Search Results (5,058)

Recidivism prediction is increasingly embedded in criminal justice decision-making, yet most deployed systems remain opaque and have been shown to exhibit discriminatory behavior against certain demographic groups. This paper presents a fairness-aware interpretable framework for recidivism prediction applied to three real-world datasets from Bulgaria, Greece, and Portugal. The classification core relies on a 1-Dimensional Convolutional Neural Network (1D-CNN), trained by a custom objective function that embeds the Equalized Odds fairness criterion as an L1-regularized penalty reflecting on gender-based disparities in false positive and false negative rates. Model-level interpretability is provided through Kernel SHAP, which decomposes individual predictions into additive feature attributions grounded in cooperative game theory. Experiments across prediction tasks, each evaluated over randomized runs, demonstrate that the baseline model exhibits statistically significant bias against the female group in all datasets. The fairness-constrained model substantially reduces these disparities across all tasks at a moderate and expected cost to classification accuracy. Kernel SHAP analysis reveals the relative contribution of static and dynamic offenders’ attributes to individual risk scores, supporting auditability and contestability. The proposed framework advances the integration of predictive performance, algorithmic fairness, and structural interpretability in criminal justice analytics. Full article

To address the dilemma of ‘non-grain use of cultivated land’ and support China’s requisition–compensation balance policy, this study developed a multi-dimensional assessment framework integrating the production, ecological, and economic dimensions (3D evaluation model), using Shenyang City as a case study to demonstrate the framework’s operational application and policy relevance. Based on 34,704 Third National Land Survey (TNLS) parcels (27,408.39 ha), we applied the constraint factor assessment method and entropy-weighted composite index model. The results show that non-cultivated agricultural land (NCAL) is generally marginally suitable (citywide average score: 2.50/4), with highly suitable areas accounting for only 4.04% (1106.30 ha). These areas exhibit a triangular spatial pattern distributed across northeastern Faku County, central Sujiatun District, and southern Xinmin City. Sensitivity tests using equal weights and ±20% dimension-weight perturbations confirm that high-suitability area remains limited (3.37–5.63% under entropy-weight scenarios; 8.54% under equal weights). Primary limiting factors include severe organic matter deficiency (average 19 g/kg), shallow soil depth, unfavorable pH, land requiring engineering restoration (94%), and punctiform heavy metal contamination (7.53% of plots, 2065.05 ha as spatially excluded areas). Consequently, we propose a five-tier sequential restoration framework: (1) near-term priority recultivation of highly suitable areas; (2) mid-term topsoil reconstruction for moderately suitable areas; (3) medium-to-long-term topsoil stripping and thickening for low-suitability areas; (4) long-term soil amelioration and slope-to-terrace conversion for marginally suitable areas; and (5) strict prohibition of restoration in unsuitable areas. This study establishes a spatially explicit decision-making system integrating “evaluation–classification–sequencing”, and distinguishes technical suitability from economic, institutional, and policy feasibility, providing a decision-support framework for scientifically implementing the cultivated land requisition–compensation balance policy. Future empirical studies using post-restoration monitoring data are needed to test its predictive accuracy against observed restoration outcomes. Full article

As a next-generation automotive display technology, Augmented Reality Head-Up Display (AR-HUD) has demonstrated immense potential in reshaping driving safety and enhancing the human–computer interaction experience. To address the challenges of barrel distortion and perspective distortion inherent in HUD systems, we propose a joint-learning-based dual-path pre-correction method. This approach employs a shared encoder to extract image features, which are then decoupled into two parallel branches: a classification branch and a distortion flow prediction branch. Building upon this architecture, a model-fitting method is introduced to estimate the distortion model parameters in the parameter space using the predicted distortion types and flows, thereby reconstructing a refined distortion flow. Finally, image rectification is achieved through a resampling method. On the ARHDD dataset, the proposed method achieves a PSNR of 24.617 dB (barrel) and 25.062 dB (perspective), an SSIM of 0.845 and 0.873, and an NRMSE of 0.163 and 0.157, respectively. On the Places 365 dataset, it achieves a PSNR of 23.914 dB (barrel) and 21.870 dB (perspective), an SSIM of 0.812 and 0.748, and an NRMSE of 0.174 and 0.211, respectively. Both quantitative and qualitative comparative experiments against other state-of-the-art methods demonstrate that the proposed approach achieves superior correction performance for both types of distortion. Finally, the simulation verification of the HUD system proved that this correction method demonstrated excellent potential, but further verification is still needed in a real or semi-real environment. Full article

