Difficulties In Classification Research Articles

Skin lesions and the role of the internist: a series of clinical cases

The skin often acts as an indicator of underlying internal diseases and skin lesions that are difficult to diagnose are one of the reasons for hospitalization in internal medicine departments where patients frequently arrive after a long journey between different dermatology specialists and after ineffective or sometimes responsible for diagnostic delays. In this article, we will illustrate three clinical cases observed in our General Medicine Unit of the Ospedale del Mare in Naples in the last year that are characterized by diagnostic errors that misled our attention, the notable difficulty in the diagnostic classification, and the duration of hospitalization necessary to reach the correct diagnosis. The first case is that of a patient diagnosed with violin spider lesions who, after a 2-week hospitalization, was discharged with the diagnosis of non-Hodgkin’s lymphoma. In the second case, the patient was hospitalized for suspected lesions caused by Burger’s disease and was instead discharged with a lesion caused by mycosis fungoides. The third case is that of a patient hospitalized for suspected vasculitis and discharged with the diagnosis of drug-induced toxic epidermal necrolysis.

Advancing user classification models: A comparative analysis of machine learning approaches to enhance faculty password policies at the University of Buraimi

In this paper, we assess the results of experiment with different machine learning algorithms for the data classification on the basis of accuracy, precision, recall and F1-Score metrics. We collected metrics like Accuracy, F1-Score, Precision, and Recall: From the Neural Network model, it produced the highest Accuracy of 0.129526 also highest F1-Score of 0.118785, showing that it has the correct balance of precision and recall ratio that can pick up important patterns from the dataset. Random Forest was not much behind with an accuracy of 0.128119 and highest precision score of 0.118553 knit a great ability for handling relations in large dataset but with slightly lower recall in comparison with Neural Network. This ranked the Decision Tree model at number three with a 0.111792, Accuracy Score while its Recall score showed it can predict true positives better than Support Vector Machine (SVM), although it predicts more of the positives than it actually is a majority of the times. SVM ranked fourth, with accuracy of 0.095465 and F1-Score of 0.067861, the figure showing difficulty in classification of associated classes. Finally, the K-Neighbors model took the 6th place, with the predetermined accuracy of 0.065531 and the unsatisfactory results with the precision and recall indicating the problems of this algorithm in classification. We found out that Neural Networks and Random Forests are the best algorithms for this classification task, while K-Neighbors is far much inferior than the other classifiers.

AER-Net: Attention-Enhanced Residual Refinement Network for Nuclei Segmentation and Classification in Histology Images.

The acurate segmentation and classification of nuclei in histological images are crucial for the diagnosis and treatment of colorectal cancer. However, the aggregation of nuclei and intra-class variability in histology images present significant challenges for nuclei segmentation and classification. In addition, the imbalance of various nuclei classes exacerbates the difficulty of nuclei classification and segmentation using deep learning models. To address these challenges, we present a novel attention-enhanced residual refinement network (AER-Net), which consists of one encoder and three decoder branches that have same network structure. In addition to the nuclei instance segmentation branch and nuclei classification branch, one branch is used to predict the vertical and horizontal distance from each pixel to its nuclear center, which is combined with output by the segmentation branch to improve the final segmentation results. The AER-Net utilizes an attention-enhanced encoder module to focus on more valuable features. To further refine predictions and achieve more accurate results, an attention-enhancing residual refinement module is employed at the end of each encoder branch. Moreover, the coarse predictions and refined predictions are combined by using a loss function that employs cross-entropy loss and generalized dice loss to efficiently tackle the challenge of class imbalance among nuclei in histology images. Compared with other state-of-the-art methods on two colorectal cancer datasets and a pan-cancer dataset, AER-Net demonstrates outstanding performance, validating its effectiveness in nuclear segmentation and classification.

Differential diagnosis of dizziness: what's the contribution of Artificial Intelligence?

Dizziness is one of the most common reasons for medical consultations. The interdisciplinary range of differential diagnoses often leads to difficulties in proper classification. Artificial Intelligence and machine learning can assist through data-driven algorithms and facilitate the collection of important clinical signs as digital biomarkers. The article will present and critically discuss the current evidence on the topic.

