Search Results (80)

Mitotic figures in tumor tissues are an important criterion for diagnosing malignant lesions, and physicians often search for the presence of mitosis in whole slide imaging (WSI). However, prolonged visual inspection by doctors may increase the likelihood of human error. With the advancement of deep learning, AI-based automatic cytopathological diagnosis has been increasingly applied in clinical settings. Nevertheless, existing diagnostic models often suffer from high computational costs and suboptimal detection accuracy. More importantly, when assessing cellular abnormalities, doctors frequently compare target cells with their surrounding cells—an aspect that current models fail to capture due to their lack of intercellular information modeling, leading to the loss of critical medical insights. To address these limitations, we conducted an in-depth analysis of existing models and propose an Inter–Intra Hypergraph Neural Network (II-HGNN). Our model introduces a block-based feature extraction mechanism to efficiently capture deep representations. Additionally, we leverage hypergraph convolutional networks to process both intracellular and intercellular information, leading to more precise diagnostic outcomes. We evaluate our model on publicly available datasets under varying imaging conditions, and experimental results demonstrate that our approach consistently outperforms baseline models in terms of accuracy. Full article

The Palma Cathedral, a landmark of Mediterranean Gothic architecture, features some of the most structurally daring slender piers in European ecclesiastical design. This study examines the role of marés stone—a local marine calcarenite—in enabling such architectural feats despite its inherent fragility. A multi-technique, non-invasive diagnostic campaign was conducted, including visual inspection, portable microscopy, and infrared thermography, to evaluate the physical condition and behavior of the stone under structural and environmental stress. The results reveal widespread deterioration processes—granular disintegration, alveolization, biological colonization, and structural cracking—exacerbated by the stone’s high porosity and exposure to marine aerosols and thermal fluctuations. Thermographic analysis highlighted moisture retention zones and hidden material discontinuities, while crack monitoring confirmed long-standing, localized structural strain. These findings demonstrate that the Cathedral’s formal audacity was grounded in a refined empirical understanding of marés’ properties. The study underscores the importance of material-based diagnostics for the sustainable conservation of Gothic heritage architecture. Full article

Subtle locomotor asymmetries are common in horses and may go unnoticed during routine pre-race clinical inspections, particularly when based solely on subjective evaluation. This study aimed to describe vertical head and pelvic movement asymmetries in racehorses that passed official pre-race inspections at a traditional racing event. Twenty-four horses were analysed using a markerless AI-based gait analysis system while trotting in-hand and during lungeing in both directions. Asymmetry parameters (HDmin, HDmax, PDmin, and PDmax) were extracted from video recordings, with values ≥0.5 considered clinically relevant. Vertical asymmetries were detected in 71% of horses during straight-line evaluation and in 79% during at least one lungeing direction. Some horses showed relevant asymmetries only under specific movement conditions, underscoring the complementary role of straight-line and lungeing assessments in comprehensive gait evaluation. These results suggest that objective gait analysis could enhance pre-race veterinary assessments, especially in traditional racing, where horses are subjected to significant biomechanical stress, including variable surface properties and repetitive directional loading. In such complex and dynamic environments, relying solely on visual assessment may result in the underdiagnosis of subtle locomotor alterations. The AI-based tools offer potential to improve the detection of subtle irregularities and support evidence-based decisions in performance horse management. Further investigations are warranted to validate the clinical relevance of currently adopted asymmetry thresholds, refine their diagnostic value, and support their integration into standardized pre-race evaluation protocols. Full article

Rising raw material costs and complex global supply chains have reduced the durability and availability of conveyor belts. In response, condition-based maintenance (CBM) with in situ diagnostics has become essential. This case study from a Polish lignite mine shows how subjective visual inspections were replaced with objective, repeatable measurements of belt core condition and thickness. Shifting refurbishment decisions from the plant to the conveyor improved success rates from 70% to over 90% and optimized belt lifecycle management. Sensor-based monitoring enables predictive maintenance, reduces premature or delayed replacements, increases belt reuse, lowers costs, and supports the circular economy by extending belt core life and reducing raw material demand. The study demonstrates how real-time, sensor-based diagnostics using inductive and ultrasonic technologies supports predictive maintenance of conveyor belts, improving refurbishment efficiency and lifecycle management. Full article