Facial emotion recognition (FER) in real-world settings is limited by the semantic mismatch between discrete emotion categories and continuous Valence–Arousal–Dominance (V-A-D) dimensions and the lack of reliable uncertainty estimates under incomplete facial evidence. Existing uncertainty-aware FER studies mainly address annotation ambiguity or training-time reliability, leaving the behavior of predictive uncertainty under progressive input degradation insufficiently examined. This paper proposes BGDC (Bayesian Gaussian-mixture Distributional Consistency), a multi-task FER framework that integrates a GMM-based soft consistency module with a context-conditioned Bayesian regression head and explicitly models aleatoric and epistemic uncertainty. To evaluate predictive reliability, a controlled masking protocol is introduced to remove facial information under different spatial configurations. On FER2013-VAD, BGDC attains the highest classification accuracy of 0.6943 and the highest mean V-A-D CCC of 0.6079 among the compared configurations, and it yields a stronger epistemic uncertainty-error correspondence than MC Dropout in a single-model setting. Controlled masking further shows that the epistemic uncertainty of BGDC tracks task-relevant facial information loss rather than masking ratio alone: it rises with regression error when diagnostically important regions are removed, and it contracts when the masked region is largely task-irrelevant. Combining Bayesian uncertainty with the GMM-based distributional prior thus enables reliability-aware multi-task FER, in which controlled masking serves as a diagnostic intervention rather than as a benchmark of accuracy degradation alone. Full article

Background: Dedicated artificial intelligence (AI) systems for fracture detection already exist, yet general-purpose multimodal models are increasingly accessible to clinicians despite not being developed or formally validated as medical devices. Their behavior in focused orthopedic imaging tasks remains insufficiently characterized. Purpose: To characterize how two accessible general-purpose multimodal models interpret AP pelvis radiographs with hip fractures, focusing on context dependence, overconfidence, and complementary error patterns within a surgically confirmed positive-only cohort. This was a behavioral characterization study of a fracture-positive cohort, not a diagnostic accuracy evaluation. Methods: In April 2026, we retrospectively studied 214 surgically confirmed hip fractures on AP pelvis radiographs using two general-purpose multimodal models under six prompting conditions. In runs A–D, the models were explicitly told that a hip fracture was present and were asked to classify it; in runs E–F, they were not told whether a hip fracture was present. Each image was rerun de novo in a separate chat session through vendor APIs using a fixed base prompt and no image preprocessing. We recorded hip-fracture detection, correct laterality, coarse fracture pattern, intracapsular displacement, AO/OTA grading, subtrochanteric identification, and self-reported confidence. Because the cohort contained hip fractures only, we report fracture-detection rates and classification performance within a positive-only cohort rather than full diagnostic-accuracy metrics. Results: Using the more conservative endpoint of hip-fracture detection with correct laterality, GPT-5.4 was correct in 79.0% and 86.4% of cases in runs E and F, whereas Gemini was correct in 80.4% and 93.5%, respectively. When outputs from both models were combined, this endpoint reached 89.7% in run E and 96.7% in run F, indicating complementary rather than redundant error patterns. Incorrect laterality cues markedly degraded performance, from 90.7% to 66.4% in GPT-5.4 and from 97.7% to 57.0% in Gemini. Performance remained limited for treatment-relevant subtyping, particularly AO/OTA grading and subtrochanteric identification. Both models frequently remained highly confident when wrong, and self-reported confidence did not reliably distinguish correct from incorrect outputs. Conclusions: Accessible general-purpose multimodal models showed partial capability for coarse hip-fracture interpretation, but they remained context-sensitive, unreliable for treatment-relevant subtyping, and highly confident even when incorrect. Their complementary error patterns are hypothesis-generating rather than evidence of clinical readiness. On the basis of these findings, we do not support unvalidated or uncontrolled clinical use of such models. As access to these tools expands, explicit usage boundaries, minimum performance expectations, repeated local revalidation, and sustained human oversight become increasingly necessary. Full article

