Search Results (12,749)

Objective: To develop machine learning models by integrating transvaginal ultrasound (TVUS) with clinical indicators, conduct visual analysis of the models, and systematically assess their diagnostic efficacy in differentiating early cervical neoplastic lesions. Methods: A total of 144 eligible patients (84 cases of early cervical cancer and 60 cases of cervical myoma) admitted to the First Affiliated Hospital of Chongqing Medical University from January 2018 to August 2025 were retrospectively enrolled in this study. Their clinical data, human papillomavirus (HPV) test results, Thinprep Cytologic Test (TCT) findings, TVUS images and magnetic resonance (MR) imaging data were collected and subjected to comprehensive statistical analysis. Univariate and multivariate Logistic Regression analyses were performed to identify independent differentiating factors for lesion classification. Eleven machine learning models were subsequently constructed, and their diagnostic performance was evaluated using receiver operating characteristic (ROC) curves, decision curve analysis (DCA), and the DeLong test. Finally, a nomogram was developed based on the optimal-performing model for clinical visualization. Results: The TVUS–clinical indicator integration model identified five independent differentiating factors: HPV status, TCT findings, menopausal status, ultrasonic tumor blood supply, and ultrasonic tumor morphology. In contrast, the MR–clinical indicator integration model screened out three independent factors: HPV status, TCT findings, and intratumoral signal intensity on MR T2-weighted imaging (T2WI). The TVUS integration model demonstrated marginally superior diagnostic performance, with a sensitivity of 0.988, specificity of 0.983, and an area under the ROC curve (AUC) of 0.991, compared with the MR integration model (sensitivity: 0.952, specificity: 0.950, AUC: 0.975); however, this difference in AUC values was not statistically significant (p = 0.911). Among the 11 machine learning models, the Logistic Regression model exhibited optimal classification performance and stability. DCA curves confirmed that all constructed models outperformed single-index diagnostic strategies in clinical decision-making for lesion differentiation. A nomogram was further established based on the Logistic Regression model for intuitive clinical application. Conclusions: Multiple machine learning models integrating TVUS with clinical indicators are successfully developed, and a corresponding nomogram is constructed in this study. Full article

Background: Dyspnea is a common reason for emergency medical service (EMS) interventions and is associated with a substantial risk of severe clinical course, complications, and hospital admission. Its differential diagnosis in the prehospital setting remains challenging due to the limited availability of imaging modalities. Point-of-care ultrasound (POCUS), including lung ultrasound (LUS), is a rapid, field-applicable technique recommended in numerous acute respiratory diagnostic scenarios. Objective: To evaluate the use of lung ultrasound in the prehospital setting and its association with the precision of diagnoses related to respiratory failure, the frequency of transport to the emergency department (ED) among patients presenting with dyspnea/respiratory failure, and to characterize the profile of sonographic findings with their correlation to clinical diagnostic categories. Additionally, transport rates in the study population were compared with aggregated regional data for the Masovian Voivodeship (excluding the analyzed county). Methods: A retrospective observational study was conducted on EMS interventions performed between 01 January 2025 and 30 June 2025 in Legionowo County (N = 353). The analysis included ICD-10 codes assigned in prehospital documentation (one primary code and up to two additional codes) in patients presenting with dyspnea and/or respiratory failure, the performance of ultrasound examination, and resulting LUS findings (absence of pleural sliding and/or lung point; B-lines; consolidations; C-lines; pleural effusion). Descriptive analyses, frequency comparison tests (χ²/Fisher), estimation of relative risk (RR) with 95% confidence intervals (CI), and agreement analysis using Cohen’s kappa coefficient (κ) between etiological categories derived from ICD-10 codes and those inferred from LUS profiles were performed (κ with 95% CI estimated using bootstrap resampling). The study was reported in accordance with the STROBE guidelines for observational studies. Additionally, the distribution of ICD-10 coding and the proportion of hospital transports across the entire Masovian Voivodeship were compared with those observed in the analyzed area. Results: Ultrasound examination was performed in 72/353 (20.4%) EMS interventions; transport to the emergency department occurred in 239/353 (67.7%) cases. The most frequent clinical categories based on ICD-10 codes were: general/symptom-based 182/353 (51.6%), inflammatory 77/353 (21.8%), obstructive 66/353 (18.7%), and cardiological 20/353 (5.7%). Among abnormal LUS findings, the most common were B-lines (43/72; 61.4%) and consolidations (29/72; 41.4%). Consolidations were strongly associated with the inflammatory category (OR 9.72; p < 0.001), whereas B-lines were associated with the cardiological category (OR 23.41; p = 0.0011) among cases in which LUS was performed. Ultrasound use was associated with a higher frequency of assigning at least one targeted (non-symptom-based) diagnosis within ICD coding: 53/72 (73.6%) vs. 111/278 (39.9%), RR 1.84 (95% CI 1.51–2.25; p < 0.001). Agreement between the ICD-10 etiological category (inflammatory/cardiological/obstructive/other) and the category inferred from the LUS profile was moderate: κ = 0.36 (95% CI 0.21–0.51), with an observed agreement of 54.2%. Compared with aggregated regional data (Masovian Voivodeship excluding the analyzed county), the overall transport rate for comparable ICD-10 codes was lower in the study unit: 279/409 (68.2%) vs. 11,351/13,785 (82.3%), RR 0.83 (95% CI 0.78–0.89; p < 0.001). The largest differences were observed for dyspnea (R06.0: 72.9% vs. 88.2%; RR 0.83) and obstructive codes (J44/J45/J46 combined: 43.1% vs. 67.0%; RR 0.64). Conclusions: In this retrospective analysis, an EMS unit with systematically implemented ultrasound demonstrated a lower frequency of hospital transport for selected dyspnea/respiratory failure codes compared with regional data and greater precision in ICD-10 diagnostic coding in cases where ultrasound was performed. The profile of LUS findings correlated with clinical categories in a manner consistent with existing literature. Full article