Skin diseases represent a major worldwide health hazard affecting millions of people yearly and substantially compromising healthcare systems. Particularly in areas where dermatologists are scarce, standard diagnostic techniques, which mostly rely on visual inspection and clinical experience, are frequently subjective, time-consuming, and prone to mistakes. This investigation undertakes a comparative analysis of four state-of-the-art deep learning architectures, YOLO11, YOLOv8, VGG16, and ResNet50, in the context of skin disease identification. This study evaluates the performance of these models using pivotal metrics, building upon the foundation of the YOLO paradigm, which revolutionized spatial attention and multi-scale representation. A properly selected collection of 900 high-quality dermatological images with nine disease categories was used for investigation. Robustness and generalizability were guaranteed by using data augmentation and hyperparameter adjustment. By varying benchmark models in balancing accuracy and recall while limiting false positives and false negatives, YOLO11 obtained a test accuracy of 80.72%, precision of 88.7%, recall of 86.7%, and an F1 score of 87.0%. The expedition performance of YOLO11 signifies a promising trajectory in the development of highly accurate skin disease detection models. Our analysis not only highlights the strengths and weaknesses of the model but also underscores the rapid development of deep learning techniques in medical imaging. Full article

Patients with chronic obstructive pulmonary disease (COPD) often experience acute exacerbations characterized by elevated neutrophilic inflammation in the lungs. Currently, this condition is diagnosed through visual inspection of sputum color and volume, a method prone to personal bias and unsuitable for patients who are unable to expectorate spontaneously. In this manuscript, we present a novel approach for measuring and monitoring exhaled myeloperoxidase (MPO), a biomarker of neutrophilic airway inflammation, without the need for sputum analysis. The method involves analyzing an unmodified surgical facemask worn by the patient for 30 min using biosensing decals that transfer antibody-coated nanoparticles. These colloids specifically interact with MPO trapped by the facemask in a dose-dependent manner, enabling the quantification of MPO levels, with a dynamic range up to 3 · 10¹ µg·mL⁻¹. The proposed diagnostic approach successfully differentiated patients with acute exacerbations from stable patients with 100% sensitivity and specificity. Healthy individuals also showed significantly lower MPO levels compared to COPD patients. Our results suggest that facemask analysis could be a non-invasive diagnostic tool for airway diseases, particularly in patients unable to expectorate. Full article

As an important economic crop, tomato is highly susceptible to diseases that, if not promptly managed, can severely impact yield and quality, leading to significant economic losses. Traditional diagnostic methods rely on expert visual inspection, which is not only laborious but also prone to subjective bias. In recent years, object detection algorithms have gained widespread application in tomato disease detection due to their efficiency and accuracy, providing reliable technical support for crop disease identification. In this paper, we propose an improved tomato leaf disease detection method based on the YOLOv10n algorithm, named BED-YOLO. We constructed an image dataset containing four common tomato diseases (early blight, late blight, leaf mold, and septoria leaf spot), with 65% of the images sourced from field collections in natural environments, and the remainder obtained from the publicly available PlantVillage dataset. All images were annotated with bounding boxes, and the class distribution was relatively balanced to ensure the stability of training and the fairness of evaluation. First, we introduced a Deformable Convolutional Network (DCN) to replace the conventional convolution in the YOLOv10n backbone network, enhancing the model’s adaptability to overlapping leaves, occlusions, and blurred lesion edges. Second, we incorporated a Bidirectional Feature Pyramid Network (BiFPN) on top of the FPN + PAN structure to optimize feature fusion and improve the extraction of small disease regions, thereby enhancing the detection accuracy for small lesion targets. Lastly, the Efficient Multi-Scale Attention (EMA) mechanism was integrated into the C2f module to enhance feature fusion, effectively focusing on disease regions while reducing background noise and ensuring the integrity of disease features in multi-scale fusion. The experimental results demonstrated that the improved BED-YOLO model achieved significant performance improvements compared to the original model. Precision increased from 85.1% to 87.2%, recall from 86.3% to 89.1%, and mean average precision (mAP) from 87.4% to 91.3%. Therefore, the improved BED-YOLO model demonstrated significant enhancements in detection accuracy, recall ability, and overall robustness. Notably, it exhibited stronger practical applicability, particularly in image testing under natural field conditions, making it highly suitable for intelligent disease monitoring tasks in large-scale agricultural scenarios. Full article

