Artificial Intelligence Paradigm Research Articles

Data-Centric Artificial Intelligence, Preprocessing, and the Quest for Transformative Artificial Intelligence Systems Development

Some data-centric artificial intelligence (AI) practices often resemble the preprocessing used in traditional AI systems; therefore, there is a possible misconception about the two approaches. This article differentiates these two aspects to guide the AI community to unlock the hidden potential of the data-centric AI paradigm.

A new paradigm for artificial intelligence based on group equivariant non-expansive operators

The recent frantic surge of machine learning and, more broadly, of artificial intelligence (AI) brings to light old and new open issues, and among them, the so-called eXplainable artificial intelligence (XAI) – AI that humans can understand – as opposed to black-box learning systems where even their designers cannot explain AI decisions. One of the major XAI questions is how to design transparent learning systems that incorporate prior knowledge. These topics are becoming more relevant and pervasive as AI systems become more unfathomable and entangled with human factors. Recently a new paradigm for XAI has been introduced in literature, based on group equivariant non-expansive operators (GENEOs), which are able to inject prior knowledge in a learning system. Hence, the use of GENEOs dramatically reduces the number of unknown parameters to be identified and the size of the related training set, providing both computational advantages and an increased degree of interpretability of the results. Here we will illustrate the main characteristics of GENEOs and the encouraging results already obtained on a couple of industrial case studies.

Foundation models for generalist medical artificial intelligence.

The exceptionally rapid development of highly flexible, reusable artificial intelligence (AI) models is likely to usher in newfound capabilities in medicine. We propose a new paradigm for medical AI, which we refer to as generalist medical AI (GMAI). GMAI models will be capable of carrying out a diverse set of tasks using very little or no task-specific labelled data. Built through self-supervision on large, diverse datasets, GMAI will flexibly interpret different combinations of medical modalities, including data from imaging, electronic health records, laboratory results, genomics, graphs or medical text. Models will in turn produce expressive outputs such as free-text explanations, spoken recommendations or image annotations that demonstrate advanced medical reasoning abilities. Here we identify a set of high-impact potential applications for GMAI and lay out specific technical capabilities and training datasets necessary to enable them. We expect that GMAI-enabled applications will challenge current strategies for regulating and validating AI devices for medicine and will shift practices associated with the collection of large medical datasets.

On potential limitations of differential expression analysis with non-linear machine learning models

Recently, there has been a growing interest in bioinformatics toward the adoption of increasingly complex machine learning models for the analysis of next-generation sequencing data with the goal of disease subtyping (i.e., patient stratification based on molecular features) or risk-based classification for specific endpoints, such as survival. With gene-expression data, a common approach consists in characterising the emerging groups by exploiting a differential expression analysis, which selects relevant gene sets coupled with pathway enrichment analysis, providing an insight into the underlying biological processes. However, when non-linear machine learning models are involved, differential expression analysis could be limiting since patient groupings identified by the model could be based on a set of genes that are hidden to differential expression due to its linear nature, affecting subsequent biological characterisation and validation. The aim of this study is to provide a proof-of-concept example demonstrating such a limitation. Moreover, we suggest that this issue could be overcome by the adoption of the innovative paradigm of eXplainable Artificial Intelligence, which consists in building an additional explainer to get a trustworthy interpretation of the model outputs and building a reliable set of genes characterising each group, preserving also non-linear relations, to be used for downstream analysis and validation.

Physics-Driven Synthetic Data Learning for Biomedical Magnetic Resonance: The imaging physics-based data synthesis paradigm for artificial intelligence

Deep learning (DL) has driven innovation in the field of computational imaging. One of its bottlenecks is unavailable or insufficient training data. This article reviews an emerging paradigm, imaging physics-based data synthesis (IPADS), that can provide huge training data in biomedical magnetic resonance (MR) without or with few real data. Following the physical law of MR, IPADS generates signals from differential equations or analytical solution models, making learning more scalable and explainable and better protecting privacy. Key components of IPADS learning, including signal generation models, basic DL network structures, enhanced data generation, and learning methods, are discussed. Great IPADS potential has been demonstrated by representative applications in fast imaging, ultrafast signal reconstruction, and accurate parameter quantification. Finally, open questions and future work are discussed.