A Risk Stratification Study of Ultrasound Images of Thyroid Nodules Based on Improved DETR

ABSTRACTThe Chinese Thyroid Imaging Reporting and Data System (C‐TIRADS) standard is based on the Chinese current medical context. However, at present, there is a lack of C‐TIRADS‐based automatic computer‐aided diagnosis system for thyroid nodule ultrasound images, and the existing algorithms for detecting and recognizing thyroid nodules are basically for the dichotomous classification of benign and malignant. We used the DETR (detection transformer) model as a baseline model and carried out model enhancements to address the shortcomings of unsatisfactory classification accuracy and difficulty in detecting small thyroid nodules in the DETR model. First, to investigate the method of acquiring multi‐scale features of thyroid nodule ultrasound images, we choose TResNet‐L as the feature extraction network and introduce multi‐scale feature information and group convolution, thereby enhancing the model's multi‐label classification accuracy. Second, a parallel decoder architecture for multi‐label thyroid nodule ultrasound image classification is designed to enhance the learning of correlation between pathological feature class labels, aiming to improve the multi‐label classification accuracy of the detection model. Third, the loss function of the detection model is improved. We propose a linear combination of Smooth L1‐Loss and CIoU Loss as the model's bounding box loss function and asymmetric loss as the model's multi‐label classification loss function, aiming to further improve the detection model's detection accuracy for small thyroid nodules. The experiment results show that the improved DETR model achieves an AP of 92.4% and 81.6% with IoU thresholds of 0.5 and 0.75, respectively.

Design of Time Series Fuzzy Segmentation Algorithm for Communication Data Classification

In order to solve the problem that the imbalance of communication data sets leads to a significant increase in classification difficulty, a classification algorithm for fuzzy segmentation of time series is proposed. The principal component analysis method is used to obtain the eigenvector with the largest eigenvalue. The time series of data is established according to the interval number theory. The segmentation target of the communication data time series is characterized by the Langmuir distance measurement function between data and categories. The classification result of fuzzy segmentation is obtained based on the judgment relationship between the difference of fuzzy classification matrix and the closing condition. The experimental results show that the number of data in the three classification situations of the algorithm in this paper is always at the corresponding ideal level, with high accuracy and low error and failure.

Enhancing MeshNet for 3D shape classification with focal and regularization losses

With the development of deep learning and computer vision, an increasing amount of research has focused on applying deep learning models to the recognition and classification of three-dimensional shapes. In classification tasks, differences in sample quantity, feature amount, model complexity, and other aspects among different categories of 3D model data cause significant variations in classification difficulty. However, simple cross-entropy loss is generally used as the loss function, but it is insufficient to address these differences. In this paper, we used MeshNet as the base model and introduced focal loss as a metric for the loss function. Additionally, to prevent deep learning models from developing a preference for specific categories, we incorporated regularization loss. The combined use of focal loss and regularization loss in optimizing the MeshNet model’s loss function resulted in a classification accuracy of up to 92.46%, representing a 0.20% improvement over the original model’s highest accuracy of 92.26%. Furthermore, the average accuracy over the final 50 epochs remained stable at a higher level of 92.01%, reflecting a 0.71% improvement compared to the original MeshNet model’s 91.30%. These results indicate that our method performs better in 3D shape classification task.

Age differences in BOLD modulation to task difficulty as a function of amyloid burden.

Effective cognitive performance often requires the allocation of additional neural resources (i.e. blood-oxygen-level-dependent [BOLD] activation) as task demands increase, and this demand-related modulation is affected by amyloid-beta deposition and normal aging. The present study investigated these complex relationships between amyloid, modulation, and cognitive function (i.e. fluid ability). Participants from the Dallas Lifespan Brain Study (DLBS, n = 252, ages 50-89) completed a semantic judgment task during functional magnetic resonance imaging (fMRI) where the judgments differed in classification difficulty. Amyloid burden was assessed via positron emission tomography (PET) using 18F-florbetapir. A quadratic relationship between amyloid standardized value uptake ratios (SUVRs) and BOLD modulation was observed such that modulation was weaker in those with moderately elevated SUVRs (e.g. just reaching amyloid-positivity), whereas those with very high SUVRs (e.g. SUVR > 1.5) showed strong modulation. Greater modulation was related to better fluid ability, and this relationship was strongest in younger participants and those with lower amyloid burden. These results support the theory that effective demand-related modulation contributes to healthy cognitive aging, especially in the transition from middle age to older adulthood, whereas high modulation may be dysfunctional in those with substantial amyloid deposition.

