Machine Learning (ML) plays a crucial role in Industry 4.0, enabling predictive fault detection (FD) by analyzing vast amounts of log data. However, current ML approaches often rely on single-task learning, neglecting the diverse nature of log data and the prediction of interrelated faults. Moreover, multi-view learning (MVL) and multi-task learning (MTL) are usually applied disjointly without applying joint learning across tasks and views. To address these gaps, we a novel approach which leverages multi-task and multi-view learning frameworks, augmented by a multi-label Cross Entropy loss (MTMVL-CE). MTMVL-CE improves generalization performance between and within different fault types, enabling the classification of multiple faults in complex industrial machines. Indeed, MTMVL-CE optimizes classification performance by learning across numerous faults simultaneously, achieving an accurate representation of heterogeneous log data, robust fault classification, and feasible generalization over time. We tested our approach through extensive experiments on an real use case involving FD in a complex banknote recirculator device inside Automated Teller Machines. Our results demonstrate MTMVL-CE’s superior performance compared to MTL and MVL competitors in capturing fault interdependencies and providing accurate, reliable predictions. • MTMVL-CE: unified framework for multi-source industrial fault detection. • Jointly optimizes within-task and between-task fault dependencies. • Correlation-based regularizer models task relationships. • Validated on 7400 heterogeneous real-world ATM log records. • Superior temporal generalization compared to state-of-the-art approaches.

Hepatic steatosis, or non-alcoholic fatty liver disease (NAFLD), affects a significant portion of the global population and can lead to more severe liver conditions, including hepatic fibrosis. Early and accurate risk prediction of fibrosis is crucial for timely intervention. Traditional diagnostic methods are invasive and carry risks, while imaging techniques and blood-based biomarkers have limitations in routine general practice. This study presents a machine learning-based clinical decision support system designed to assess the risk of hepatic fibrosis in patients with NAFLD using routine laboratory tests. The framework is developed using electronic health record data collected over 15 years, initially encompassing 1,272,572 patients from general practice. After applying clinical selection criteria, two cohorts of 12,960 and 25,478 patients were used for model development and evaluation. The proposed approach provides a robust foundation for monitoring fibrosis risk by implementing a novel screening method, which preprocesses predictors by leveraging well-established clinical indicators (e.g., hepatic steatosis index, fibrosis-4 index), alongside a selected minimal number of predictors, making it practical and cost-effective for widespread clinical use. The study’s findings indicate promising results for screening and monitoring fibrosis risk in NAFLD patients, achieving the best AUC of 92.97%, PRAUC of 75.44%, and Sensitivity of 79.63%.

Real-world classification problems may disclose different hierarchical levels where the categories are displayed in an ordinal structure. However, no specific deep learning (DL) models simultaneously learn hierarchical and ordinal constraints while improving generalization performance. To fill this gap, we propose the introduction of two novel ordinal-hierarchical DL methodologies, namely, the hierarchical cumulative link model (HCLM) and hierarchical-ordinal binary decomposition (HOBD), which are able to model the ordinal structure within different hierarchical levels of the labels. In particular, we decompose the hierarchical-ordinal problem into local and global graph paths that may encode an ordinal constraint for each hierarchical level. Thus, we frame this problem as simultaneously minimizing global and local losses. Furthermore, the ordinal constraints are set by two approaches [ordinal binary decomposition (OBD) and cumulative link model (CLM)] within each global and local function. The effectiveness of the proposed approach is measured on four real-use case datasets concerning industrial, biomedical, computer vision, and financial domains. The extracted results demonstrate a statistically significant improvement to state-of-the-art nominal, ordinal, and hierarchical approaches.

