household objects with ten grasp types. A continuous grasp-type classification based on surface electromyography served as both intent detector and classifier. At the same time, the information provided by eye-hand coordination parameters, gaze data and object recognition in first-person videos allowed to identify the object a person aims to grasp. The results show that the inclusion of visual information significantly increases the average offline classification accuracy by up to 15.61 ± 4.22% for the transradial amputees and of up to 7.37 ± 3.52% for the able-bodied subjects, allowing trans-radial amputees to reach average classification accuracy comparable to intact subjects and suggesting that the robustness of hand prosthesis control based on grasp-type recognition can be significantly improved with the inclusion of visual information extracted by leveraging natural eye-hand coordination behavior and without placing additional cognitive burden on the user.

Semantic Integration of Multi-Modal Data and Derived Neuroimaging Results Using the Platform for Imaging in Precision Medicine (PRISM) in the Arkansas Imaging Enterprise System (ARIES)

Abstract: Neuroimaging is among the most active research domains for the creation and management of open-access data repositories. Notably lacking from most data repositories are integrated capabilities for semantic representation. The Arkansas Imaging Enterprise System (ARIES) is a research data management system which features integrated capabilities to support semantic representations of multi-modal data from disparate sources (imaging, behavioral, or cognitive assessments), across common image-processing stages (preprocessing steps, segmentation schemes, analytic pipelines), as well as derived results (publishable findings). These unique capabilities ensure greater reproducibility of scientific findings across large-scale research projects. The current investigation was conducted with three collaborating teams who are using ARIES in a project focusing on neurodegeneration. Datasets included magnetic resonance imaging (MRI) data as well as non-imaging data obtained from a variety of assessments designed to measure neurocognitive functions (performance scores on neuropsychological tests). We integrate and manage these data with semantic representations based on axiomatically rich biomedical ontologies. These instantiate a knowledge graph that combines the data from the study cohorts into a shared semantic representation that explicitly accounts for relations among the entities that the data are about. This knowledge graph is stored in a triple-store database that supports reasoning over and querying these integrated data. Semantic integration of the non-imaging data using background information encoded in biomedical domain ontologies has served as a key feature-engineering step, allowing us to combine disparate data and apply analyses to explore associations, for instance, between hippocampal volumes and measures of cognitive functions derived from various assessment instruments.

Improving plant disease classification by adaptive minimal ensembling

Abstract: A novel method for improving plant disease classification, a challenging and time-consuming process, is proposed. First, using as baseline EfficientNet, a recent and advanced family of architectures having an excellent accuracy/complexity trade-off, we have introduced, devised, and applied refined techniques based on transfer learning, regularization, stratification, weighted metrics, and advanced optimizers in order to achieve improved performance. Then, we go further by introducing adaptive minimal ensembling, which is a unique input to the knowledge base of the proposed solution. This represents a leap forward since it allows improving the accuracy with limited complexity using only two EfficientNet-b0 weak models, performing ensembling on feature vectors by a trainable layer instead of classic aggregation on outputs. To the best of our knowledge, such an approach to ensembling has never been used before in literature. Our method was tested on PlantVillage, a public reference dataset used for benchmarking models' performances for crop disease diagnostic, considering both its original and augmented versions. We noticeably improved the state of the art by achieving 100% accuracy in both the original and augmented datasets. Results were obtained using PyTorch to train, test, and validate the models; reproducibility is granted by providing exhaustive details, including hyperparameters used in the experimentation. A Web interface is also made publicly available to test the proposed methods.

Models of Language and Multiword Expressions

Abstract: Traditional accounts of language postulate two basic components: words stored in a lexicon, and rules that govern how they can be combined into meaningful sentences, a grammar. But, although this words-and-rules framework has proven itself to be useful in natural language processing and cognitive science, it has also shown important shortcomings when faced with actual language use. In this article, we review evidence from language acquisition, sentence processing, and computational modeling that shows how multiword expressions such as idioms, collocations, and other meaningful and common units that comprise more than one word play a key role in the organization of our linguistic knowledge. Importantly, multiword expressions straddle the line between lexicon and grammar, calling into question how useful this distinction is as a foundation for our understanding of language. Nonetheless, finding a replacement for the foundational role the words-and-rules approach has played in our theories is not straightforward. Thus, the second part of our article reviews and synthesizes the diverse approaches that have attempted to account for the central role of multiword expressions in language representation, acquisition, and processing.

Cutting-edge communication and learning assistive technologies for disabled children: An artificial intelligence perspective

Abstract: In this study we provide an in-depth review and analysis of the impact of artificial intelligence (AI) components and solutions that support the development of cutting-edge assistive technologies for children with special needs. Various disabilities are addressed and the most recent assistive technologies that enhance communication and education of disabled children, as well as the AI technologies that have enabled their development, are presented. The paper summarizes with an AI perspective on future assistive technologies and ethical concerns arising from the use of such cutting-edge communication and learning technologies for children with disabilities.