Search Results (28)

Alzheimer’s disease and related dementias are projected to affect more than 150 million people worldwide by 2050. Early staging with validated instruments such as the Clinical Dementia Rating (CDR) scale is essential for timely intervention, yet access to clinician-administered CDR assessment remains constrained by workforce, time, and geographic barriers. This study complements a previously published machine learning pipeline for Alzheimer’s disease prediction by addressing the downstream task of dementia staging. Because the global CDR score is already derived from the six sub-domain ratings through an established rule-based procedure, the contribution reported here lies not in discovering that mapping but in encoding it in a transparent, deployable form: an explainable decision tree classifier embedded in DiAbot, a large-language-model-fronted conversational system that supports self-administered CDR-style assessment. We extracted 13,453 CDR records from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), removed administrative variables, invalid entries, and missing rows (final n = 13,290), and trained decision tree classifiers under two impurity criteria, Information Gain and Gini Index, using a 70/30 stratified record-level hold-out and ten-fold stratified record-level cross-validation. This classifier-level evaluation uses the six domain scores as recorded during ADNI’s clinician-administered assessment, not scores elicited by the DiAbot chatbot; the trained classifier was separately embedded in a web application in which a prompt-engineered large language model conducts a CDR-style interview and normalizes responses to ordinal domain scores, but the end-to-end accuracy of that full conversational pipeline (chatbot elicitation through to final CDGLOBAL) has not yet been measured, and is not what the headline accuracy figures below report. The Information Gain Decision Tree reproduced the established mapping from the six CDR sub-domain scores to the CDGLOBAL with 99.86% accuracy under the record-level hold-out protocol (matching macro-averaged precision, recall, and F1-score), with a ten-fold record-level cross-validated mean of 99.81% (SD 0.07); this result represents fidelity to the established CDR scoring rule rather than independent dementia-diagnosis accuracy. Gini-based trees performed almost identically (99.79% hold-out, 99.74% cross-validated). Memory dominated feature importance, consistent with its role as the primary domain in the official CDR scoring algorithm. Residual misclassifications were confined to adjacent CDR stages. Because the CDGLOBAL is deterministically derived from the six sub-domain scores, these figures should be read throughout as evidence of high-fidelity reconstruction of the established CDR scoring relationship, not as general dementia-diagnosis accuracy comparable to imaging- or biomarker-based classifiers; further, participant-independent generalization remains unverified under the record-level protocol evaluated here. An interpretable classifier embedded in a conversational front-end can nonetheless make standardized CDR-style staging more widely accessible while preserving clinical inspectability; the resulting system is positioned as a screening-stage adjunct to, and not a replacement for, clinician-administered CDR assessment. Full article

The prompt spread of misleading information through recent information and communication technologies (ICT) admonishes social convention and credence. Developing trustworthy algorithms that can automatically identify fake content becomes increasingly difficult. We investigate a hybrid artificial intelligence (AI) strategy that integrates machine learning (ML) and deep learning (DL) to enhance fake news detection. The model’s deep learning entity evaluates confined text arrangements and inclusive text values using a Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) with an attention layer. Conventional machine learning classifiers, mostly Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression (LR), are trained synchronously employing Term Frequency–Inverse Document Frequency (TF-IDF). A simple ensemble averaging strategy is used on both machine learning and deep learning predictions. The model demonstrates strong generalization across various text types when evaluated on the LIAR dataset and a Kaggle-style fake news dataset. The combined system performs noticeably better than each of the separate models in terms of accuracy, precision, recall, F1, and AUC. Full article

