Search Results (65)

Background/Objectives: Predicting individual response to depression therapy prior to treatment initiation remains a critical clinical challenge, as the response rate to both selective serotonin reuptake inhibitors (SSRIs) and repetitive transcranial magnetic stimulation (rTMS) is approximately 50%, leaving treatment selection largely trial-based. This study presents a computer-aided decision (CAD) framework that predicts depression therapy outcomes from pre-treatment electroencephalogram (EEG) signals using advanced time-frequency representations and pretrained convolutional neural networks (CNNs). Methods: EEG signals from 30 SSRI patients and 46 rTMS patients are transformed into time-frequency images using Continuous Wavelet Transform (CWT), Variational Mode Decomposition (VMD), and their pixel-level fusion. Four pretrained CNN architectures, including ResNet-18, MobileNet-V3, EfficientNet-B0, and TinyViT-Hybrid, are fine-tuned and evaluated under both image-independent and subject-independent 6-fold cross-validation (CV). Results: Results reveal a clear therapy-specific pattern: CWT-based representations yield superior discrimination for SSRI outcome prediction, with ResNet-18 achieving 99.43% image-level accuracy, while VMD-based representations are statistically superior for rTMS outcome prediction, with ResNet-18 reaching 98.77%. Pixel-level fusion of CWT and VMD does not consistently improve performance over the best individual representation in either therapy context. Pairwise Wilcoxon signed-rank tests confirm a two-tier architectural hierarchy in which ResNet-18 and TinyViT-Hybrid significantly outperform MobileNet-V3 and EfficientNet-B0 across all conditions, while remaining statistically indistinguishable from each other. At the subject level, the framework achieves 82.50% and 83.53% accuracy for SSRI and rTMS, respectively, under strict subject-independent evaluation. Per-channel analysis reveals occipital dominance for SSRI under CWT and frontotemporal dominance for rTMS under VMD, consistent with known neurophysiological mechanisms. Conclusions: These findings demonstrate that the choice of time-frequency representation is therapy-specific and at least as important as architectural complexity, and that competitive performance can be achieved without recurrent or attention layers by combining well-designed spectral images with a simple pretrained residual network. Full article

This study tackles the challenge of data scarcity in medical AI, focusing on Non-Small-Cell Lung Cancer (NSCLC) diagnosis from Positron Emission Tomography (PET) and Computed Tomography (CT) images. We introduce MedScanGAN, a conditional Generative Adversarial Network designed to generate high-fidelity synthetic PET and CT images of Solitary Pulmonary Nodules (SPNs) to enhance computer-aided diagnosis systems. The framework incorporates advanced architectural features, including residual blocks, spectral normalization, and stabilized training strategies. MedScanGAN produces realistic images—particularly for PET representations—capable of plausibly misleading medical professionals. More importantly, when used to augment training datasets for established deep learning models such as YOLOv8, VGG-16, ResNet, and MobileNet, the synthetic data significantly improves NSCLC classification performance. Accuracy gains of up to +5.8 absolute percentage points were observed, with YOLOv8 achieving the best results at 94.14% accuracy, 93.12% specificity, and 95.33% sensitivity using the augmented dataset. The conditional generation mechanism enables the targeted synthesis of underrepresented classes, effectively addressing class imbalance. Overall, this work demonstrates both state-of-the-art medical image synthesis and its practical value in improving real-world diagnostic systems, bridging generative AI research and clinical pulmonary oncology. Full article

