Standard Classification Models Research Articles

High frequency oscillations may improve somatosensory evoked potential detection of good outcomes in disorders of consciousness secondary to acute neurologic injury

BackgroundSomatosensory evoked potentials (SEPs) are highly specific predictors of poor prognosis in hypoxic-ischemic coma when cortical responses (N20s) are absent. However, bilateral N20 presence is nonspecific for good outcomes. High-frequency oscillations (HFOs) in the SEP waveform predict neurologic recovery in animals, but clinical applications are poorly understood. We sought to develop a clinical measure of HFOs to potentially improve detection of good outcomes in coma. Materials and MethodsWe collected SEP waveform data from all comatose inpatients (GCS<=8) who underwent neurologic prognostication from 2020 to 2022 at Johns Hopkins Hospital. We developed a novel measure − HFO evoked to spontaneous ratios (HFO-ESRs) – and applied this to those patients with bilaterally present N20s using both standard univariate classification and cubic kernal vector machine (SVM) models to predict the last documented in-hospital Glasgow Coma Scale (GCS) prior to discharge or death. ResultsOf 58 total patients, 34 (58.6%) had bilaterally present N20s. Of these, 14 had final GCS>=9, and 20 had final GCS<=8. Mean age was 52 (+/− 17) years, 20.1% female. Etiologies of coma were primarily global hypoxic-ischemic brain injury (79.4%), intracranial hemorrhage (11.8%), and traumatic brain injury (2.9%). In univariate classification, the addition of averaged HFO-ESRs to bilaterally present N20s predicted final GCS>=9 with 68% specificity. The SVM model further improved specificity to 85%. ConclusionsIn this pilot investigation, we developed a novel clinical measure of SEP HFOs. Incorporation of this measure may improve the specificity of the SEP to predict in-hospital GCS outcomes in coma, but requires further validation in specific neurologic injuries and with longitudinal outcomes.

Predicting progression from subjective cognitive decline to mild cognitive impairment or dementia based on brain atrophy patterns

BackgroundAlzheimer’s disease (AD) is a progressive neurodegenerative disorder where pathophysiological changes begin decades before the onset of clinical symptoms. Analysis of brain atrophy patterns using structural MRI and multivariate data analysis are an effective tool in identifying patients with subjective cognitive decline (SCD) at higher risk of progression to AD dementia. Atrophy patterns obtained from models trained to classify advanced AD versus normal subjects, may not be optimal for subjects at an early stage, like SCD. In this study, we compared the accuracy of the SCD progression prediction using the ‘severity index’ generated using a standard classification model trained on patients with AD dementia versus a new model trained on β-amyloid (Aβ) positive patients with amnestic mild cognitive impairment (aMCI).MethodsWe used structural MRI data of 504 patients from the Swedish BioFINDER-1 study cohort (cognitively normal (CN), Aβ-negative = 220; SCD, Aβ positive and negative = 139; aMCI, Aβ-positive = 106; AD dementia = 39). We applied multivariate data analysis to create two predictive models trained to discriminate CN individuals from either individuals with Aβ positive aMCI or AD dementia. Models were applied to individuals with SCD to classify their atrophy patterns as either high-risk “disease-like” or low-risk “CN-like”. Clinical trajectory and model accuracy were evaluated using 8 years of longitudinal data.ResultsIn predicting progression from SCD to MCI or dementia, the standard, dementia-based model, reached 100% specificity but only 10.6% sensitivity, while the new, aMCI-based model, reached 72.3% sensitivity and 60.9% specificity. The aMCI-based model was superior in predicting progression from SCD to MCI or dementia, reaching a higher receiver operating characteristic area under curve (AUC = 0.72; P = 0.037) in comparison with the dementia-based model (AUC = 0.57).ConclusionWhen predicting conversion from SCD to MCI or dementia using structural MRI data, prediction models based on individuals with milder levels of atrophy (i.e. aMCI) may offer superior clinical value compared to standard dementia-based models.