Need for UAI–Anatomy of the Paradigm of Usable Artificial Intelligence for Domain-Specific AI Applicability

Data-driven methods based on artificial intelligence (AI) are powerful yet flexible tools for gathering knowledge and automating complex tasks in many areas of science and practice. Despite the rapid development of the field, the existing potential of AI methods to solve recent industrial, corporate and social challenges has not yet been fully exploited. Research shows the insufficient practicality of AI in domain-specific contexts as one of the main application hurdles. Focusing on industrial demands, this publication introduces a new paradigm in terms of applicability of AI methods, called Usable AI (UAI). Aspects of easily accessible, domain-specific AI methods are derived, which address essential user-oriented AI services within the UAI paradigm: usability, suitability, integrability and interoperability. The relevance of UAI is clarified by describing challenges, hurdles and peculiarities of AI applications in the production area, whereby the following user roles have been abstracted: developers of cyber–physical production systems (CPPS), developers of processes and operators of processes. The analysis shows that target artifacts, motivation, knowledge horizon and challenges differ for the user roles. Therefore, UAI shall enable domain- and user-role-specific adaptation of affordances accompanied by adaptive support of vertical and horizontal integration across the domains and user roles.

Federated Learning for Privacy Preservation in Smart Healthcare Systems: A Comprehensive Survey.

Recent advances in electronic devices and communication infrastructure have revolutionized the traditional healthcare system into a smart healthcare system by using internet of medical things (IoMT) devices. However, due to the centralized training approach of artificial intelligence (AI), mobile and wearable IoMT devices raise privacy issues concerning the information communicated between hospitals and end-users. The information conveyed by the IoMT devices is highly confidential and can be exposed to adversaries. In this regard, federated learning (FL), a distributive AI paradigm, has opened up new opportunities for privacy preservation in IoMT without accessing the confidential data of the participants. Further, FL provides privacy to end-users as only gradients are shared during training. For these specific properties of FL, in this paper, we present privacy-related issues in IoMT. Afterwards, we present the role of FL in IoMT networks for privacy preservation and introduce some advanced FL architectures by incorporating deep reinforcement learning (DRL), digital twin, and generative adversarial networks (GANs) for detecting privacy threats. Moreover, we present some practical opportunities for FL in IoMT. In the end, we conclude this survey by discussing open research issues and challenges while using FL in future smart healthcare systems.

Systematic Review of Machine Learning applied to the Prediction of Obesity and Overweight

Obesity and overweight has increased in the last year and has become a pandemic disease, the result of sedentary lifestyles and unhealthy diets rich in sugars, refined starches, fats and calories. Machine learning (ML) has proven to be very useful in the scientific community, especially in the health sector. With the aim of providing useful tools to help nutritionists and dieticians, research focused on the development of ML and Deep Learning (DL) algorithms and models is searched in the literature. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol has been used, a very common technique applied to carry out revisions. In our proposal, 17 articles have been filtered in which ML and DL are applied in the prediction of diseases, in the delineation of treatment strategies, in the improvement of personalized nutrition and more. Despite expecting better results with the use of DL, according to the selected investigations, the traditional methods are still the most used and the yields in both cases fluctuate around positive values, conditioned by the databases (transformed in each case) to a greater extent than by the artificial intelligence paradigm used. Conclusions: An important compilation is provided for the literature in this area. ML models are time-consuming to clean data, but (like DL) they allow automatic modeling of large volumes of data which makes them superior to traditional statistics.