Large-scale symbolic melody datasets are essential for data-driven music information retrieval and generation, yet traditional-style Chinese melodies remain scattered across heterogeneous score formats and image sources. Existing extraction pipelines typically focus on single modalities—either MIDI archives or standard staff notation—and lack unified handling for numbered musical notation (Jianpu) and automated quality assurance. We propose the Multi-Source Melody Pipeline (MSMP), a systems-integration prototype whose front-end admits MIDI, MusicXML, Jianpu images, and staff images, and whose back-end converges on a standardized event-level representation; the present case study exercises the image branch—in particular the Jianpu branch, through a Gemini-2.5-flash vision language model—and treats the MIDI/MusicXML ingestion paths as architectural slots that are wired in but not experimentally validated in this submission. The system employs notation-aware routing to direct score images to appropriate backends (a VLM for Jianpu and rule-based OMR for staff) and enforces a structural validity gate (schema conformance plus at least one melodic track with at least one musical event) on every candidate segment. Validation on a 292-page representative prototype cohort yielded an 80.1% structural-acceptance rate—explicitly not a transcription accuracy number—and a newly added ground-truth benchmark on 50 manually annotated Jianpu pages reports 95.8% time-signature exact accuracy, 77.1% tonal-pitch-class key accuracy, 100% tempo agreement within $\pm 5$ BPM, and, on a 10-page note-level subset, a mean first-16-note pitch F1 of 0.898 (octave-sensitive) with a Symbol Error Rate of 0.150. A companion 10-page K = 3 self-consistency audit indicates that metadata errors are systematic rather than stochastic. This work, therefore, contributes a reproducible integration architecture and a quantitative baseline on the Jianpu branch, rather than a new OMR algorithm, a new dataset release, or a fully benchmarked multi-format corpus; ongoing work addresses out-of-distribution classifier evaluation, comparison against dedicated Jianpu OMR baselines, and release of a copyright-cleared corpus. Full article

Background/Objectives: Variations in staining styles—arising from differences in tissue preparation, scanners, and laboratory protocols—severely compromise the robustness of automated cell segmentation algorithms in digital pathology. Moreover, manual nucleus annotation is extremely labor-intensive, leading to a scarcity of large-scale, fully annotated datasets for supervised nucleus segmentation. This study proposes a novel framework that simultaneously mitigates staining variability and achieves high-accuracy nucleus segmentation using only minimal annotations. Methods: We present 3SGAN, a multi-task dual-branch generative adversarial network (GAN) that jointly performs stain normalization and nucleus segmentation in a semi-supervised manner. The framework adopts a teacher–student paradigm: a lightweight teacher model (AttCycle) equipped with attention gates generates reliable pseudo-labels, while a high-capacity student model (TransCycle) leveraging a hybrid CNN–Transformer architecture further refines performance. 3SGAN was trained and evaluated on a large dataset of 1408 Whole-Slide Images (WSIs) from two medical institutions, encompassing 101 distinct staining styles, with nucleus-level annotations required for only 5% of the data. Results: 3SGAN significantly outperformed state-of-the-art methods, achieving superior segmentation accuracy with an F1-score of 0.8140, mean IoU of 0.8201, and AJI of 0.6915. Simultaneously, it demonstrated substantial improvements in stain normalization quality, yielding a low RMSE of 0.0908, high PSNR of 21.0615, and SSIM of 0.8556 on the internal test set. External validation on independent MoNuSeg and PanNuke datasets, as well as on previously untested tumor-rich non-ROI regions from our in-house WSIs, confirmed strong generalizability with excellent stain normalization and top-tier segmentation accuracy across diverse staining protocols, tissue types, and pathological patterns. Conclusions: The proposed 3SGAN framework demonstrates that high-performance nucleus segmentation and stain normalization can be achieved with minimal annotation requirements, offering a practical and scalable solution for digital pathology applications across diverse clinical settings and staining protocols. Full article