The early and accurate identification of breast cancer is a significant healthcare issue, largely because the traditional machine learning approaches rely on handcrafted features that are unable to fully capture the spatial and textural complexity found in mammograms. Even with the advancements made possible through deep learning and improvements in diagnostic performance, most computational-aided diagnosis (CAD) systems based on Convolutional Neural Networks (CNNs) still only rely on single-domain features, normally spatial features, while neglecting some important spectral and spatial–spectral features, leading to limitations in generalisability, redundancy, and loss of performative interpretability. Inspired by these limitations, this research proposes MERGE, a novel CAD framework that combines spatial, spectral, and spatial–spectral information—all part of a single multistage architecture taking advantage of three fine-tuned CNN models (ResNet-50, Xception, and Inception). This system utilises Discrete Stationary Wavelet Transform (DSWT) to enhance spectral–spatial features; Discrete Cosine Transform (DCT) to fuse the features optimally, resulting in enhanced spatial and spatial–spectral representations; and, finally, Non-Negative Matrix Factorisation (NNMF) for reduced-dimensional features. Finally, the Linear Discriminant Analysis (LDA), support vector machine (SVM), and k-nearest neighbours (KNN) classifiers provide a robust diagnosis. Using the INBreast and MIAS datasets in evaluations of the experimental research design, evaluation metrics of accuracy, sensitivity, specificity, and AUC were around 99%, with performance surpassing state-of-the-art paradigms. The findings of the suggested MERGE indicate significant promise as a dependable and effective diagnostic tool, enhancing the consistency and interpretability of breast cancer screening results. Full article

Thyroid nodule evaluation relies heavily on ultrasound imaging, yet it suffers from significant inter-operator variability. To address this, we present a preliminary validation of the Synergy-Net platform, an AI-driven Computer-Aided Diagnosis (CAD) system designed to standardize acquisition and improve diagnostic accuracy. The system integrates a U-Net architecture for anatomical segmentation and a ResNet-50 classifier for lesion characterization within a Human-in-the-Loop (HITL) workflow. The study enrolled 110 patients (71 benign, 39 malignant) undergoing surgery. Performance was evaluated against histopathological ground truth. The system achieved an Accuracy of 90.35% (95% CI: 88.2–92.5%), Sensitivity of 90.64% (95% CI: 87.9–93.4%), and an AUC of 0.90. Furthermore, the framework introduces a multimodal approach, performing late fusion of imaging features with genomic profiles (TruSight One panel). While current results validate the 2D diagnostic pipeline, the discussion outlines the transition to the ANTHEM framework, incorporating future 3D volumetric analysis and digital pathology integration. These findings suggest that AI-assisted standardization can significantly enhance diagnostic precision, though multi-center validation remains necessary. Full article

Recent text to image systems can synthesize Islamic heritage elements with high visual fidelity, but their outputs rarely translate into fabricable geometry or integrate into interiors without substantial redrawing. We present an end-to-end workflow that links historically grounded precedent retrieval, controllable tileable generation, semantic segmentation and vectorization, and geometry-aware mapping into Computer-Aided Design (CAD) environments. Contributions include the following: (i) a license-audited dataset schema and a retrieval classifier for common Islamic motif families and architectural elements; (ii) precedent retrieval via a ResNet 50 and Vision Transformer (ViT) embedding pipeline; (iii) a Low-Rank Adaptation (LoRA) tuned diffusion model that generates tileable motifs with motif/region controls; (iv) a raster-to-vector pipeline that enforces curve closure and minimum feature widths for CNC/laser fabrication; and (v) a rubric and domain metrics (symmetry coherence, seam/tileability error, spline closure and junction valence, UV distortion, feature width compliance) that quantify “depth of integration” beyond surface texture. Quantitative metrics and blinded expert ratings compare the workflow against strong parametric baselines, while scripts translate images to fabrication-ready vectors/solids across walls, ceilings, partitions, floors, and furniture. Cultural safeguards cover calligraphy handling, regional balance audits, and provenance/credit. The workflow advances heritage-aware generative design by carrying imagery across the last mile into buildable detail and by providing practical checklists for adoption in interior architecture and conservation. Full article

Hybrid computer-aided design and computer-aided manufacturing (CAD/CAM) materials have gained prominence in restorative dentistry due to their advantageous mechanical and esthetic properties; however, their long-term performance may be adversely affected by acidic oral environments, such as those associated with gastroesophageal reflux disease (GERD). This in vitro study aimed to investigate the effects of simulated gastric acid exposure on the surface roughness, gloss, color stability, and microhardness of two hybrid CAD/CAM materials: Vita Enamic and Cerasmart. Standardized rectangular specimens (2 mm thickness) were prepared and polished using a clinically relevant intraoral protocol. Baseline measurements were obtained for surface roughness, gloss, color change (ΔE), and Vickers microhardness. All specimens were then immersed in hydrochloric acid (pH 1.2) for 24 h to simulate prolonged gastric acid exposure, after which the same properties were re-evaluated. Post-immersion analysis revealed significant increases in surface roughness and reductions in gloss and microhardness for both materials (p < 0.05), with Vita Enamic demonstrating greater susceptibility to degradation. Color changes remained below the clinically perceptible threshold, with no significant differences between materials. These findings highlight the potential vulnerability of hybrid CAD/CAM materials to acidic environments and underscore the importance of careful material selection in patients predisposed to acid-related challenges. Full article