Early diagnosis of skin cancer can significantly improve patient survival. Currently, skin lesion classification faces challenges such as lesion–background semantic entanglement, high intra-class variability, artifactual interference, and more, while existing classification models lack modeling of physicians’ diagnostic paradigms. To this end, we propose DermViT, a medically driven deep learning architecture that addresses the above issues through a medically-inspired modular design. DermViT consists of three main modules: (1) Dermoscopic Context Pyramid (DCP), which mimics the multi-scale observation process of pathological diagnosis to adapt to the high intraclass variability of lesions such as melanoma, then extract stable and consistent data at different scales; (2) Dermoscopic Hierarchical Attention (DHA), which can reduce computational complexity while realizing intelligent focusing on lesion areas through a coarse screening–fine inspection mechanism; (3). Dermoscopic Feature Gate (DFG), which simulates the observation–verification operation of doctors through a convolutional gating mechanism and effectively suppresses semantic leakage of artifact regions. Our experimental results show that DermViT significantly outperforms existing methods in terms of classification accuracy (86.12%, a 7.8% improvement over ViT-Base) and number of parameters (40% less than ViT-Base) on the ISIC2018 and ISIC2019 datasets. Our visualization results further validate DermViT’s ability to locate lesions under interference conditions. By introducing a modular design that mimics a physician’s observation mode, DermViT achieves more logical feature extraction and decision-making processes for medical diagnosis, providing an efficient and reliable solution for dermoscopic image analysis. Full article

Smallholder potato farmers in Ethiopia do not realize the theoretical yield potential of the crop because they do not benefit from the advantages of using quality seed potato of improved varieties. The high disease incidence in seed potatoes has large implications on the potato farming system since the country lacks appropriate seed quality assurance mechanisms. Seed potato quality assurance relies more on the technical support provided by the national research and extension systems than the official seed certification agency. This paper elaborates systematic challenges and opportunities within the potato seed system and poses two research questions: (1) What type of seed quality assurance mechanisms (informal, quality declared, certified) are under implementation in Ethiopia? (2) How does the current seed quality assurance system operate in terms of reliability, accessibility, and quality standards to deliver quality seed potato? The data were collected through face-to-face in-depth key informant interviews with various seed regulatory laboratory managers and technicians in the Oromia, SNNP, and SWEP regions in the main seed- and ware-producing areas of Ethiopia. This was complemented by a comprehensive analysis of relevant documents. The findings show that currently there is no established procedure in place to officially certify early-generation seed potatoes. Two out of six seed quality control laboratories assessed for this study inspected seed potato fields in 2021 but as quality declared seed (QDS), and approved the fields inspected based on visual inspection alone. Our study revealed a weak linkage between early-generation seed (EGS) potato producers, commercial, and QDS seed potato producers, and seed quality control laboratories. Seed potato quality assurance operations were carried out by only a few seed regulatory laboratories with several concerns raised over the effectiveness of quality standards since seed-borne diseases, such as bacterial wilt, have been found at high frequency in the country’s seed potato system. Hence, the current procedures and challenges call for the necessity of upgrading current quality assurance in seed potato certification. Our study underlines the need for policymakers, development partners, and researchers to collaborate and pool efforts to consider transforming the quality declared system to appropriate seed certification. We recommended that institutionalizing novel plant disease diagnostics into seed regulatory frameworks is needed for sustainable potato production and food security in Ethiopia. Full article

This systematic review examines the capabilities, challenges, and practical implementations of the most widely utilized and emerging sensing technologies in structural health monitoring (SHM) for infrastructures, addressing a critical research gap. While many existing reviews focus on individual methods, comprehensive cross-method comparisons have been limited due to the highly tailored nature of each technology. We address this by proposing a novel framework comprising five specific evaluation criteria—deployment suitability in SHM, hardware prerequisites, characteristics of the acquired signals, sensitivity metrics, and integration with Digital Twin environments—refined with subcriteria to ensure transparent and meaningful performance assessments. Applying this framework, we analyze both the advantages and constraints of established sensing technologies, including infrared thermography, electrochemical sensing, strain measurement, ultrasonic testing, visual inspection, vibration analysis, and acoustic emission. Our findings highlight critical trade-offs in scalability, environmental sensitivity, and diagnostic accuracy. Recognizing these challenges, we explore next-generation advancements such as self-sensing structures, unmanned aerial vehicle deployment, IoT-enabled data fusion, and enhanced Digital Twin simulations. These innovations aim to overcome existing limitations by enhancing real-time monitoring, data management, and remote accessibility. This review provides actionable insights for researchers and practitioners while identifying future research opportunities to advance scalable and adaptive SHM solutions for large-scale infrastructure. Full article