Stroke is a leading cause of death and disability. However, brain computed tomography image classification using machine learning and deep learning algorithms frequently suffers from a class imbalance problem, making it difficult to effectively extract deep-detailed features from instances of minority stroke lesions. In this study, we systematically implement three style-based generative adversarial network (StyleGAN)-based data augmentation approaches: StyleGAN2, StyleGAN3, and conditional StyleGAN3 to address class imbalance in brain stroke classification. Furthermore, we deploy an ensemble learning-based deep neural network to enhance the effect of those data augmentation algorithms on downstream classification tasks. Experimental results show that StyleGAN3 effectively outperforms the other two StyleGAN data augmentation approaches in terms of precision, recall, and F1-score when addressing highly imbalanced brain stroke classification. Overall, this paper demonstrates the efficacy of three StyleGAN-based data augmentation approaches in addressing imbalanced brain stroke detection. Full article

Direct methods for solving protein crystal structures from X-ray diffraction data provide an essential approach for validating predicted models while avoiding external model bias. Nevertheless, traditional iterative projection algorithms, including the widely used Difference Map (DiffMap), are often limited by modest phase retrieval success rates. To address this limitation, we introduce a novel Hybrid Difference Map (HDM) algorithm that synergistically combines the strengths of DiffMap and the Hybrid Input–Output (HIO) method through six distinct iterative update rules. HDM retains an optimized DiffMap-style relaxation term for fine-grained density modulation in protein regions while adopting HIO’s efficient negative feedback mechanism for enforcing the solvent flatness constraint. Using the transmembrane photosynthetic reaction center 2uxj as a test case, the first HDM formula, HDM-f1, successfully recovered an atomic-resolution structure directly from random phases under a conventional full-resolution phasing scheme, demonstrating the robust phasing capability of the approach. Systematic evaluation across 22 protein crystal structures (resolution 1.5–3.0 Å, solvent content ≥ 60%) revealed that all six HDM variants outperformed DiffMap, achieving 1.8–3.5× higher success rates (average 2.8×), performing on par with or exceeding HIO under a conventional phasing scheme. Further performance gains were achieved by integrating HDM with advanced strategies: resolution weighting and a genetic algorithm-based evolutionary scheme. The genetic evolution strategy boosted the success rate to nearly 100%, halved the median number of iterations required for convergence, and reduced the final phase error to approximately 35° on average across test structures through averaging of multiple solutions. The resulting electron density maps were of high interpretability, enabling automated model building that produced structures with a backbone RMSD of less than 0.5 Å when compared to their PDB-deposited counterparts. Collectively, the HDM algorithm suite offers a robust, efficient, and adaptable framework for direct phasing, particularly for challenging cases where conventional methods struggle. Our implementation supports all space groups providing an accessible tool for the broader structural biology community. Full article

Traditional scoring of makeup face templates in beauty skill assessments heavily relies on manual judgment, leading to inconsistencies and subjective bias. Hand-drawn templates often exhibit proportion distortions, asymmetry, and occlusions that reduce the accuracy of conventional facial landmark detection algorithms. This study proposes a novel approach that integrates Geometric Feature Enhancement (GFE) with Dlib’s 68-landmark detection to improve the robustness and precision of landmark localization. A comprehensive comparison among Haar Cascade, MTCNN-MobileNetV2, and Dlib was conducted using a curated dataset of 11,600 hand-drawn facial templates. The proposed GFE-enhanced Dlib achieved 60.5% accuracy—outperforming MTCNN (23.4%) and Haar (20.3%) by approximately 37 percentage points, with precision and F1-score improvements exceeding 20% and 25%, respectively. The results demonstrate that the proposed method significantly enhances detection accuracy and scoring consistency, providing a reliable framework for automated beauty skill evaluation, and laying a solid foundation for future applications such as digital archiving and style-guided synthesis. Full article