This paper explores the integration of computing and multimodal technologies into personal reading practices to enhance student engagement and knowledge assimilation in higher education. In response to a documented decline in voluntary academic reading, we investigated how technology-enhanced reading environments can re-engage students through interactive and personalized experiences. Central to this research is SmartRead, a proposed multimodal eReading platform that incorporates gamification, adaptive content delivery, and real-time feedback mechanisms. Drawing on empirical data collected from students at a higher education institution, we examined how features such as progress tracking, motivational rewards, and interactive comprehension aids influence reading behavior, engagement levels, and information retention. Results indicate that such multimodal interventions can significantly improve learner outcomes and user satisfaction. This paper contributes actionable insights into the design of innovative, accessible, and pedagogically sound digital reading tools and proposes a framework for future eReading technologies that align with multimodal interaction principles. Full article

Due to the rapid advancements in coordinate measuring systems, data processing software, and additive manufacturing (AM) techniques, it has become possible to create copies of existing models through the reverse engineering (RE) process. However, the lack of precise estimates regarding the accuracy of the RE process—particularly at the measurement, reconstruction, and computer-aided design (CAD) modeling stages—poses significant challenges. Additionally, the assessment of dimensional and geometrical errors during the manufacturing stage using AM techniques limits the practical implementation of product replicas in the industry. This paper provides an estimation of the errors encountered in the RE process and the AM stage of various models. It includes examples of an electrical box, a lampshade for a standing lamp, a cover for a vacuum unit, and a battery cover. The geometry of these models was measured using a GOM Scan 1 (Carl Zeiss AG, Jena, Germany). Following the measurement process, data processing was performed, along with CAD modeling, which involved primitive detection, profile extraction, and auto-surface methods using Siemens NX 2406 software (Siemens Digital Industries, Plano, TX, USA). The models were produced using a Fortus 360-mc 3D printer (Stratasys, Eden Prairie, MN, USA) with ABS-M30 material. After fabrication, the models were scanned using a GOM Scan 1 scanner to identify any manufacturing errors. The research findings indicated that overall, 95% of the points representing reconstruction errors are within the maximum deviation range of ±0.6 mm to ±1 mm. The highest errors in CAD modeling were attributed to the auto-surfacing method, overall, 95% of the points are within the average range of ±0.9 mm. In contrast, the lowest errors occurred with the detect primitives method, averaging ±0.6 mm. Overall, 95% of the points representing the surface of a model made using the additive manufacturing technology fall within the deviation range ±0.2 mm on average. The findings provide crucial insights for designers utilizing RE and AM techniques in creating functional model replicas. Full article