Toward the third generation artificial intelligence

There have been two competing paradigms in artificial intelligence (AI) development ever since its birth in 1956, i.e., symbolism and connectionism (or sub-symbolism). While symbolism dominated AI research by the end of 1980s, connectionism gained momentum in the 1990s and is gradually displacing symbolism. This paper considers symbolism as the first generation of AI and connectionism as the second generation. However, each of these two paradigms simulates the human mind from only one perspective. AI cannot achieve true human behaviors by relying on only one paradigm. In order to develop novel AI technologies that are safe, reliable, and extensible, it is necessary to establish a new explainable and robust AI theory. To this end, this paper looks toward developing a third generation artificial intelligence by combining the current paradigms.

Sensors and machine learning and AI operation-constrained process control method for sensor-aided industrial internet of things and smart factories

The Industrial Internet of Things (IIoT) incorporates intelligent computing and artificial intelligence paradigms in smart factories for ease of production and human-less interventions. The controlling unit is integrated into the smart operations in managing, controlling, and monitoring industrial operations. This article introduces an Operation-Constrained Process Control (OCPC) for preventing time-lag errors between smart machine operation cycles. The proposed method considers the operational cycles, completion time, and output efficiency metrics for identifying time lags and errors in production. Depending on the maximum productivity-based outcomes, further allocations or modifications in the regular operational cycles are identified. The productivity is analyzed based on the previous outcomes using federated learning. This learning implies multi-sensor knowledge update and production efficiency through repeated training. Therefore, the errors in cycle assignment and production completion are synchronized under controlled error. This method is analyzed using time lag, error, production efficiency, and training instances.

A New Paradigm of Artificial Intelligence to Disabilities

A New Paradigm of Artificial Intelligence to Disabilities

Autonomous and sustainable machine learning: pursuing new horizons of intelligent systems

The paradigm of Artificial Intelligence and Machine Learning has resulted in an amazingly diverse plethora of models operating in various environments and quite often exhibiting numerous successes. There is a growing spectrum of challenging application areas of high criticality where one has to meet a number of fundamental requirements. Those manifest evidently when Machine Learning constructs have to function autonomously and any decisions being rendered entail far reaching implications. The carefully crafted learning process has to result with advanced models. Along with the developed models, they have to come hand-in-hand with credibility measures that are crucial to assess an extent to which the results generated by such measures are meaningful, trustworthy and credible. The credibility of the Machine Learning models becomes of paramount importance given the nature of application domains. Autonomous systems including autonomous vehicles, user identification (both using audio and video channels), financial systems (calling for sound mechanisms to quantify risk levels) require the ML system making classification or prediction decisions some level of self-awareness. Among others, this translates to forming sound answers to the following crucial questions emerging within the design process: How much confidence could be associated with the result? Could any action /decision be taken on a basis of obtained result? Given the reported level of credibility, is there any other experimental evidence one could acquire to validate the decision? In this study, we advocate that a general way to achieve such goals is to engage the mechanism of Granular Computing; subsequently, the granularity endowing the results are sought as a vehicle use to quantify the credibility level. Sustainable (or green) Machine Learning gives rise to the agenda of knowledge reuse, namely exploring possibilities of potential reuse of the already designed models in a spectrum of current environments where computing overhead as one of the ways to contribute to the agenda of sustainable Machine Learning and discuss a crucial role of information granularity in this context.

Repositioning the Base Level of Bibliographic Relationships: Or, a Cataloguer, a Post-Modernist and a Chatbot Walk Into a Bar

Designers and maintainers of library catalogues are facing fresh challenges representing bibliographic relationships, due both to changes in cataloguing standards and to a broader information environment that has grown increasingly diverse, sophisticated and complex. This paper presents three different paradigms, drawn from three different fields of study, for representing relationships between bibliographic entities beyond the FRBR/LRM models: superworks, as developed in information studies; adaptation, as developed in literary studies; and artificial intelligence, as developed in computer science. Theories of literary adaptation remain focused on “the work,” as traditionally conceived. The concept of the superwork reminds us that there are some works which serve as ancestors for entire families of works, and that those familial relationships are still useful. Crowd-sourcing projects often make more granular connections, a trend which has escalated significantly with current and emerging artificial intelligence systems. While the artificial intelligence paradigm is proving more pervasive outside conventional library systems, it could lead to a seismic shift in knowl­edge organization, a shift in which the power both to arrange information and to use it are moving beyond the control of users and intermediaries alike.