Behavior recognition and multi-object tracking of dairy cows in free-style cow barn environments play a crucial role in monitoring their health status and serve as an essential means for intelligent scientific farming. This study proposes an efficient embedded algorithm, MTW-BYTE, for dairy cow behavior recognition and multi-object tracking. It addresses challenges in free-style cow barn environments, including the impact of lighting variations and common occlusions on behavior recognition, as well as trajectory interruptions and identity ID switching during multi-object tracking. First, the MTW-YOLO cow behavior recognition model is constructed based on the YOLOv11n object detection algorithm. Replacing parts of the backbone network and neck network with MANet and introducing the Task Dynamic Align Detection Head (TDADH). The CIoU loss function of YOLOv11n is replaced with the WIoU loss. The improved model not only effectively handles variations in lighting conditions but also addresses common occlusion issues in cows, enhancing multi-scale behavior recognition capabilities and improving overall detection performance. The improved MTW-YOLO algorithm improves Precision, Recall, mAP50 and F1 score by 4.5%, 0.1%, 1.6% and 2.2%, respectively, compared to the original YOLOv11n model. Second, the ByteTrack multi-object tracking algorithm is enhanced by designing a dynamic buffer and re-detection mechanism to address cow trajectory interruptions and identity ID switching. The MTW-YOLO algorithm is cascaded with the improved ByteTrack to form the multi-target tracking algorithm MTW-BYTE. Compared with the original multi-target tracking algorithm YOLOv11n-ByteTrack (a combination of YOLOv11n and the original ByteTrack), this algorithm improves HOTA by 1.1%, MOTA by 3.6%, MOTP by 0.2%, and IDF1 by 1.9%, reduces the number of ID changes by 11, and achieves a frame rate of 43.11 FPS, which can meet the requirements of multi-target tracking of dairy cows in free-style cow barn environments. Finally, to verify the model’s applicability in real-world scenarios, the MTW-BYTE algorithm is deployed on an NVIDIA Jetson AGX Orin edge device. Based on real-time monitoring of cow behavior on the edge device, the pure inference time for a single frame is 16.62 ms, achieving an FPS of 29.95, demonstrating efficient and stable real-time behavior detection and tracking. The ability of MTW-BYTE to be deployed on edge devices to identify and continuously track cow behavior in various scenarios provides hardware feasibility verification and algorithmic support for the subsequent deployment of intelligent monitoring systems in free-style cow barn environments. Full article

Providing students with effective learning resources is essential for improving educational outcomes—especially in complex and conceptually diverse fields such as Mathematics and Computer Science. To better understand how these subjects are communicated, this study investigates the linguistic structures embedded in academic texts from selected subfields within both disciplines. In particular, we focus on meta-languages—the linguistic tools used to express definitions, axioms, intuitions, and heuristics within a discipline. The primary objective of this research is to identify which subfields of Mathematics and Computer Science share similar meta-languages. Identifying such correspondences may enable the rephrasing of content from less familiar subfields using styles that students already recognize from more familiar areas, thereby enhancing accessibility and comprehension. To pursue this aim, we compiled text corpora from multiple subfields across both disciplines. We compared their meta-languages using a combination of supervised (Neural Network) and unsupervised (clustering) learning methods. Specifically, we applied several clustering algorithms—K-means, Partitioning around Medoids (PAM), Density-Based Clustering, and Gaussian Mixture Models—to analyze inter-discipline similarities. To validate the resulting classifications, we used XLNet, a deep learning model known for its sensitivity to linguistic patterns. The model achieved an accuracy of 78% and an F1-score of 0.944. Our findings show that subfields can be meaningfully grouped based on meta-language similarity, offering valuable insights for tailoring educational content more effectively. To further verify these groupings and explore their pedagogical relevance, we conducted both quantitative and qualitative research involving student participation. This paper presents findings from the qualitative component—namely, a content analysis of semi-structured interviews with software engineering students and lecturers. Full article