Warpage is among the most prevalent defects affecting injection molded parts. In this study, we aimed to develop methods to minimize warpage through mold design. Common strategies include matching the cavity geometry to the intended shape of the part, adjusting cavity dimensions to offset material shrinkage, and optimizing the cooling system and critical injection molding parameters. These optimization methods can offer significant improvements, but recently introduced methods that optimize the molded part and mold cavity shape result in higher levels of warpage reduction. In these methods, optimization of the shape of the molded part is achieved by shaping it in the opposite direction of warpage—a method known as inverse contouring. Inverse contouring of molded parts is a design technique in which mold cavities are intentionally modified to incorporate compensatory geometric deviations in regions anticipated to exhibit significant warpage. The final result after molded part ejection and warpage is a significant reduction in deviations between the warped and reference molded part geometries. In this study, a two-step approach for minimizing warpage was used: the first step was optimizing the most significant injection molding parameters, and the second was inverse contouring. In the first step, Response Surface Methodology (RSM) and Autodesk Moldflow Insight 2023 simulations were used to optimize molded part warpage based on three processing parameters: melt temperature, target mold temperature, and coolant temperature. For improved accuracy, a Computer-Aided Design (CAD) model of the warped molded part was exported into ZEISS Inspect 2023 software and aligned with the reference CAD geometry of the molded part. The maximal warpage value after the initial simulation was 1.85 mm based on Autodesk Moldflow Insight simulations and 1.67 mm based on ZEISS Inspect alignment. After RSM optimization, the maximal warpage was 0.73 mm. In the second step, inverse contouring was performed on the molded part, utilizing the initial injection molding simulation results to further reduce warpage. In this step, the CAD model of the redesigned, inverse-contoured molded part was imported into Moldflow Insight to conduct a second iteration of the injection molding simulation. The simulation results were exported into ZEISS Inspect software for a final analysis and comparison with the reference CAD model. The warpage values after inverse contouring were reduced within the range of ±0.30 mm, which represents a significant decrease in warpage of approximately 82%. Both steps are presented in a case study on an injection molded part made of polybutylene terephthalate (PBT) with 30% glass fiber (GF). Full article

Background/Objectives: Diabetic foot ulcers (DFUs) remain a critical complication of diabetes, with high rates of amputation when not diagnosed early. Despite advancements in medical imaging, current DFU detection methods are often limited by their computational complexity, poor generalizability, and delayed diagnostic performance. This study presents a novel hybrid diagnostic framework that integrates traditional feature extraction methods with deep learning (DL) to improve the early real-time computer-aided detection (CAD) of DFUs. Methods: The proposed model leverages plantar thermograms to detect early thermal asymmetries associated with DFUs. It uniquely combines the oriented FAST and rotated BRIEF (ORB) algorithm with the Bag of Features (BOF) method to extract robust handcrafted features while also incorporating deep features from pretrained convolutional neural networks (ResNet50, AlexNet, and EfficientNet). These features were fused and input into a lightweight deep neural network (DNN) classifier designed for binary classification. Results: Our model demonstrated an accuracy of 98.51%, precision of 100%, sensitivity of 98.98%, and AUC of 1.00 in a publicly available plantar thermogram dataset (n = 1670 images). An ablation study confirmed the superiority of ORB + DL fusion over standalone approaches. Unlike previous DFU detection models that rely solely on either handcrafted or deep features, our study presents the first lightweight hybrid framework that integrates ORB-based descriptors with deep CNN representations (e.g., ResNet50 and EfficientNet). Compared with recent state-of-the-art models, such as DFU_VIRNet and DFU_QUTNet, our approach achieved a higher diagnostic performance (accuracy = 98.51%, AUC = 1.00) while maintaining real-time capability and a lower computational overhead, making it highly suitable for clinical deployment. Conclusions: This study proposes the first integration of ORB-based handcrafted features with deep neural representations for DFU detection from thermal images. The model delivers high accuracy, robustness to noise, and real-time capabilities, outperforming existing state-of-the-art approaches and demonstrating strong potential for clinical deployment. Full article

The early detection of Alzheimer’s disease (AD) is essential for improving patient outcomes, enabling timely intervention, and slowing disease progression. However, the complexity of neuroimaging data presents significant obstacles to accurate classification. This study introduces a computationally efficient AI framework designed to enhance AD staging using structural MRI. The proposed method integrates discrete wavelet transform (DWT) for multi-scale feature extraction, a novel reduced kernel partial least squares (Red-KPLS) algorithm for feature reduction, and ResNet-50 for classification. The proposed technique, referred to as Red-KPLS-CNN, refines MRI features into discriminative biomarkers while minimizing redundancy. As a result, the framework achieves 96.9% accuracy and an F1-score of 97.8% in the multiclass classification of AD cases using the Kaggle dataset. The dataset was strategically partitioned into 60% training, 20% validation, and 20% testing sets, preserving class balance throughout all splits. The integration of Red–KPLS enhances feature selection, reducing dimensionality without compromising diagnostic sensitivity. Compared to conventional models, our approach improves classification robustness and generalization, reinforcing its potential for scalable and interpretable AD diagnostics. These findings emphasize the importance of hybrid wavelet–kernel–deep learning architectures, offering a promising direction for advancing computer-aided diagnosis (CAD) in clinical applications. Full article