Background: Childhood oral hygiene behaviors are crucial to preventing oral diseases and can influence the development and progression of malocclusions. The World Health Organization (WHO) Oral Health Questionnaire is a standardized tool for assessing oral hygiene behaviors, oral health-related behaviors, and preventive dental awareness in children. Aim: This study aimed to assess oral hygiene behaviours and examine associations between WHO Oral Health Questionnaire variables and Angle malocclusion classes among children aged 6–9 years. Materials and Methods: This cross-sectional study included 90 children aged 6–9 years from the Pristina region, Kosovo. Data were collected using the WHO Oral Health Questionnaire for Children, which assessed oral hygiene habits, toothbrushing frequency, fluoride awareness, dental attendance, dietary behaviors, oral symptoms, and oral-health-related quality of life. Malocclusion was classified according to Angle classification into Class I, II, and III malocclusions with 3D intraoral scanners, Aerolscan 3. Descriptive statistical analysis, Chi-square (χ²) test, Spearman correlation analysis, and reliability analysis using Cronbach’s Alpha were performed using SPSS Statistics 23.0 (IBM Corp., Armonk, NY, USA) and Statistica 7.1 (StatSoft Inc., Tusla, OK, USA). Results: Most participants reported regular oral hygiene practices, with 46.7% brushing their teeth two or more times daily. However, limited awareness regarding fluoride-containing toothpaste was observed, as most children answered “don’t know” regarding fluoride use. Occasional toothache or oral discomfort was reported by 33.3% of participants, while 23.3% reported dissatisfaction with dental appearance. Difficulty biting hard foods was present in 34.4% of children. Reliability analysis of the Q10 section demonstrated moderate internal consistency (Cronbach’s Alpha = 0.500). Chi-square analysis demonstrated no statistically significant association between Angle malocclusion classes and WHO questionnaire variables (p > 0.05). The highest χ² value was observed for tooth-cleaning frequency (Q7) (χ² = 11.97; p = 0.152), although the association remained statistically non-significant. Psychosocial impact questions and oral health-related quality of life questions also demonstrated no statistically significant association with malocclusion classes. Conclusions: oral hygiene practices, preventative oral health practices, and oral health-related experiences were comparatively similar among children in different Angle malocclusion classes. Although there were no statistically significant correlations found between malocclusion classes and WHO questionnaire variables, the results show that some children have psychosocial concerns about their dental appearance and insufficient awareness of preventive oral health. The WHO Oral Health Questionnaire is a useful epidemiological tool for evaluating pediatric oral health behaviors and may help build youth orthodontic and preventive oral health policies. Full article

Optical performance monitoring is essential for adaptive and intelligent coherent optical communication systems. In this paper, a Transformer-based multi-task meta-learning framework is proposed for joint modulation format identification and electrical signal-to-noise ratio (ESNR) estimation from original received waveforms. A simulated coherent optical communication system is established to generate QPSK, 16QAM, and 32QAM signals under different launch-power conditions. The received I/Q waveforms are directly used as model inputs, avoiding handcrafted feature extraction or constellation-image conversion. The proposed model employs a shared one-dimensional Transformer encoder to extract temporal waveform representations. A prototypical classification branch is used for few-shot modulation format identification, while an ESNR regression branch is introduced for continuous signal-quality estimation. The two tasks are jointly optimized under an episodic support-query training mechanism. Experimental results show that the proposed method achieves 99.99% modulation identification accuracy on the test episodes. For ESNR estimation, the model obtains an MAE of 0.1194 dB, an RMSE of 0.1738 dB, and an R² value of 99.83%. These results demonstrate that the proposed framework can simultaneously provide accurate modulation decisions and reliable ESNR estimation, showing its potential for waveform-based optical performance monitoring. Full article