An estimated 60,000 people die annually from skin cancer, predominantly melanoma. The diagnosis of skin lesions primarily relies on visual inspection, but around half of lesions pose diagnostic challenges, often necessitating a biopsy. Non-invasive detection methods like Computer-Aided Diagnosis (CAD) using Deep Learning (DL) are becoming more prominent. This study focuses on the use of multispectral (MS) imaging to improve skin lesion classification of DL models. We trained two convolutional neural networks (CNNs)—a simple CNN with six two-dimensional (2D) convolutional layers and a custom VGG-16 model with three-dimensional (3D) convolutional layers—using a dataset of MS images. The dataset included spectral cubes from 327 nevi, 112 melanomas, and 70 basal cell carcinomas (BCCs). We compared the performance of the CNNs trained with full spectral cubes versus using only three spectral bands closest to RGB wavelengths. The custom VGG-16 model achieved a classification accuracy of 71% with full spectral cubes and 45% with RGB-simulated images. The simple CNN achieved an accuracy of 83% with full spectral cubes and 36% with RGB-simulated images, demonstrating the added value of spectral information. These results confirm that MS imaging provides complementary information beyond traditional RGB images, contributing to improved classification performance. Although the dataset size remains a limitation, the findings indicate that MS imaging has significant potential for enhancing skin lesion diagnosis, paving the way for further advancements as larger datasets become available. Full article

The assessment and rehabilitation of upper-limb functionality are crucial for addressing motor disorders in individuals with multiple sclerosis (PwMS). Traditional methods often lack the sensitivity to quantify subtle motor impairments, with cerebellar tremor diagnosis typically based on subjective visual inspections by clinicians. This study explored the feasibility of using Microsoft HoloLens2 for motion capture to assess upper-limb function in PwMS. Using the ROCKapp application, kinematic metrics such as movement quality and oculomotor coordination were recorded during pick-and-place tasks. Data from twelve healthy individuals served as benchmarks, while nine PwMS, including three with cerebellar tremor and one with ataxia, were tested to evaluate the tool’s diagnostic potential. Clustering algorithms applied to the kinematic data classified participants into distinct groups, showing that PwMS without cerebellar symptoms sometimes displayed behavior similar to healthy controls. However, those with cerebellar conditions, like tremor and ataxia, were more easily differentiated. While the HoloLens2 shows promise in detecting motor impairments, further refinement is required to improve sensitivity for those without overt cerebellar symptoms. Despite these challenges, this approach offers potential for personalized rehabilitation, providing detailed feedback that could improve interventions and enhance quality of life for PwMS. In conclusion, these findings highlight the potential of mixed-reality tools to refine diagnostic accuracy, suggesting future studies to validate their integration in clinical rehabilitation programs. Full article

Background: The objective in epithelial ovarian cancer is to reach maximal cytoreduction with no visible residual tumor. Tumor detection during cytoreductive surgery depends on visual inspection, palpation, or blind biopsy, methods that lack reliability for identifying microscopic disease. Although the importance of microscopic disease in epithelial ovarian cancer is controversial, it may harbor chemoresistant cells and explain the high recurrence rates. Fluorescence-guided surgery (FGS) is an emerging approach. However, the potential in ovarian cancer remains underexplored; the majority of the existing evidence pertains to gastrointestinal tumors and a limited group of ovarian cancer patients. Their comparative effectiveness is still uncertain. Objective: To systematically review and evaluate the role of fluorescence-guided surgical techniques in detecting microscopic disease in ovarian cancer and compare their efficacy to total peritonectomy. Data Sources: A systematic search was made in three databases (PubMed, Web of Science, and Embase). The search was conducted from 1975 to 2024, including randomized controlled trials, observational studies, and conference abstracts in the last 25 years. Study Selection: Clinical studies published in English involving ovarian cancer patients undergoing FGS or total peritonectomy were included. Case reports, reviews, animal studies, and studies involving mixed cancer populations without ovarian cancer-specific data were excluded. Two independent reviewers screened 631 studies, yielding 12 eligible studies for final analysis. Data Extraction and Synthesis: Data were extracted and synthesized in accordance with PRISMA and MOOSE guidelines, using random-effects models for independent analysis. Sensitivity, specificity, positive predictive value (PPV), and odds ratios (ORs) were grouped, accompanied by subgroup analyses based on the fluorescence agent employed. For quality assessment, we utilized the NIH quality tool. Main Outcome(s) and Measure(s): The primary outcome was the rate of change in surgical management due to fluorescence guidance or total peritonectomy. Secondary outcomes comprised lesion-level sensitivity, specificity, and PPV. Safety outcomes included adverse events associated with fluorescence agents. Results: There were 12 studies involving 429 ovarian cancer patients. FGS improved the detection of microscopic disease compared to standard visualization methods, with a pooled sensitivity of 0.77. Folate receptor-targeted agents had high sensitivity (84%) but low specificity (26%). Aminolevulinic acid (5-ALA) showed superior diagnostic accuracy with a sensitivity of 84% and a specificity of 96%. Total peritonectomy showed no significant advantage over FGS for detecting microscopic disease. The adverse events were mild, with no serious events reported. We observed a high heterogeneity across studies and methodologies. Conclusions and Relevance: Fluorescence-guided surgery utilizing fluorescence tracers demonstrates potential in improving the detection of microscopic disease and may change surgical management in epithelial ovarian cancer, particularly with 5-ALA. Variability in performance and limited data on survival outcomes necessitates additional research. Total peritonectomy does not offer further advantage in the detection of microscopic disease. Future trials should focus on standardizing methodology and evaluating the effects of microscopic disease removal on survival outcomes. Registration: The study was registered to PROSPERO as CRD42024578274. Full article