Reverse engineering (RE) is essential in recreating 3D models of existing manufactured parts. It is widely used for repairing damaged components, improving used parts, and designing new items based on older models. One of the most common methods in RE is photogrammetry, which enables 3D reconstruction by capturing multiple images. Therefore, this study aimed to explore the application of mobile photogrammetry to obtain a 3D model of a train brake pad. The process started with capturing images of objects in a quick and professional manner to ensure visualization of data. This was followed by processing 2D images using Agisoft Metashape 2.2.1 software and Artificial Intelligence (AI) algorithms to create a precise 3D model. Subsequently, assessment was performed using feasibility in terms of cost, time, and accuracy. The results show that mobile photogrammetry provided an accessible and cost-effective method, alongside maximum contact stress after reducing optimization by approximately 28.42%, with maximum error value measured by the virtual model with the reference value of 0.30 mm (on Metashape) and 0.46 mm (on AI). This suggested that reverse parameterization significantly accelerated computer-aided design (CAD) model reconstruction and reduced the part redesign development cycle. By using photogrammetry and parametric modeling, engineers could accurately analyze and optimize train brake pads, ensuring safety as well as sustainability in railway operations. Additionally, RE and parametric modeling could assist in creating durable, cost-effective alternatives, and predicting appropriate replacements. Full article

Additive manufacturing technologies have allowed the production of complex geometries that are typically obtained by applying topology optimization techniques. The outcome of the optimization process is a tessellated geometry, which has reduced aesthetic quality and unwanted spikes and cusps. Filters can be applied to improve the surface quality, but volume shrinking and geometry modification can be noticed. The design practice suggests manually re-designing the object in Computer-Aided Design (CAD) software, imitating the shape suggested by topology optimization. However, this operation is tedious and a lot of time is wasted. This paper proposes a methodology to automate the conversion from topology optimization output to a CAD-compatible design for industrial components. Topology optimization usually produces a dense triangle mesh with a high topological genus for those objects. We present a method to automatically generate a collection of spline (tensor-product) patches joined watertight and test the approach on real-wise industrial components. The methodology is based on the use of quadrilateral patches which are built on the external surface of the components. Based on the tests carried out, promising results have been obtained. It constitutes a first step towards the automatic generation of shapes that can readily be imported and edited in a CAD system. Full article

Background/Objectives: Chest disease identification for Tuberculosis and Pneumonia diseases presents diagnostic challenges due to overlapping radiographic features and the limited availability of expert radiologists, especially in developing countries. The present study aims to address these challenges by developing a Computer-Aided Diagnosis (CAD) system to provide consistent and objective analyses of chest X-ray images, thereby reducing potential human error. By leveraging the complementary strengths of convolutional neural networks (CNNs) and vision transformers (ViTs), we propose a hybrid model for the accurate detection of Tuberculosis and for distinguishing between Tuberculosis and Pneumonia. Methods: We designed a two-step hybrid model that integrates the ResNet-50 CNN with the ViT-b16 architecture. It uses the transfer learning on datasets from Guangzhou Women’s and Children’s Medical Center for Pneumonia cases and datasets from Qatar and Dhaka (Bangladesh) universities for Tuberculosis cases. CNNs capture hierarchical structures in images, while ViTs, with their self-attention mechanisms, excel at identifying relationships between features. Combining these approaches enhances the model’s performance on binary and multi-class classification tasks. Results: Our hybrid CNN-ViT model achieved a binary classification accuracy of 98.97% for Tuberculosis detection. For multi-class classification, distinguishing between Tuberculosis, viral Pneumonia, and bacterial Pneumonia, the model achieved an accuracy of 96.18%. These results underscore the model’s potential in improving diagnostic accuracy and reliability for chest disease classification based on X-ray images. Conclusions: The proposed hybrid CNN-ViT model demonstrates substantial potential in advancing the accuracy and robustness of CAD systems for chest disease diagnosis. By integrating CNN and ViT architectures, our approach enhances the diagnostic precision, which may help to alleviate the burden on healthcare systems in resource-limited settings and improve patient outcomes in chest disease diagnosis. Full article