The opioid epidemic in the United States has significantly impacted pregnant women with opioid use disorder (OUD), leading to increased health and social complications. This study explores the feasibility of using machine learning algorithms with consumer-grade smartwatches to identify medication-taking gestures. The research specifically focuses on treatments for OUD, investigating methadone and buprenorphine taking gestures. Participants (n = 16, all female university students) simulated medication-taking gestures in a controlled lab environment over two weeks, with data collected via Ticwatch E and E3 smartwatches running custom ASPIRE software. The study employed a RegNet-style 1D ResNet model to analyze gesture data, achieving high performance in three classification scenarios: distinguishing between medication types, separating medication gestures from daily activities, and detecting any medication-taking gesture. The model’s overall F1 scores were 0.89, 0.88, and 0.96 for each scenario, respectively. These findings suggest that smartwatch-based gesture recognition could enhance real-time monitoring and adherence to medication regimens for OUD treatment. Limitations include the use of simulated gestures and a small, homogeneous participant pool, warranting further real-world validation. This approach has the potential to improve patient outcomes and management strategies. Full article

Secondary systems in electrical engineering often rely on traditional CAD software (AutoCAD v2024.1.6) or non-structured, paper-based diagrams for fieldwork, posing challenges for digital transformation. Electrical diagram recognition technology bridges this gap by converting traditional diagram operations into a “digital” model, playing a critical role in power system scheduling, operation, and maintenance. However, conventional recognition methods, which primarily rely on partition detection, face significant limitations such as poor adaptability to diverse diagram styles, interference among recognition objects, and reduced accuracy in handling complex and varied electrical diagrams. This paper introduces a novel layered framework for electrical diagram recognition that sequentially extracts the element layer, text layer, and connection relationship layer to address these challenges. First, an improved YOLOv7 model, combined with a multi-scale sliding window strategy, is employed to accurately segment large and small diagram objects. Next, PaddleOCR, trained with electrical-specific terminology, and PaddleClas, using multi-angle classification, are utilized for robust text recognition, effectively mitigating interference from diagram elements. Finally, clustering and adaptive FcF-inpainting algorithms are applied to repair the connection relationship layer, resolving local occlusion issues and enhancing the overall coupling of the diagram. Experimental results demonstrate that the proposed method outperforms existing approaches in robustness and universality, particularly for complex diagrams, providing technical support for intelligent power grid construction and operation. Full article

Line charts are prevalent in scientific documents and commercial data visualization, serving as essential tools for conveying data trends. Automatic detection and tracing of line paths in these charts is crucial for downstream tasks such as data extraction, chart quality assessment, plagiarism detection, and visual question answering. However, line graphs present unique challenges due to their complex backgrounds and diverse curve styles, including solid, dashed, and dotted lines. Existing curve detection algorithms struggle to address these challenges effectively. In this paper, we propose ChartLine, a novel network designed for detecting and tracing curves in line graphs. Our approach integrates a Spatial-Sequence Attention Feature Pyramid Network (SSA-FPN) in both the encoder and decoder to capture rich hierarchical representations of curve structures and boundary features. The model incorporates a Spatial-Sequence Fusion (SSF) module and a Channel Multi-Head Attention (CMA) module to enhance intra-class consistency and inter-class distinction. We evaluate ChartLine on four line chart datasets and compare its performance against state-of-the-art curve detection, edge detection, and semantic segmentation methods. Extensive experiments demonstrate that our method significantly outperforms existing algorithms, achieving an F-measure of 94% on a synthetic dataset. Full article