The use of multiple raters to label datasets is an established practice in affective computing. The principal goal is to reduce unwanted subjective bias in the labelling process. Unfortunately, this leads to the key problem of identifying a ground truth for training the affect recognition system. This problem becomes more relevant in a sparsely-crossed annotation where each rater only labels a portion of the full dataset to ensure a manageable workload per rater. In this paper, we introduce a Multi-Rater Consensus Learning (MRCL) method which learns a representative affect recognition model that accounts for each rater's agreement with the other raters. MRCL combines a multitask learning (MTL) regularizer and a consensus loss. Unlike standard MTL, this approach allows the model to learn to predict each rater's label while explicitly accounting for the consensus among raters. We evaluated our approach on two different datasets based on spontaneous affective body movement expressions for pain behaviour detection and laughter type recognition respectively. The two naturalistic datasets were chosen for the different forms of labelling (different in affect, observation stimuli, and raters) that they together offer for evaluating our approach. Empirical results demonstrate that MRCL is effective for modelling affect from datasets with sparsely-crossed multi-rater annotation.

Predictive Maintenance (PdM) methods aim to facilitate the scheduling of maintenance work before equipment failure. In this context, detecting early faults in automated teller machines (ATMs) has become increasingly important since these machines are susceptible to various types of unpredictable failures. ATMs track execution status by generating massive event-log data that collect system messages unrelated to the failure event. Predicting machine failure based on event logs poses additional challenges, mainly in extracting features that might represent sequences of events indicating impending failures. Accordingly, feature learning approaches are currently being used in PdM, where informative features are learned automatically from minimally processed sensor data. However, a gap remains to be seen on how these approaches can be exploited for deriving relevant features from event-log-based data. To fill this gap, we present a predictive model based on a convolutional kernel (MiniROCKET and HYDRA) to extract features from the original event-log data and a linear classifier to classify the sample based on the learned features. The proposed methodology is applied to a significant real-world collected dataset. Experimental results demonstrated how one of the proposed convolutional kernels (i.e. HYDRA) exhibited the best classification performance (accuracy of 0.759 and AUC of 0.693). In addition, statistical analysis revealed that the HYDRA and MiniROCKET models significantly overcome one of the established state-of-the-art approaches in time series classification (InceptionTime), and three non-temporal ML methods from the literature. The predictive model was integrated into a container-based decision support system to support operators in the timely maintenance of ATMs.

Abstract Nowadays, decision support systems (DSSs) are widely used in several application domains, from industrial to healthcare and medicine fields. Concerning the industrial scenario, we propose a DSS oriented to the aesthetic quality control (AQC) task, which has quickly established itself as one of the most crucial challenges of Industry 4.0. Taking into account the increasing amount of data in this domain, the application of machine learning (ML) and deep learning (DL) techniques offers great opportunities to automatize the overall AQC process. State-of-the-art is mainly oriented to approach this problem with a nominal DL classification method which does not exploit the ordinal structure of the AQC task, thus not penalizing the error among distant AQC classes (which is a relevant aspect for the real use case). The paper introduces a DL ordinal methodology for the AQC classification. Differently from other deep ordinal methods, we combined the standard categorical cross-entropy with the cumulative link model and we imposed the ordinal constraint via the thresholds and slope parameters. Experimental results were performed for solving an AQC task on a novel image dataset originated from a specific company’s demand (i.e., aesthetic assessment of wooden stocks). We demonstrated how the proposed methodology is able to reduce misclassification errors (up to 0.937 quadratic weight kappa loss) among distant classes while overcoming other state-of-the-art deep ordinal models and reducing the bias factor related to the item geometry. The proposed DL approach was integrated as the main core of a DSS supported by Internet of Things (IoT) architecture that can support the human operator by reducing up to 90% the time needed for the qualitative analysis carried out manually in this specific domain.