Invoice level forecasting with discrete survival methods for effective forecasting of account receivables in supply chain

The objective of this study is to obtain accurate and timely Account Receivables forecasts that can feed into the Cash Flow Forecasting (CFF) for a generic wholesales scenario in a supply chain company. The main components of the CFF are the Account Receivables (AR), Account Payables (AP) and Working Capital (WC). The focus of this work is on AR, in particular on how they contribute to the overall cash flow forecasting. The prediction of AR is based on the predicted payment date of each invoice, which can be obtained from the time to payment given the issue date. In this context, we propose the use of discrete survival methods for predicting the time to payment of issued invoices. To obtain more accurate predictions for time to payment, we fit a generalized linear mixed effects model with a time-varying baseline hazard function. This approach provides greater precision for subsequent account receivables predictions compared to the prediction based on simply the payment terms of each invoice, standard time series and classification models, and continuous-time survival methods. We demonstrate the approach on two financial data sets.

Efficient plant leaf detection through machine learning approach based on corn leaf image classification

Since maize is a staple diet for people, especially vegetarians and vegans, maize leaf disease has a significant influence here on the food industry including maize crop productivity. Therefore, it should be understood that maize quality must be optimal; yet, to do so, maize must be safeguarded from several illnesses. As a result, there is a great demand for such an automated system that can identify the condition early on and take the appropriate action. Early disease identification is crucial, but it also poses a major obstacle. As a result, in this research project, we adopt the fundamental k-nearest neighbor (KNN) model and concentrate on building and developing the improved k-nearest neighbor (EKNN) model. EKNN aids in identifying several classes of disease. To gather discriminative, boundary, pattern, and structurally linked information, additional high-quality fine and coarse features are generated. This information is then used in the classification process. The classification algorithm offers high-quality gradient-based features. Additionally, the proposed model is assessed using the Plant-Village dataset, and a comparison with many standard classification models using various metrics is also done.

Three-way imbalanced learning based on fuzzy twin SVM

Three-way decision (3WD) is a powerful tool for granular computing to deal with uncertain data, commonly used in information systems, decision-making, and medical care. Three-way decision gets much research in traditional rough set models. However, three-way decision is rarely combined with the currently popular field of machine learning to expand its research. In this paper, three-way decision is connected with SVM, a standard binary classification model in machine learning, for solving imbalanced classification problems that SVM needs to improve. A new three-way fuzzy membership function and a new fuzzy twin support vector machine with three-way membership (TWFTSVM) are proposed. The new three-way fuzzy membership function is defined to increase the certainty of uncertain data in both input space and feature space, which assigns higher fuzzy membership to minority samples compared with majority samples. To evaluate the effectiveness of the proposed model, comparative experiments are designed for forty-seven different datasets with varying imbalance ratios. In addition, datasets with different imbalance ratios are derived from the same dataset to further assess the proposed model’s performance. The results show that the proposed model significantly outperforms other traditional SVM-based methods.

A Weakly Supervised Gradient Attribution Constraint for Interpretable Classification and Anomaly Detection.

The lack of interpretability of deep learning reduces understanding of what happens when a network does not work as expected and hinders its use in critical fields like medicine, which require transparency of decisions. For example, a healthy vs pathological classification model should rely on radiological signs and not on some training dataset biases. Several post-hoc models have been proposed to explain the decision of a trained network. However, they are very seldom used to enforce interpretability during training and none in accordance with the classification. In this paper, we propose a new weakly supervised method for both interpretable healthy vs pathological classification and anomaly detection. A new loss function is added to a standard classification model to constrain each voxel of healthy images to drive the network decision towards the healthy class according to gradient-based attributions. This constraint reveals pathological structures for patient images, allowing their unsupervised segmentation. Moreover, we advocate both theoretically and experimentally, that constrained training with the simple Gradient attribution is similar to constraints with the heavier Expected Gradient, consequently reducing the computational cost. We also propose a combination of attributions during the constrained training making the model robust to the attribution choice at inference. Our proposition was evaluated on two brain pathologies: tumors and multiple sclerosis. This new constraint provides a more relevant classification, with a more pathology-driven decision. For anomaly detection, the proposed method outperforms state-of-the-art especially on difficult multiple sclerosis lesions segmentation task with a 15 points Dice improvement.