The accurate detection of wind power outliers plays a crucial role in wind power forecasting, while the inherited strong randomness and high fluctuations bring great challenges to this issue. This work investigates the way to improve the outlier detection accuracy based on support vector machine (SVM). Although SVM can achieve good results for outlier detection in theory, its performance is heavily dependent on the hyper-parameters. Parameter optimization is not an easy task due to its complex nonlinear multi-optimum nature; an improved Harris hawk optimization (IHHO) is proposed to optimize the parameters of SVM for more accurate outlier detection. HHO takes the cooperative behavior and chasing style of Harris’ hawks in nature called surprise pounce and can effectively search the optimal one in large parameter space, but it tends to fall into local optimum. To solve this issue, an improved Harris hawk optimization algorithm (IHHO) was proposed to obtain the optimal parameters of SVM. First, Hammersley sequence initialization is carried out to acquire good initial solutions. Then, a nonlinear factor control mode and an adaptive Gaussian–Cauchy mutation perturbation strategy are proposed to avoid getting trapped in local optima. In this way, a novel wind power outlier detection method named IHHO-SVM was constructed. The results on several wind power data with outliers show that IHHO-SVM outperforms SVM and HHO-SVM, which achieves the highest average F1 score of 96.63% and exhibits the smallest standard deviation. Compared to commonly used models for detecting outliers in wind power, such as isolation forest (IF), local outlier factor (LOF), SVM with grey wolf optimization (GWO-SVM), and SVM with particle swarm optimization (PSO-SVM), the proposed IHHO-SVM model shows the best overall performance with precision, recall, and F1 scores of 95.76%, 96.94%, and 96.35%, respectively. Full article

Nowadays, the dissemination of news information has become more rapid, liberal, and open to the public. People can find what they want to know more and more easily from a variety of sources, including traditional news outlets and new social media platforms. However, at a time when our lives are glutted with all kinds of news, we cannot help but doubt the veracity and legitimacy of these news sources; meanwhile, we also need to guard against the possible impact of various forms of fake news. To combat the spread of misinformation, more and more researchers have turned to natural language processing (NLP) approaches for effective fake news detection. However, in the face of increasingly serious fake news events, existing detection methods still need to be continuously improved. This study proposes a modified proof-of-concept model named NER-SA, which integrates natural language processing (NLP) and named entity recognition (NER) to conduct the in-domain and cross-domain analysis of fake news detection with the existing three datasets simultaneously. The named entities associated with any particular news event exist in a finite and available evidence pool. Therefore, entities must be mentioned and recognized in this entity bank in any authentic news articles. A piece of fake news inevitably includes only some entitlements in the entity bank. The false information is deliberately fabricated with fictitious, imaginary, and even unreasonable sentences and content. As a result, there must be differences in statements, writing logic, and style between legitimate news and fake news, meaning that it is possible to successfully detect fake news. We developed a mathematical model and used the simulated annealing algorithm to find the optimal legitimate area. Comparing the detection performance of the NER-SA model with current state-of-the-art models proposed in other studies, we found that the NER-SA model indeed has superior performance in detecting fake news. For in-domain analysis, the accuracy increased by an average of 8.94% on the LIAR dataset and 19.36% on the fake or real news dataset, while the F1-score increased by an average of 24.04% on the LIAR dataset and 19.36% on the fake or real news dataset. In cross-domain analysis, the accuracy and F1-score for the NER-SA model increased by an average of 28.51% and 24.54%, respectively, across six domains in the FakeNews AMT dataset. The findings and implications of this study are further discussed with regard to their significance for improving accuracy, understanding context, and addressing adversarial attacks. The development of stylometric detection based on NLP approaches using NER techniques can improve the effectiveness and applicability of fake news detection. Full article

Humans often demonstrate diverse behaviors due to their personal preferences, for instance, related to their individual execution style or personal margin for safety. In this paper, we consider the problem of integrating both path and velocity preferences into trajectory planning for robotic manipulators. We first learn reward functions that represent the user path and velocity preferences from kinesthetic demonstration. We then optimize the trajectory in two steps, first the path and then the velocity, to produce trajectories that adhere to both task requirements and user preferences. We design a set of parameterized features that capture the fundamental preferences in a pick-and-place type of object transportation task, both in the shape and timing of the motion. We demonstrate that our method is capable of generalizing such preferences to new scenarios. We implement our algorithm on a Franka Emika 7-DoF robot arm and validate the functionality and flexibility of our approach in a user study. The results show that non-expert users are able to teach the robot their preferences with just a few iterations of feedback. Full article