Non-Intrusive Load Monitoring (NILM) enables estimation of the energy use of individual appliances in smart buildings from a single aggregate meter. In practice, however, this task is not straightforward. Signals from different appliances can overlap, and the measured data may also include distortions such as spikes, drops, and noise. To address these issues, this study presents a multi-task triple-hybrid deep learning framework that handles appliance classification and anomaly detection together. The model brings together 1D-CNN, BiLSTM, and Transformer Attention so that local patterns, temporal dependencies, and wider contextual information can be learned within the same structure. It also uses a dual-output design to classify appliance categories and detect anomaly types simultaneously. Experiments were carried out on Building 1 of the UK-DALE dataset with four appliances: kettle, microwave, washer dryer, and fridge freezer. For the anomaly task, synthetic disturbances were added to segmented signal windows and grouped as normal, spike, drop, and noise. To check how well the proposed framework handled different scenarios, it was tested on both the UK-DALE and REDD datasets. Looking at the main UK-DALE results, the model correctly identified appliances 99.48% of the time and spotted anomalies with 98.80% accuracy. A secondary test on the REDD dataset yielded an 86.44% classification score. This proves the architecture can adjust to completely new power grid environments without losing its edge. On top of that, when pitted against standard benchmark models like Seq2Point, this triple-hybrid design clearly does a better job of mapping out complex signal changes. As a result, it yields much stronger anomaly detection metrics. Full article

In precision livestock farming, 3D point clouds provide important data support for analyzing pig behavior and monitoring their health. However, due to environmental occlusions, limited sensor viewpoints, and mutual shielding between pigs, the acquired point clouds are often severely partial, which affects the accuracy of body shape modeling and behavior recognition. To address these challenges, this study constructed a pig pose point cloud dataset using multi-view depth camera acquisition and point cloud registration techniques. Based on this dataset, an improved point cloud completion model, IIR-PoinTr, is proposed to enhance the reconstruction of geometric and topological structures in pig bodies. By strengthening local geometric perception and high-dimensional feature representation, the model improves the reconstruction quality of partial pig point clouds and produces more structurally consistent pig body shapes. Experimental results show that, on the self-constructed pig posture dataset, the proposed method reduces Chamfer Distance (CD-L1) by 3.6%, CD-L2 by 6.9%, and Earth Mover’s Distance (EMD) by 2.0%, while improving the F-score by 5.4% compared with the baseline model. In single-view point cloud completion tasks, the method is capable of reconstructing geometrically consistent pig body structures and increases downstream classification accuracy by 34.9%. These results indicate that the proposed method can improve the reconstruction quality of partial pig point clouds and provide preliminary technical support for posture analysis under occlusion. Full article

Accurate waveform classification in noisy environments is an important task in modern communications, radar signal analysis, biomedical signal interpretation, industrial monitoring and other signal processing systems. This paper investigates the performance of three neural network architectures: Multilayer Perceptron (MLP), one-dimensional Convolutional Neural Network (1D-CNN), and Long Short-Term Memory (LSTM) for multiclass waveform classification in the presence of Additive White Gaussian Noise (AWGN). A time series dataset consisting of multiple waveform classes is generated and corrupted with AWGN across a wide range of signal-to-noise ratio (SNR) levels to simulate noisy signal distortion conditions. The three models are trained and evaluated under identical conditions to ensure a fair comparison. Their classification performance is evaluated in terms of accuracy, Confusion Matrix (CM), Receiver Operating Characteristic (ROC) curve and the Area Under the ROC curve (AUC) across varying SNR values. Simulation results demonstrate that the 1D-CNN effectively captures local temporal patterns and achieves superior robustness in classification at moderate and high SNR levels. The LSTM model demonstrates the ability to capture temporal dependencies in sequential waveform data but exhibits sensitivity to waveform variations due to amplitude, phase and frequency changes and noise at lower SNR values. The MLP, although computationally simpler, shows comparatively limited performance in low-SNR conditions due to its lack of temporal feature extraction capability. For the case of multiclass deterministic waveforms, the accuracy of classification for the 1D-CNN and LSTM is nearly 100% at SNR = 5 dB showing their robustness in classification, whereas the accuracy of MLP is approximately 70% that shows poor classification in noisy conditions. When there is random amplitude, frequency and phase variations in the waveforms, the accuracy of the 1D-CNN and MLP increases with SNR, and 1D-CNN superior to MLP. However, the LSTM accuracy fails to improve with SNR, resulting in poor classification performance in such a scenario. The results provide an insight into the suitability of different neural architectures for waveform classification tasks in noisy communication or other time series applications and highlight the advantages of convolutional feature extraction for robust signal recognition. Full article