NCLWO: Newton’s cooling law-based weighted oversampling algorithm for imbalanced datasets with feature noise

Imbalanced datasets pose challenges to standard classification algorithms. Although oversampling techniques can balance the number of samples across different classes, the difficulties of imbalanced classification is not solely imbalanced data itself but other factors, such as small disjuncts and overlapping regions, especially in the presence of noise. Traditional oversampling techniques are not effectively address these intricacies. To this end, we propose a novel oversampling method called Newton’s Cooling Law-Based Weighted Oversampling (NCLWO). The proposed method initially calculates the weight of the minority class based on density and closeness factors to identify hard-to-learn samples, assigning them higher heat. Subsequently, Newton’s Cooling Law is applied to each minority class sample by using it as the center and expanding the sampling region outward, gradually decreasing the heat until reaching a balanced state. Finally, majority class samples within the sampling region are translated to eliminate overlapping areas, and a weighted oversampling approach is employed to synthesize informative minority class samples. The experimental study, carried out on a set of benchmark datasets, confirm that the proposed method not only outperforms state-of-the-art oversampling approaches but also shows greater robustness in the presence of feature noise.

An Adaptive Active Learning Method for Multiclass Imbalanced Data Streams with Concept Drift

Learning from multiclass imbalanced data streams with concept drift and variable class imbalance ratios under a limited label budget presents new challenges in the field of data mining. To address these challenges, this paper proposes an adaptive active learning method for multiclass imbalanced data streams with concept drift (AdaAL-MID). Firstly, a dynamic label budget strategy under concept drift scenarios is introduced, which allocates label budgets reasonably at different stages of the data stream to effectively handle concept drift. Secondly, an uncertainty-based label request strategy using a dual-margin dynamic threshold matrix is designed to enhance learning opportunities for minority class instances and those that are challenging to classify, and combined with a random strategy, it can estimate the current class imbalance distribution by accessing only a limited number of instance labels. Finally, an instance-adaptive sampling strategy is proposed, which comprehensively considers the imbalance ratio and classification difficulty of instances, and combined with a weighted ensemble strategy, improves the classification performance of the ensemble classifier in imbalanced data streams. Extensive experiments and analyses demonstrate that AdaAL-MID can handle various complex concept drifts and adapt to changes in class imbalance ratios, and it outperforms several state-of-the-art active learning algorithms.

A new data complexity measure for multi-class imbalanced classification tasks

The skewed class distribution and data complexity may severely affect the imbalanced classification results. The cost of classification can be significantly reduced if these data complexity are measured and pre-processed prior to training, particularly when dealing with large-scale and high-dimensional datasets. Although many methods have been proposed to evaluate data complexity, most of them fail to fully consider the interaction among different data characteristics, or the connection between class imbalance and these characteristics, thus posing a serious challenge to effectively evaluate the difficulty of classification. This paper presents a new data complexity measure MFII (multi-factor imbalance index), which measures the combined effects of the skewed class distribution and data characteristics on classification difficulty. In particular, it further comprehensively investigates the impact of overlap size, confusion degree, and sub-cluster structure. VoR (value of resolution) and DoC (degree of consistency) are also proposed to evaluate the resolution and interpretability of complexity measures. The experimental results demonstrate that MFII has lower VoR and a stronger correlation with classification metrics, which indicates that MFII can more accurately evaluate the difficulty of multi-class imbalanced classification tasks.

Fault diagnosis using variational autoencoder GAN and focal loss CNN under unbalanced data

For the poor model generalization and low diagnostic efficiency of fault diagnosis under imbalanced distributions, a novel fault diagnosis method using variational autoencoder generation adversarial network and improved convolutional neural network, named VGAIC-FDM, is proposed in this paper. First, to capture local features of vibration signals, continuous wavelet transform is employed to convert the original one-dimensional fault signals into wavelet time–frequency images. Second, for the data dimensionality reduction and model simplification, the time–frequency wavelet images are processed in grayscale to generate single-channel grayscale time–frequency images. Then, sample augmentation is performed on grayscale time–frequency images to balance the dataset by using a variational autoencoder generation adversarial network. Finally, the generated images and the original images are fused and trained by using a focus-loss-optimized CNN classifier to achieve fault diagnosis under unbalanced conditions. The experimental results show that the VGAIC-FDM effectively captures the potential spatial distribution of real samples and alleviates the impact caused by the inconsistent difficulty of sample classification. As a result, it enhances the fault diagnosis performance of the model when dealing with unbalanced datasets, leading to higher accuracy and F1-score values.