NORHA: A NORmal Hippocampal Asymmetry Deviation Index Based on One-Class Novelty Detection and 3D Shape Features.

Radiologists routinely analyze hippocampal asymmetries in magnetic resonance (MR) images as a biomarker for neurodegenerative conditions like epilepsy and Alzheimer's Disease. However, current clinical tools rely on either subjective evaluations, basic volume measurements, or disease-specific models that fail to capture more complex differences in normal shape. In this paper, we overcome these limitations by introducing NORHA, a novel NORmal Hippocampal Asymmetry deviation index that uses machine learning novelty detection to objectively quantify it from MR scans. NORHA is based on a One-Class Support Vector Machine model learned from a set of morphological features extracted from automatically segmented hippocampi of healthy subjects. Hence, in test time, the model automatically measures how far a new unseen sample falls with respect to the feature space of normal individuals. This avoids biases produced by standard classification models, which require being trained using diseased cases and therefore learning to characterize changes produced only by the ones. We evaluated our new index in multiple clinical use cases using public and private MRI datasets comprising control individuals and subjects with different levels of dementia or epilepsy. The index reported high values for subjects with unilateral atrophies and remained low for controls or individuals with mild or severe symmetric bilateral changes. It also showed high AUC values for discriminating individuals with hippocampal sclerosis, further emphasizing its ability to characterize unilateral abnormalities. Finally, a positive correlation between NORHA and the functional cognitive test CDR-SB was observed, highlighting its promising application as a biomarker for dementia.

An Intelligent Approach Based on Cleaning up of Inutile Contents for Extremism Detection and Classification in Social Networks

Extremism is a growing threat worldwide that presents a significant danger to public safety and national security. Social networks provide extremists with spaces to spread their ideas through commentaries or tweets, often in Asian English. In this paper, we propose an intelligent approach that cleans the text’s content, analyzes its sentiment, and extracts its features after converting it to digital data for machine learning treatments. We apply 16 intelligent machine learning classifiers for extremism detection and classification. The proposed artificial intelligence methods for Asian English language data are used to extract the essential features from the text. Our evaluation of the proposed model with an extremism dataset proves its effectiveness compared to the standard classification models based on various performance metrics. The proposed model achieves 93,6% accuracy for extremism detection and 97,0% for extremism classification.

Towards the Monitoring of Violent Events in Social Media through Visual Information

Violence is a latent threat for individuals, this is an even more concerning issue in Latin American cities. The detection and monitoring of events reported in social media could help to build maps of zones with incident of violent events. This could eventually lead to the automatic generation of risk maps which could be of great help to users and even authorities. This paper aims to detect violent incidents reported in social media using visual information only. While most of the related work focuses on the text modality, the goal of this paper is to assess the feasibility of detection when only visual information is available. CNN based feature extraction and standard classification models are implemented and evaluated in a recently released corpus. Experimental results show that distinguishing images depicting violent events is feasible, but the fine grain recognition of categories is still an open problem.

Data Analysis of Impaired Renal and Cardiac Function Using a Combination of Standard Classifiers

We examine the significance of the predictive potential of EPI cystatin C (EPI CysC) in combination with NTproBNP, sodium, and potassium in the evaluation of renal function in patients with cardiorenal syndrome using standard mathematical classification models from the domain of artificial intelligence. The criterion for the inclusion of subjects with combined impairment of heart and kidney function in the study was the presence of newly discovered or previously diagnosed clinically manifest cardiovascular disease and acute or chronic kidney disease in different stages of evolution. In this paper, five standard classifiers from the field of machine learning were used for the analysis of the obtained data: ensemble of neural networks (MLP), ensemble of k-nearest neighbors (k-NN) and naive Bayes classifier, decision tree, and a classifier based on logistic regression. The results showed that in MLP, k-NN, and naive Bayes, EPI CysC had the highest predictive potential. Thus, our approach with utility classifiers recognizes the essence of the disorder in patients with cardiorenal syndrome and facilitates the planning of further treatment.