Vision Transformer (ViT) has been widely adopted in the computer vision community. However, the standard ViT often contains many parameters, usually performs poorly when trained from scratch on medium-scale datasets, and does not explicitly preserve the local spatial and channel-wise structures within each token. This work proposes a novel model called the Token-Shared Convolutional Projection Vision Transformer (TSCP-ViT). The core idea of TSCP-ViT is to integrate convolutional layers into the multi-head attention mechanism and to apply the same convolutional operation independently to each token, where each token exhibits spatial 2D multi-channel characteristics. In addition, this work introduces a Transformer decoder immediately after each Transformer encoder, enabling the classification tokens to aggregate information from all tokens and be updated using statistical information. Moreover, a trainable Non-Reversing Gate GELU (NRG-GELU) activation is also proposed. Comparative experiments on CIFAR-100, Food-101, and ImageNet100 show that, under comparable parameter counts and without pretraining or knowledge distillation, TSCP-ViT substantially surpasses ViT, outperforms CvT, outperforms ResNet on Food-101, and approaches ResNet on CIFAR-100 and ImageNet100, although with considerably higher FLOPs. Full article

The digitization of existing building stock often depends on legacy 2D raster floor plans (scanned drawings, PDF exports, or photographs) because structured building information models are frequently unavailable for older properties. Manual measurement and visual inspection of such documents are time consuming and error prone. This paper presents an integrated deep learning pipeline that extracts semantic information from unstructured two-dimensional floor plan images of existing structures and supports preliminary compliance screening via locally deployed large language models. The pipeline employs YOLOv8 for the localization and classification of 18 architectural symbols and furniture items, and a U-Net with a ResNet34 encoder for the semantic segmentation of walls and interior room spaces. To translate pixel-level predictions into physical metrics, we implement an area calculation module based on user-defined reference scale calibration. An LLM evaluation module, deployed locally via Ollama with a retrieval-augmented generation pipeline, interprets extracted room metrics and flags potential non-compliance against referenced residential design guidelines; it is intended for the assessment of existing layouts rather than generative co-design. We expand a core dataset of 101 manually annotated source floor plans to 303 augmented instances using label-aligned geometric transformations, while reporting generalization in terms of the 101 unique source plans. On the held-out validation split (10 source plans), YOLOv8 achieves 92.3% mAP50 versus 87.2% for a Faster R-CNN reference model on the same data split (detection baselines differ in training epochs and pretraining; see Experiments); U-Net achieves 95.71% mIoU, surpassing DeepLabv3+ (93.2%) under matched segmentation training settings. The system is deployed as an interactive web application for legacy building survey and preliminary regulatory review when only two-dimensional documentation is available. Full article