Polygenic Risk Score for Cardiovascular Diseases in Artificial Intelligence Paradigm: A Review.

Cardiovascular disease (CVD) related mortality and morbidity heavily strain society. The relationship between external risk factors and our genetics have not been well established. It is widely acknowledged that environmental influence and individual behaviours play a significant role in CVD vulnerability, leading to the development of polygenic risk scores (PRS). We employed the PRISMA search method to locate pertinent research and literature to extensively review artificial intelligence (AI)-based PRS models for CVD risk prediction. Furthermore, we analyzed and compared conventional vs. AI-based solutions for PRS. We summarized the recent advances in our understanding of the use of AI-based PRS for risk prediction of CVD. Our study proposes three hypotheses: i) Multiple genetic variations and risk factors can be incorporated into AI-based PRS to improve the accuracy of CVD risk predicting. ii) AI-based PRS for CVD circumvents the drawbacks of conventional PRS calculators by incorporating a larger variety of genetic and non-genetic components, allowing for more precise and individualised risk estimations. iii) Using AI approaches, it is possible to significantly reduce the dimensionality of huge genomic datasets, resulting in more accurate and effective disease risk prediction models. Our study highlighted that the AI-PRS model outperformed traditional PRS calculators in predicting CVD risk. Furthermore, using AI-based methods to calculate PRS may increase the precision of risk predictions for CVD and have significant ramifications for individualized prevention and treatment plans.

Online data poisoning attack against edge AI paradigm for IoT-enabled smart city.

The deep integration of edge computing and Artificial Intelligence (AI) in IoT (Internet of Things)-enabled smart cities has given rise to new edge AI paradigms that are more vulnerable to attacks such as data and model poisoning and evasion of attacks. This work proposes an online poisoning attack framework based on the edge AI environment of IoT-enabled smart cities, which takes into account the limited storage space and proposes a rehearsal-based buffer mechanism to manipulate the model by incrementally polluting the sample data stream that arrives at the appropriately sized cache. A maximum-gradient-based sample selection strategy is presented, which converts the operation of traversing historical sample gradients into an online iterative computation method to overcome the problem of periodic overwriting of the sample data cache after training. Additionally, a maximum-loss-based sample pollution strategy is proposed to solve the problem of each poisoning sample being updated only once in basic online attacks, transforming the bi-level optimization problem from offline mode to online mode. Finally, the proposed online gray-box poisoning attack algorithms are implemented and evaluated on edge devices of IoT-enabled smart cities using an online data stream simulated with offline open-grid datasets. The results show that the proposed method outperforms the existing baseline methods in both attack effectiveness and overhead.

Answering Topical Information Needs Using Neural Entity-Oriented Information Retrieval and Extraction