Generalizable electroencephalographic classification of Parkinson's disease using deep learning

Growing interest surrounds the use of electroencephalography (EEG) and deep learning for diagnosing neurological conditions like Parkinson's Disease (PD). Despite the existing proof-of-concept literature demonstrating the potential of deep learning in classifying PD from EEG data, neurologists have been slow to adopt these tools due to insufficient evidence of their real-world diagnostic generalizable performance. Our study aimed to evaluate the potential of deep learning for inter-subject PD classification using a conservative training approach and testing on an external independent dataset. Specifically, we utilized publicly available resting-state EEG data from PD patients at two separate centers, the University of New Mexico (n = 54) and the University of Iowa (n = 28), as our training and testing sets, respectively. Each of these recordings had a minimum of 2 minutes of data. We implemented a channel-wise convolutional neural network, tuning it with a leave-one-subject-out cross-validation approach. Our approach achieved a patient-level accuracy of 80.4% (epoch-level accuracy = 72.7%), which remained consistent when tested on the external dataset (patient-level accuracy = 82.8%, epoch-level accuracy = 75.7%). Our model performs equal-or-better than other standard classification models and our approach compares favourably to similar works. Our publicly available code serves as a foundation for future research exploring different deep learning architectures, investigating other pathologies, and involving larger datasets with the hope of accelerating the adoption of objective computational approaches for the diagnosis and monitoring of neurological disorders.

Noise Invariant Convolution Neural Network for Segmentation of Multiple Sclerosis Lesions from Brain Magnetic Resonance Imaging

The objective of the research work is to accurately segment multiple sclerosis (MS) lesions in brain Magnetic Resonance Imaging (MRI) of varying sizes and also to classify its types. Designing effective automatic segmentation and classification tool aid the doctors in better understanding MS lesion progressions. In meeting research challenges, this paper presents Noise Invariant Convolution Neural Network (NICNN). The NICNN model is efficient in the detection and segmentation of MS lesions of varying sizes in comparison with standard CNN-based segmentation methods. Further, this paper introduced a new cross-validation scheme to address the class imbalance issue by selecting effective features for classifying the type of MS lesion. The experiment outcome shows the proposed method provides improved Dice Similarity Coefficient (DSC), Positive Predicted Value (PPV), and True Positive Rate (TPR) value compared to the state-of-art CNN-based MS lesion segmentation method. Further, achieves better accuracy in classifying MS lesion types compared to standard MS lesion type classification models.

Classification model with weighted regularization to improve the reproducibility of neuroimaging signature selection.

Machine learning (ML) has been extensively applied in brain imaging studies to aid the diagnosis of psychiatric disorders and the selection of potential biomarkers. Due to the high dimensionality of imaging data and heterogeneous subtypes of psychiatric disorders, the reproducibility of ML results in brain imaging studies has drawn increasing attention. The reproducibility in brain imaging has been primarily examined in terms of prediction accuracy. However, achieving high prediction accuracy and discovering relevant features are two separate but related goals. An important yet under-investigated problem is the reproducibility of feature selection in brain imaging studies. We propose a new metric to quantify the reproducibility of neuroimaging feature selection via bootstrapping. We estimate the reproducibility index (R-index) for each feature as the reciprocal coefficient of variation of absolute mean difference across a larger number of bootstrap samples. We then integrate the R-index in regularized classification models as penalty weight. Reproducible features with a larger R-index are assigned smaller penalty weights and thus are more likely to be selected by our proposed models. Both simulated and multimodal neuroimaging data are used to examine the performance of our proposed models. Results show that our proposed R-index models are effective in separating informative features from noise features. Additionally, the proposed models yield similar or higher prediction accuracy than the standard regularized classification models while further reducing coefficient estimation error. Improvements achieved by the proposed models are essential to advance our understanding of the selected brain imaging features as well as their associations with psychiatric disorders.

Toward a Comprehensive Domestic Dirt Dataset Curation for Cleaning Auditing Applications.