Robust two-stage instance-level cost-sensitive learning method for class imbalance problem

The class imbalance problem is a pervasive challenge in various applications, prompting researchers to design specialized loss functions for instance-level cost-sensitive learning models. However, the performance deteriorates when datasets are contaminated by label noise, especially in scenarios with uneven noise distribution. Additionally, specific hard-to-classify samples can further impede performance. Addressing these issues and establishing a robust instance-level cost-sensitive learning framework remains a formidable challenge. In this paper, we propose a robust Two-Stage instance-level cost-sensitive learning method with the Bounded Quadratic Type Squared Error (BQTSE) loss function, termed TSBQT. The asymmetric and bounded nature of the BQTSE loss function not only allows for distinct cost assignments to majority and minority instances, but also enhances the robustness to uneven noise. Moreover, our two-stage learning strategy prioritizes hard-to-classify instances by considering their classification difficulty. We solve TSBQT using the alternating direction method of multipliers and gradient descent algorithm. The generalization capability of TSBQT is analyzed with the Rademacher complexity theory. Extensive experiments validate the effectiveness of TSBQT.

EL-Net: An efficient and lightweight optimized network for object detection in remote sensing images

Object detection in Unmanned Aerial Vehicles (UAV) optical remote sensing imagery presents a formidable challenge in computer vision due to the minuscule size of targets, which occupy fewer pixels and provide limited feature information, complicating accurate recognition and classification. Furthermore, the overlapping of dense targets exacerbates the difficulty of precise classification and localization. Meanwhile, classical detection networks often struggle to balance recognition accuracy with model complexity. Addressing these issues, this paper introduces EL-Net, an efficient and lightweight network model based on improvements to the YOLOv7-tiny architecture. First, the network structure is streamlined through a lightweight design that maintains performance while reducing complexity. Additionally, a feature perception enhancement module (FPEM) using attention mechanisms and dilated convolution significantly improves the model’s capability to extract key features from complex backgrounds. Finally, the optimized network structure is compressed by a structured pruning algorithm. EL-Net was evaluated in challenging scenarios on the VisDrone2019 dataset, where it achieved a mean Average Precision (mAP) of 38.7%, demonstrating high detection accuracy at minimal model complexity. Meanwhile, evaluation of the UA-DETRAC dataset has demonstrated the model’s remarkable generalization capacity. The outcomes suggest that EL-Net effectively balances accuracy and efficiency, making it ideal for deployment on resource-limited mobile edge devices while offering an innovative approach to object detection in UAV imagery.

AI-Enhanced Medicinal Plant Identification System with Multilingual Social Media Integration

Sri Lanka is a country with a Ayurvedic culture which cannot be experienced anywhere in the world. This cultural system is based on a series of knowledge passed on from generations over 3000 years that could treat a variety of diseases. This traditional ayurvedic system consist of a vast herbal plant collection. Most of the information about this ayurvedic system is written in manuscripts for thousands of years. Sri Lanka lacks a proper system which is specific to ayurvedic sector is a major concern that should be addressed at present. Absence of a system has lead to problems and difficulties in identification and classification of herbal plants, to transfer knowledge about herbal plants and to conserve these ayurvedic plants for the future generation. Another concern is that Ayurvedic undergraduate students face many difficulties when gathering knowledge of these herbal plants and medicinal practices. Sri Lanka does not comprise with a full ayurvedic plant inventory system is another major concern that identified in the country. By considering all the problems an intelligent system has been recognized as a solution. The system will be based on Deep Learning, CNN, GIS, Artificial Intelligence and Machine Learning based principals to cater all the identified problems. The system will be able to identify ayurvedic plant with an image of a leave, flower, or fruit as input. And also, system will be able to classify and provide a detailed description about the identified plant including medicinal value and the distribution of the plant in the island. System will provide a crowdsourcing social media facility with both English and Sinhala languages to share information with fellow herbalist in the country.

Genre Classification of Books in Russian with Stylometric Features: A Case Study

Within the literary domain, genres function as fundamental organizing concepts that provide readers, publishers, and academics with a unified framework. Genres are discrete categories that are distinguished by common stylistic, thematic, and structural components. They facilitate the categorization process and improve our understanding of a wide range of literary expressions. In this paper, we introduce a new dataset for genre classification of Russian books, covering 11 literary genres. We also perform dataset evaluation for the tasks of binary and multi-class genre identification. Through extensive experimentation and analysis, we explore the effectiveness of different text representations, including stylometric features, in genre classification. Our findings clarify the challenges present in classifying Russian literature by genre, revealing insights into the performance of different models across various genres. Furthermore, we address several research questions regarding the difficulty of multi-class classification compared to binary classification, and the impact of stylometric features on classification accuracy.