In the modern world, search engines are an integral part of human lives. The field of Information Retrieval (IR) is concerned with finding material (usually documents) of an unstructured nature (usually text) that satisfies an information need (query) from within large collections (usually stored on computers). The search engine then displays a ranked list of results relevant to our query. Traditional document retrieval algorithms match a query to a document using the overlap of words in both. However, the last decade has seen the focus shifting to leveraging the rich semantic information available in the form of entities. Entities are uniquely identifiable objects or things such as places, events, diseases, etc. that exist in the real or fictional world. Entity-oriented search systems leverage the semantic information associated with entities (e.g., names, types, etc.) to better match documents to queries. Web search engines would provide better search results if they understand the meaning of a query. This dissertation advances the state-of-the-art in IR by developing novel algorithms that understand text (query, document, question, sentence, etc.) at the semantic level. To this end, this dissertation aims to understand the fine-grained meaning of entities from the context in which the entities have been mentioned, for example, "oysters" in the context of food versus ecosystems. Further, this dissertation aims to automatically learn (vector) representations of entities that incorporate this fine-grained knowledge and knowledge about the query. This dissertation refines the automatic understanding of text passages using deep learning, a modern artificial intelligence paradigm. This dissertation utilizes the semantic information extracted from entities to retrieve materials (text and entities) relevant to a query. The interplay between text and entities in the text is studied by addressing three related prediction problems: (1) Identify entities that are relevant for the query, (2) Understand an entity's meaning in the context of the query, and (3) Identify text passages that elaborate the connection between the query and an entity. The research presented in this dissertation may be integrated into a larger system designed for answering complex topical queries such as dark chocolate health benefits which require the search engine to automatically understand the connections between the query and the relevant material, thus transforming the search engine into an answering engine. Awarded by: University of New Hampshire, Durham, USA on 1 September 2022. Supervised by: Laura Dietz. Available at: https://scholars.unh.edu/dissertation/2714/.

Novel AI Techniques for Automatic Determination of Radiation-Induced Lung Injury from Post-SBRT Images

<h3>Purpose/Objective(s)</h3> Treating lung tumors with stereotactic body radiotherapy (SBRT) unavoidably damages normal lung tissue surrounding the target tumor. As a result, physicians must closely monitor follow-up computed tomography (CT) scans for radiographic signs of radiation-induced lung injury (RILI). The differentiation between the different types of RILI, tumor recurrence, and/or tumor scar tissue can be difficult for clinicians to evaluate visually. Recent studies have shown the effectiveness of modern convolutional neural networks (CNN) for imaging analysis with artificial intelligence (AI) methods. This work explored the utilization of such AI methods with post-SBRT surveillance CT images from a database generated from three separate institutions for their ability to help clinicians recognize RILI. <h3>Materials/Methods</h3> This work utilized a database of post-SBRT follow-up CT scans which radiation oncologists had previously evaluated for RILI. Due to the relative paucity of training data, the proposed approach uses a powerful AI paradigm called self-supervised learning (SSL) to initialize a CNN on a widely available diagnostic CT database of lung cancer. Using the derived radiomic features and network weights as part of a 3D ResNet-10, a fully connected classifier layer was then trained for discriminating between RILI and normal lung tissue on 222 labeled post-SBRT CT scans of 64 subjects from three independent institutions. On each of these scans, RILI had been independently evaluated by a radiation oncologist. <h3>Results</h3> The final classification performance of the best performing CNN achieved an area under ROC curve (ROC-AUC) of 0.894 where 0.5 would be achieved by randomly guessing and 1.0 would be a perfect discrimination of RILI on the test set. The ROC-AUC score was obtained as the average ROC-AUC across folds when using 5-fold cross-validation on the dataset. <h3>Conclusion</h3> This work suggests that valuable radiomic features exist within 3D CT imagery that allows for the determination of post-SBRT RILI, and this can be learned automatically using modern artificial intelligence techniques. Furthermore, training a deep neural network for RILI classification produces noticeably high classification performance which may prove useful in the clinical setting to automatically flag scans with high probabilities of RILI for closer analysis by clinical experts.