The problem of continuous emotion recognition has been the subject of several studies. The proposed affective computing approaches employ sequential machine learning algorithms for improving the classification stage, accounting for the time ambiguity of emotional responses. Modeling and predicting the affective state over time is not a trivial problem because continuous data labeling is costly and not always feasible. This is a crucial issue in real-life applications, where data labeling is sparse and possibly captures only the most important events rather than the typical continuous subtle affective changes that occur. In this work, we introduce a framework from the machine learning literature called Multiple Instance Learning, which is able to model time intervals by capturing the presence or absence of relevant states, without the need to label the affective responses continuously (as required by standard sequential learning approaches). This choice offers a viable and natural solution for learning in a weakly supervised setting, taking into account the ambiguity of affective responses. We demonstrate the reliability of the proposed approach in a gold-standard scenario and towards real-world usage by employing an existing dataset (DEAP) and a purposely built one (Consumer). We also outline the advantages of this method with respect to standard supervised machine learning algorithms.

This article presents a fault diagnosis algorithm for rotating machinery based on the Wasserstein distance. Recently, the Wasserstein distance has been proposed as a new research direction to find better distribution mapping when compared with other popular statistical distances and divergences. In this work, first, frequency- and time-based features are extracted by vibration signals, and second, the Wasserstein distance is considered for the learning phase to discriminate the different machine operating conditions. Specifically, the 1-D Wasserstein distance is considered due to its low computational burden because it can be evaluated directly by the order statistics of the extracted features. Furthermore, a distance weighting stage based on neighborhood component features selection (NCFS) is exploited to achieve robust fault diagnosis at low signal-to-noise ratio (SNR) conditions and with high-dimensional features. In detail, the NCFS framework is here adapted to weight 1-D Wasserstein distances evaluated from time/frequency features. Experiments are conducted on two benchmark data sets to verify the effectiveness of the proposed fault diagnosis method at different SNR conditions. The comparison with state-of-the-art fault diagnosis algorithms shows promising results.Note to Practitioners—This article was motivated by the problem of fault diagnosis of rotating machinery under low SNR and different machine operating conditions. The algorithm employs a statistical distance-based fault diagnosis technique, which permits to obtain an estimation of the fault signature without the need for training a classifier. The algorithm is computationally efficient during the training and testing stages, and thus, it can be used in embedded hardware. Finally, the proposed methodology can be applied to other application domains such as system monitoring and prognostics, which can help to schedule the maintenance of rotating machinery.

As COVID-19 hounds the world, the common cause of finding a swift solution to manage the pandemic has brought together researchers, institutions, governments, and society at large. The Internet of Things (IoT), artificial intelligence (AI)-including machine learning (ML) and Big Data analytics-as well as Robotics and Blockchain, are the four decisive areas of technological innovation that have been ingenuity harnessed to fight this pandemic and future ones. While these highly interrelated smart and connected health technologies cannot resolve the pandemic overnight and may not be the only answer to the crisis, they can provide greater insight into the disease and support frontline efforts to prevent and control the pandemic. This article provides a blend of discussions on the contribution of these digital technologies, propose several complementary and multidisciplinary techniques to combat COVID-19, offer opportunities for more holistic studies, and accelerate knowledge acquisition and scientific discoveries in pandemic research. First, four areas, where IoT can contribute are discussed, namely: 1) tracking and tracing; 2) remote patient monitoring (RPM) by wearable IoT (WIoT); 3) personal digital twins (PDTs); and 4) real-life use case: ICT/IoT solution in South Korea. Second, the role and novel applications of AI are explained, namely: 1) diagnosis and prognosis; 2) risk prediction; 3) vaccine and drug development; 4) research data set; 5) early warnings and alerts; 6) social control and fake news detection; and 7) communication and chatbot. Third, the main uses of robotics and drone technology are analyzed, including: 1) crowd surveillance; 2) public announcements; 3) screening and diagnosis; and 4) essential supply delivery. Finally, we discuss how distributed ledger technologies (DLTs), of which blockchain is a common example, can be combined with other technologies for tackling COVID-19.