Pedestrian Pose Recognition Based on Frequency-Modulated Continuous-Wave Radar with Meta-Learning.

With the continuous advancement of autonomous driving and monitoring technologies, there is increasing attention on non-intrusive target monitoring and recognition. This paper proposes an ArcFace SE-attention model-agnostic meta-learning approach (AS-MAML) by integrating attention mechanisms into residual networks for pedestrian gait recognition using frequency-modulated continuous-wave (FMCW) millimeter-wave radar through meta-learning. We enhance the feature extraction capability of the base network using channel attention mechanisms and integrate the additive angular margin loss function (ArcFace loss) into the inner loop of MAML to constrain inner loop optimization and improve radar discrimination. Then, this network is used to classify small-sample micro-Doppler images obtained from millimeter-wave radar as the data source for pose recognition. Experimental tests were conducted on pose estimation and image classification tasks. The results demonstrate significant detection and recognition performance, with an accuracy of 94.5%, accompanied by a 95% confidence interval. Additionally, on the open-source dataset DIAT-μRadHAR, which is specially processed to increase classification difficulty, the network achieves a classification accuracy of 85.9%.

Comparison between parameter-efficient techniques and full fine-tuning: A case study on multilingual news article classification.

Adapters and Low-Rank Adaptation (LoRA) are parameter-efficient fine-tuning techniques designed to make the training of language models more efficient. Previous results demonstrated that these methods can even improve performance on some classification tasks. This paper complements existing research by investigating how these techniques influence classification performance and computation costs compared to full fine-tuning. We focus specifically on multilingual text classification tasks (genre, framing, and persuasion techniques detection; with different input lengths, number of predicted classes and classification difficulty), some of which have limited training data. In addition, we conduct in-depth analyses of their efficacy across different training scenarios (training on the original multilingual data; on the translations into English; and on a subset of English-only data) and different languages. Our findings provide valuable insights into the applicability of parameter-efficient fine-tuning techniques, particularly for multilabel classification and non-parallel multilingual tasks which are aimed at analysing input texts of varying length.

Improving Oriented Object Detection by Scene Classification and Task-Aligned Focal Loss

Oriented object detection (OOD) can precisely detect objects with arbitrary direction in remote sensing images (RSIs). Up to now, the two-stage OOD methods have attracted more attention because of their high detection accuracy. However, the two-stage methods only rely on the features of each proposal for object recognition, which leads to the misclassification problem because of the intra-class diversity, inter-class similarity and clutter backgrounds in RSIs. To address the above problem, an OOD model combining scene classification is proposed. Considering the fact that each foreground object has a strong contextual relationship with the scene of the RSI, a scene classification branch is added to the baseline OOD model, and the scene classification result of input RSI is used to exclude the impossible categories. To focus on the hard instances and enhance the consistency between classification and regression, a task-aligned focal loss (TFL) which combines the classification difficulty with the regression loss is proposed, and TFL assigns lager weights to the hard instances and optimizes the classification and regression branches simultaneously. The ablation study proves the effectiveness of scene classification branch, TFL and their combination. The comparisons with 15 and 14 OOD methods on the DOTA and DIOR-R datasets validate the superiority of our method.

Imbalance data: The application of RUS FCM K-RBFNN Smote with XGBoost in the elderly well-being identification

Imbalanced data is a serious binary classification difficulty in forecasting the well-being of the elderly. This paper improves the Smote algorithm from the algorithm and sample dimensions to tackle the issue of imbalanced distribution of questionnaire data. The k-means Smote is combined with RBFNN as K-RBFNN Smote in the algorithm dimension and add FCM link to resample the minority set in the sample dimension as FCM K-RBFNN Smote. In order to improve the generalization of models, the RUS module is added to the algorithm. Experiments are carried out on four improved Smote technologies and two existing Smote technologies combined with XGBoost, which is superior than the other five conventional classification models. The experimental results indicate that the performance order is RUS FCM K-RBFNN Smote > K-RBFNN Smote > FCM K-RBFNN Smote > RUS K-RBFNN Smote > K-Means Smote > FCM Smote. The RUS FCM K-RBFNN method has been identified as the optimal approach for enhancing performance, resulting in a 98.58% accuracy rate. In conclusion, Smote algorithm undergoes the implementation of K-RBFNN shows greater performance and the enhancement of FCM and RUS relies on the structure of sampling.