Objectives: Computer-aided schizophrenia (SZ) detection methods mainly depend on electroencephalogram and brain magnetic resonance images, which both capture physical signals from patients’ brains. These inspection techniques take too much time and affect patients’ compliance and cooperation, while difficult for clinicians to comprehend the principle of detection decisions. This study proposes a novel method using face diagnosis images based on traditional Chinese medicine principles, providing a non-invasive, efficient, and interpretable alternative for SZ detection. Methods: An innovative face diagnosis image analysis method for SZ detection, which learns feature representations based on Vision Transformer (ViT) directly from face diagnosis images. It provides a face features distribution visualization and quantitative importance of each facial region and is proposed to supplement interpretation and to increase efficiency in SZ detection while keeping a high detection accuracy. Results: A benchmarking platform comprising 921 face diagnostic images, 6 benchmark methods, and 4 evaluation metrics was established. The experimental results demonstrate that our method significantly improves SZ detection performance with a 3–10% increase in accuracy scores. Additionally, it is found that facial regions rank in descending order according to importance in SZ detection as eyes, mouth, forehead, cheeks, and nose, which is exactly consistent with the clinical traditional Chinese medicine experience. Conclusions: Our method fully leverages semantic feature representations of first-introduced face diagnosis images in SZ, offering strong interpretability and visualization capabilities. It not only opens a new path for SZ detection but also brings new tools and concepts to the research and application in the field of mental illness. Full article

Background/Objectives: Paediatric PET/CT imaging is crucial in oncology but poses significant radiation risks due to children’s higher radiosensitivity and longer post-exposure life expectancy. This study aims to minimize radiation exposure by generating synthetic CT (sCT) images from emission PET data, eliminating the need for attenuation correction (AC) CT scans in paediatric patients. Methods: We utilized a cohort of 128 paediatric patients, resulting in 195 paired PET and CT images. Data were acquired using Siemens Biograph Vision 600 and Long Axial Field-of-View (LAFOV) Siemens Vision Quadra PET/CT scanners. A 3D parameter transferred conditional GAN (PT-cGAN) architecture, pre-trained on adult data, was adapted and trained on the paediatric cohort. The model’s performance was evaluated qualitatively by a nuclear medicine specialist and quantitatively by comparing sCT-derived PET (sPET) with standard PET images. Results: The model demonstrated high qualitative and quantitative performance. Visual inspection showed no significant (19/23) or minor clinically insignificant (4/23) differences in image quality between PET and sPET. Quantitative analysis revealed a mean SUV relative difference of −2.6 ± 5.8% across organs, with a high agreement in lesion overlap (Dice coefficient of 0.92 ± 0.08). The model also performed robustly in low-count settings, maintaining performance with reduced acquisition times. Conclusions: The proposed method effectively reduces radiation exposure in paediatric PET/CT imaging by eliminating the need for AC CT scans. It maintains high diagnostic accuracy and minimises motion-induced artifacts, making it a valuable alternative for clinical application. Further testing in clinical settings is warranted to confirm these findings and enhance patient safety. Full article