Cleaning is an important task that is practiced in every domain and has prime importance. The significance of cleaning has led to several newfangled technologies in the domestic and professional cleaning domain. However, strategies for auditing the cleanliness delivered by the various cleaning methods remain manual and often ignored. This work presents a novel domestic dirt image dataset for cleaning auditing application including AI-based dirt analysis and robot-assisted cleaning inspection. One of the significant challenges in an AI-based robot-aided cleaning auditing is the absence of a comprehensive dataset for dirt analysis. We bridge this gap by identifying nine classes of commonly occurring domestic dirt and a labeled dataset consisting of 3000 microscope dirt images curated from a semi-indoor environment. The dirt dataset gathered using the adhesive dirt lifting method can enhance the current dirt sensing and dirt composition estimation for cleaning auditing. The dataset’s quality is analyzed by AI-based dirt analysis and a robot-aided cleaning auditing task using six standard classification models. The models trained with the dirt dataset were capable of yielding a classification accuracy above 90% in the offline dirt analysis experiment and 82% in real-time test results.

Enhanced convolutional neural network based android software maintainability prediction

Advancement in mobile technology, necessitates quality in development of android applications. To improve the quality of android application, evaluation of its maintenance has to be started from the initial stage of development. There are several existing models involved in android application maintenance evaluation, but most of them suffers from the class imbalance problem and its performance depends on the size of the dataset. This paper focuses on overcoming the class imbalance while using convolutional neural network for prediction the android application maintainability. In this work both android metrics and object-oriented metrics are used for evaluating the android application maintainability. The standard convolutional neural network is enhanced by pretraining it using evolutionary algorithm known as moth flame optimization (MFO). The MFO involves in optimizing the learning parameters of convolutional layer and after the training process is completed, with the acquired knowledge it handles the small size dataset very effectively during the testing phase. Thus, the simulation results of maintainability prediction in android application development using proposed enhanced CNN achieves better result while comparing with other standard classification models.

DRIIS: Research on Image Classification of Art Education System Based on Deep Learning

With the fast advancement of computer technology, using it as an auxiliary teaching technique in art classes may not only improve the content of art classes but also give students with advice. This research develops an art education picture categorization technique and a system based on an upgraded deep learning model. To categorize artistic photos, this technique suggests using the dual-kernel squeeze-and-excitation (DKSE) module with deep separable convolutional operations to form a convolutional neural network. The suggested model has a classification accuracy of 86.49%, which is 26.29% better than the standard classification models. The classification accuracy of the DKSE module branch is 87.54% when the convolutional kernels are [Formula: see text] and [Formula: see text]. The suggested DKSE network model extracts both the overall and partial information of artistic images efficiently.

An Intrusion Detection Mechanism for Secured IoMT Framework Based on Swarm-Neural Network.

The seamless integration of medical sensors and the Internet of Things (IoT) in smart healthcare has leveraged an intelligent Internet of Medical Things (IoMT) framework to detect the criticality of the patients. However, due to the limited storage capacity and computation power of the local IoT devices, patient's health data needs to transfer to remote computing devices for analysis, which can easily result in privacy leakage due to lack of control over the patient's health data and the vulnerability of the network for various types of attacks. Motivated by this, in this paper, an Empirical Intelligent Agent (EIA) based on a unique Swarm-Neural Network (Swarm-NN) method is proposed to identify attackers in the edge-centric IoMT framework. The major outcome of the proposed strategy is to identify the attacks during data transmission through a network and analyze the health data efficiently at the edge of the network with higher accuracy. The proposed Swarm-NN strategy is evaluated with a real-time secured dataset, namely the ToN-IoT dataset that collected Telemetry, Operating systems, and Network data for IoT applications and compares the performance over the standard classification models using various performance metrics. The test results demonstrate that the proposed Swarm-NN strategy achieves 99.5% accuracy over the ToN-IoT dataset.

Hyperspherical class prototypes for adversarial robustness

This work addresses the problem of adversarial robustness in deep neural network classification from an optimal class boundary estimation perspective. It is argued that increased model robustness to adversarial attacks can be achieved when the feature learning process is monitored by geometrically-inspired optimization criteria. To this end, we propose to learn hyperspherical class prototypes in the neural feature embedding space, along with training the network parameters. Three concurrent optimization functions for the intermediate hidden layer training data activations are devised, requiring items of the same class to be enclosed by the corresponding class prototype boundaries, to have minimum distance from their class prototype vector (i.e., hypersphere center) and to have maximum distance from the remainder hypersphere centers. Our experiments show that training standard classification model architectures with the proposed objectives, significantly increases their robustness to white-box adversarial attacks, without adverse (if not beneficial) effects to their classification accuracy.