Kidney Disease (KD) may hide complex causes and is associated with a tremendous socio-economic impact. Timely identification and management from the first level of medical care represent the most effective strategy to address the growing global burden sustainably. Clinical practice guidelines suggest utilizing estimated Glomerular Filtration Rate (eGFR) for routine evaluation within a screening purpose. Accordingly, the analysis of Electronic Health Records (EHRs) using Machine Learning techniques offers great opportunities to monitor and predict the eGFR trend over time. This paper aims to propose a novel Semi-Supervised Multi-Task Learning (SS-MTL) approach for predicting short-term KD evolution on multiple General Practitioners' EHR data. We demonstrated that the SS-MTL approach can (i) capture the eGFR temporal evolution by imposing a temporal relatedness between consecutive time windows and (ii) exploit useful information from unlabeled patients when labeled patients are less numerous with a gain of up to 4.1% in terms of Recall. This situation reflects the real-case scenario, where available labeled samples are limited, but those unlabeled much more abundant. The SS-MTL approach, also given the high level of interpretability, might be the ideal candidate in general practice to get integrated within a decision support system for KD screening purposes.

The prediction of the risk profile related to the cardiopathy complication is a core research task that could support clinical decision making. However, the design and implementation of a clinical decision support system based on Electronic Health Record (EHR) temporal data comprise of several challenges. Several single task learning approaches consider the prediction of the risk profile related to a specific diabetes complication (i.e., cardiopathy) independent from other complications. Accordingly, the state-of-the-art multi-task learning (MTL) model encapsulates only the temporal relatedness among the EHR data. However, this assumption might be restricted in the clinical scenario where both spatio-temporal constraints should be taken into account. The aim of this study is the proposal of two different MTL procedures, called spatio-temporal lasso (STL-MTL) and spatio-temporal group lasso (STGL-MTL), which encode the spatio-temporal relatedness using a regularization term and a graph-based approach (i.e., encoding the task relatedness using the structure matrix). Experimental results on a real-world EHR dataset demonstrate the robust performance and the interpretability of the proposed approach.

The diagnosis of type 2 diabetes (T2D) at an early stage has a key role for an adequate T2D integrated management system and patient's follow-up. Recent years have witnessed an increasing amount of available electronic health record (EHR) data and machine learning (ML) techniques have been considerably evolving. However, managing and modeling this amount of information may lead to several challenges, such as overfitting, model interpretability, and computational cost. Starting from these motivations, we introduced an ML method called sparse balanced support vector machine (SB-SVM) for discovering T2D in a novel collected EHR dataset (named Federazione Italiana Medici di Medicina Generale dataset). In particular, among all the EHR features related to exemptions, examination, and drug prescriptions, we have selected only those collected before T2D diagnosis from an uniform age group of subjects. We demonstrated the reliability of the introduced approach with respect to other ML and deep learning approaches widely employed in the state-of-the-art for solving this task. Results evidence that the SB-SVM overcomes the other state-of-the-art competitors providing the best compromise between predictive performance and computation time. Additionally, the induced sparsity allows to increase the model interpretability, while implicitly managing high-dimensional data and the usual unbalanced class distribution.

Here, the authors present a low cost telerehabilitation system made up of a commercial red–green–blue depth (RGB‐D) camera and a web‐based platform. The authors goal is to monitor and assess subject movement providing acceptable and usable at‐home remote rehabilitation services without the presence of a clinician. Clinical goals, defined by physiotherapists, are firstly translated into motion analysis features. A Takagi Sugeno fuzzy inference system (FIS) is then proposed to evaluate and combine these features into scores. In this stage, the ‘collaborative design’ paradigm is used in depth and complete manner: the contribution of the clinician is not limited only to the rules definition but enters in the core of the evaluation algorithm through the definition of the fuzzy rules. A case study on low back pain rehabilitation involving 40 subjects, 5 exercises, and 4 physiotherapists is then presented to the effectiveness of the proposed system. Results of the validation of the system aimed at the assessment of the reliability of the proposed approach show high correlations between clinician evaluation and FIS scores. In this scenario, due to the high correlation, each FIS could represent a virtual alter‐ego of the physiotherapist which enable a real time and free second opinion.