Parkinson’s disease (PD) is a neurodegenerative disorder with an increasing incidence rate that significantly affects patients’ motor functions and quality of life. Involuntary upper limb tremors (ULTs) commonly manifest unilaterally, affecting either the left or right upper limb. Clinically, ULT frequencies can be categorized into three distinct classes: low-frequency (<4.0 Hz), mid-frequency (4.0–7.0 Hz), and high-frequency (>7.0 Hz) tremors. These tremor motions are characterized by oscillatory or rotational (angular displacement) movements, commonly referred to as the micro-Doppler effect (mDE). This study aims to develop a short-range (<1.0 m) and contactless sensing method for ULT detection based on Doppler millimeter-wave (mm-Wave) radar. The reflected electromagnetic waves indicate time-varying frequency characteristics, which can be analyzed by using time–frequency transform (TFT) methods, such as the Wigner–Ville distribution (WVD) and smoothed pseudo WVD (SPWVD). These TFT methods are employed to extract mDE features, which are subsequently visualized as color-coded spectrograms for ULT classification. Then, a two-dimensional (2D) convolutional neural network (CNN) is employed to automatically recognize the visual feature patterns for ULTs classification based on frequency and amplitude information. In the experimental setup, the W-band (76–81 GHz) Doppler mm-Wave biosensor is implemented for sensing and extracting feature patterns. The proposed classifiers based on “WVD + 2D CNN” and “SPWVD + 2D CNN” are trained and validated by using the collected datasets, with 60% randomly selected for training datasets and 40% for testing datasets in each fold validation. A 10-fold cross-validation method is applied to evaluate the classifier’s performances, achieving an average precision of 95.92 ± 0.60%, average recall of 95.89 ± 0.62%, average F1-score of 0.9588 ± 0.0060, and average accuracy of 95.89 ± 0.62%, respectively. The experimental results demonstrate the feasibility of the proposed classifier for real-time ULTs classification in PD patients using short-range (<1.0 m) and contactless sensing. Full article

Sustainable finance has evolved rapidly in tandem with digital transformation and the circular economy; however, the simultaneous integration of these three domains remains fragmented. This study analyzes the intellectual structure of the field through a bibliometric analysis of a curated corpus of 2537 articles indexed in Scopus between 2015 and 2025, of which 2471 were classified into three thematic trajectories: sustainable finance combined with digital transformation (D1), sustainable finance combined with the circular economy (D2), and triadic convergence (D3). The classification followed a deductive, rule-based procedure, with documents independently coded by the two authors and discrepancies resolved by consensus. VOSviewer was used to construct networks of keyword co-occurrence, co-citation, and bibliographic coupling, identifying four thematic clusters. A complementary keyword-overlap projection was then used to articulate the deductive classification with the inductive clusters. The results reveal a rapidly expanding field, geographically concentrated in China, in which the dyadic trajectories anchor predominantly in a single conceptual cluster, while triadic convergence (D3), which appears only in 2021 and accounts for 2.7% of the classified corpus, is the only trajectory whose documents distribute across three clusters simultaneously. This pattern provides empirical support for interpreting triadic convergence as an emerging frontier rather than a consolidated stream. The findings are interpreted under the lens of economicità, an Italian accounting concept that frames sustainability as a condition for the firm’s long-term economic equilibrium. Full article

This review aims to evaluate the potential of endometriosis models, especially patient-derived iPSC models, to gain deeper insights into the disease, thereby advancing our understanding and treatment of endometriosis. This comprehensive narrative review utilized a structured search of the PubMed, Scopus, and Web of Science databases, primarily covering literature published between January 2000 and May 2025. An expansive search strategy was employed to capture the full breadth of the field using keywords such as “endometriosis,” “induced pluripotent stem cells (iPSCs),” “patient-derived organoids,” “disease modeling,” and “epigenetics” without restrictive filtering, ensuring the integration of both foundational theories and emerging biotechnological advances. In total, over 170 peer-reviewed publications were analyzed, ranging from landmark genomic meta-analyses that have identified significant risk loci to state-of-the-art 3D-culture systems for modeling patient-specific endometrial disease. By synthesizing these diverse sources, the review bridges the gap between traditional anatomical classifications and modern molecular modeling to evaluate the potential of iPSC platforms for personalized medicine and therapeutic discovery. Endometriosis is a multifactorial gynecological condition that affects 176 million women worldwide and can significantly impair quality of life. It occurs when endometrium-like tissue grows outside the uterus, responsive to ovarian hormones, causing inflammation, pain, and discomfort, and leading to fibrotic tissue. World Health Organization estimates indicate that 6–10% of women suffer from this disorder, which can cause infertility and increase the risk of developing various types of cancer and autoimmune disorders. The use of patient-derived iPSC models serves to gain deeper insights into the disease by mimicking the endometrial tissue or lesions observed in affected individuals, thereby advancing our understanding and treatment of endometriosis. Full article