An Evolutionary-Based Sentiment Analysis Approach for Enhancing Government Decisions during COVID-19 Pandemic: The Case of Jordan

The world has witnessed recently a global outbreak of coronavirus disease (COVID-19). This pandemic has affected many countries and has resulted in worldwide health concerns, thus governments are attempting to reduce its spread and impact on different aspects of life such as health, economics, education, and politics by making emergent decisions and policies (e.g., lockdown and social distancing). These new regulations influenced people’s daily life and cast significant burdens, concerns, and disparities on various population groups. Taking the wrong actions and enforcing bad decisions by some countries result in increasing the contagion rate and more catastrophic results. People start to post their opinions and feelings about their government’s decisions on different social media networks, and the data received through these platforms present a very useful source of information that affects how governments perceive and cope with the current the pandemic. Jordan was one of the top affected countries. In this paper, we proposed a decision support system based on the sentiment analysis mechanism by combining support vector machines with a whale optimization algorithm for automatically tuning the hyperparameters and performing feature weighting. The work is based on a hybrid evolutionary approach that aims to perform sentiment analysis combined with a decision support system to study people’s posts on Facebook to investigate their attitudes and feelings toward the government’s decisions during the pandemic. The government regulations were divided into two periods: the first and latter regulations. Studying public sentiments during these periods allows decision-makers in the government to sense people’s feelings, alert them in case of possible threats, and help in making proactive actions if needed to better handle the current pandemic situation. Five different versions were generated for each of the two collected datasets. The results demonstrate the superiority of the proposed Whale Optimization Algorithm &amp; Support Vector Machines (WOA-SVM) against other metaheuristic algorithms and standard classification models as WOA-SVM has achieved 78.78% in terms of accuracy and 84.64% in term of f-measure, while other standard classification models such as NB, k-NN, J84, and SVM achieved an accuracy of 69.25%, 69.78%, 70.17%, and 69.29%, respectively, with 64.15%, 62.90%, 60.51%, and 59.09% F-measure. Moreover, when comparing our proposed WOA-SVM approach with other metaheuristic algorithms, which are GA-SVM, PSO-SVM, and MVO-SVM, WOA-SVM proved to outperform the other approaches with results of 78.78% in terms of accuracy and 84.64% in terms of F-measure. Further, we investigate and analyze the most relevant features and their effect to improve the decision support system of government decisions.

EPX: An R package for the ensemble of subsets of variables for highly unbalanced binary classification.

In binary classification problems with a rare class of interest, there is relatively little information available for the rare class to build a model. On the other hand, the number of useful variables to develop a model for classification can be high-dimensional. For example, in drug discovery, there are usually a very few bioactive compounds in a large chemical library, whereas thousands of potentially useful explanatory variables characterize a compound's chemical structure. The sparsity of information for the rare class of interest makes it difficult for the standard classification models to exploit the richness of the useful feature variables. Thus, the objective of this paper is to develop an R package which clusters the feature variables into diverse subsets to be aggregated into a powerful ensemble for the detection of a rare class object. The ensemble of phalanxes (EPX) builds a classifier by exploiting the richness of feature variables using several diverse subsets of variables, called phalanxes, and outperforms many competitive state-of-the-art classification methods in terms of predictive ranking of the rare class of interest. We present an R package EPX which implements the algorithm to form the ensemble of phalanxes as well as its associated functions. We further show how the ensemble of phalanxes can be constructed using parallel computing to lower the computational burden given high-dimensional data. The R package EPX shows a flexible way of clustering feature variable space into smaller and diverse subsets of variables to develop an ensemble of phalanxes which better ranks a rare class object in a highly unbalanced two class classification problem. The ensemble EPX will be useful to detect the rare drug-like active biomolecules for development in drug discovery (Tomal et al., Mar. 2016) [1] and homologous proteins using similarity scores of amino acid sequences in protein homology (Tomal et al., 2019) [2]. The package EPX is freely available to download from CRAN (https://CRAN.R-project.org/package=EPX).