Hybrid artificial intelligence paradigms for modeling of water-gas (pure/mixture) interfacial tension

There are many applications with the two-phase flow of gas (hydrocarbon, non-hydrocarbon, and their mixture) and water in different courses of gas recovery from natural gas resources and gas storage/sequestration programs. As the interface of gas-water is crucial in such systems, precise prediction of gas-water interfacial tension (IFT) can aid in the simulation and development of such processes. Artificial intelligence techniques (AIT) are being used to estimate IFT. In this paper, the IFT of the gas and water system was estimated based on models built using a comprehensive data set comprised of 2658 experimental data points. These cover a wide range of input parameters, i.e., specific gravity (0.5539–1.5225), temperature (278.1–477.5944 K), pressure (0.01–280 MPa), and water salinity (0–200,000 ppm). The intelligent models include Least-Squares Boosting (LS-Boost), Multilayer perceptron (MLP), Least Square Support Vector Machine (LSSVM), and Committee machine intelligent system (CMIS). The models reproduce the IFT data in 7.4–81.69 mN/m. The modeling approaches contain new hybrid forms in which Imperialist Competitive Algorithm (ICA), Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), Levenberg-Marquardt algorithm (LM), Bayesian regularization algorithm (BR), Scaled conjugate gradient algorithm (SCG), and Coupled Simulated Annealing (CSA) were used for optimization and learning purposes. Statistical and graphical analyses were implemented to check the agreement between the prediction and evaluation data. The results show a reasonable coherence for most models, among which the CMIS approach exhibited a promising performance. CMIS was accurate even in conditions of varying specific gravity, pressure, temperature, and salinity. The findings were also compared with available models in the literature and demonstrated superior predictions of the CMIS model. Also, outlier detection by the Leverage approach demonstrates the validity of the gathered dataset and, subsequently, the CMIS model.

Adopting Physical Artificial Intelligence in Soft Aerial Robots

In recent years, there has been a lot more attention towards the utilization of physically intelligent features in robotics. In this work, we provide a perspective on the physical artificial intelligence (PAI) paradigm and its impact on the conceptualization, design, and manufacturing of current and future aerial robots and infrastructure. We highlight the theory, enabling technologies, system features, and the tasks that the PAI paradigm will improve beyond the current approaches with conventional rigid aerial robots. We also discuss the multi-disciplinary effort required to collaborate with and educate researchers in the development of physically intelligent robots. PAI promises to lead the development of a new era of robust flying robotic organisms that are capable of adapting to and performing multi-functional tasks autonomously in a complex and unstructured environment. Aerial robotics is a great field of study to validate PAI as a development methodology.

Attention-Based UNet Deep Learning Model for Plaque Segmentation in Carotid Ultrasound for Stroke Risk Stratification: An Artificial Intelligence Paradigm.

Stroke and cardiovascular diseases (CVD) significantly affect the world population. The early detection of such events may prevent the burden of death and costly surgery. Conventional methods are neither automated nor clinically accurate. Artificial Intelligence-based methods of automatically detecting and predicting the severity of CVD and stroke in their early stages are of prime importance. This study proposes an attention-channel-based UNet deep learning (DL) model that identifies the carotid plaques in the internal carotid artery (ICA) and common carotid artery (CCA) images. Our experiments consist of 970 ICA images from the UK, 379 CCA images from diabetic Japanese patients, and 300 CCA images from post-menopausal women from Hong Kong. We combined both CCA images to form an integrated database of 679 images. A rotation transformation technique was applied to 679 CCA images, doubling the database for the experiments. The cross-validation K5 (80% training: 20% testing) protocol was applied for accuracy determination. The results of the Attention-UNet model are benchmarked against UNet, UNet++, and UNet3P models. Visual plaque segmentation showed improvement in the Attention-UNet results compared to the other three models. The correlation coefficient (CC) value for Attention-UNet is 0.96, compared to 0.93, 0.96, and 0.92 for UNet, UNet++, and UNet3P models. Similarly, the AUC value for Attention-UNet is 0.97, compared to 0.964, 0.966, and 0.965 for other models. Conclusively, the Attention-UNet model is beneficial in segmenting very bright and fuzzy plaque images that are hard to diagnose using other methods. Further, we present a multi-ethnic, multi-center, racial bias-free study of stroke risk assessment.