Monolithic zirconia restorations are frequently affected by the unnoticed growth of subcritical cracks, a failure process that is not captured by traditional imaging methods like radiographs and ultrasounds in sophisticated dental architectures. To address this evaluative inadequacy, this research introduces a hierarchical deep learning framework for microcrack detection and spatial localization. We promote a hierarchical deep learning system that integrates Acoustic Emission (AE) detection alongside signal processing. Raw AE signals utilized during dynamic loading are enhanced via Kalman filtering and Continuous Wavelet Transform (CWT) to construct high-fidelity time–frequency scalograms. The diagnostic pipeline operates in two stages: first, a hybrid CNN–BiGRU network with temporal attention fulfills zirconia component-level classification; second, a ResNet-18 backbone integrated with Bidirectional LSTM and Multi-Head Attention precisely localizes defects across five anatomical crown regions. This hierarchical design effectively captures the non-stationary, transient nature of fracture-induced stress waves. The framework achieved an F1-score of 99.00% and an AUC of 0.994, significantly outperforming conventional convolutional networks. By enabling predictive maintenance through early, non-invasive damage localization, this study demonstrates a promising laboratory framework for AE-based crack detection in zirconia dental structures and prosthetics and toward enhanced clinical reliability in digital dentistry. Full article

Background: Low-dose CT scanning is a key tool in lung cancer screening, enabling the detection of clinically significant abnormalities in asymptomatic individuals and often prompting further diagnostic evaluation. Case Presentation: We describe the case of an 80-year-old man with a heavy smoking history who was found to have a new right middle lobe collapse on screening CT. Subsequent positron emission tomography-computed tomography (PET/CT) imaging demonstrated mild fluorodeoxyglucose (FDG) uptake (SUVmax 2.7), raising concern for a low-grade endobronchial malignancy versus mucoid impaction. Flexible fiberoptic bronchoscopy revealed a large exophytic endobronchial mass occluding the airway. Histopathologic examination of the biopsy sample unexpectedly revealed vegetable material, consistent with chronic foreign-body aspiration. Discussion: Unrecognized aspiration events are relatively common in elderly adults and can mimic malignancy on imaging. This case highlights an important diagnostic pitfall: inflammatory endobronchial processes, including foreign-body granulomas, can demonstrate FDG uptake and mimic malignancy. Conclusion: Clinicians should maintain a broad differential diagnosis when evaluating PET-avid endobronchial lesions, especially in elderly patients. Full article

Imaging-based multi-omics derived from digital histopathology provides a valuable approach for characterizing tumor heterogeneity from routine clinical specimens. However, robust multi-cancer histopathological analysis remains challenging due to pronounced intra-tumor variability, inter-organ morphological overlap, and sensitivity to staining and acquisition variations, which can limit the generalizability of deep learning models. These limitations are largely driven by insufficient representation learning, particularly in multi-organ and multi-class diagnostic settings. In this study, we propose a hierarchically regularized representation learning framework for multi-cancer histopathological image analysis that models imaging-based features across multiple organs and diagnostic categories. The framework integrates complementary mechanisms to capture fine-grained cellular morphology, long-range tissue architecture, and organ-aware diagnostic semantics within a unified computational model. A hierarchical supervision strategy guides the network to reduce entanglement between organ-level structural characteristics and disease-specific diagnostic patterns in the learned representations. The method operates without pixel-level annotations or handcrafted morphological priors, supporting scalable experimental evaluation. We demonstrate the approach on balanced lung and colon cancer histopathology cohorts, achieving 96.5% accuracy on lung cancer classification and 96.8% accuracy on colon cancer classification. Ablation and robustness analyses further validate the contributions of hierarchical regularization and consistency learning. Overall, this work provides a demonstrated proof-of-concept framework for representation-centric imaging-based analysis in multi-organ histopathology under the evaluated dataset conditions. Full article

Structured radiology reporting can improve clinical decision support by standardizing clinical findings into hierarchical formats. However, thousands of questions in structured report templates about clinical findings are prohibitively time-consuming, which can limit clinical adoption. Furthermore, early medical VQA datasets primarily focused on free-text and independent question–answer pairs while a recent dataset, Rad-ReStruct, introduced a hierarchical VQA, but the accompanying model still relies heavily on flattened embedding representations and single-path text–image fusion mechanisms that inadequately handle complex hierarchical dependencies in responses. In this paper, we propose DPA-HiVQA (Dual-Path Cross-Attention for Hierarchical VQA), addressing these limitations through two key contributions: (1) multi-scale image embedding representing global semantic embeddings with patch-level spatial features from domain-specific BioViL encoder; (2) dual-path cross-attention mechanism enabling simultaneous holistic semantic understanding and fine-grained spatial reasoning. Evaluated on the Rad-ReStruct benchmark, the model substantially outperforms the established benchmark baseline with an overall F1-score and Level 3 F1-score improvement by 21.2% and 31.9%, respectively. The proposed model demonstrates that dual-path cross-attention architectures can effectively connect holistic semantic understanding and fine-grained spatial detail, paving the way for practical AI-assisted structured reporting systems that reduce radiologist burden while maintaining diagnostic accuracy. Full article

Background: Pancreatic neuroendocrine neoplasms (PanNENs) represent a heterogeneous tumor entity with a steadily rising incidence, mainly due to advances in imaging and growing diagnostic awareness. In pancreatic ductal adenocarcinoma (PDAC), the mesopancreas (MP) has been identified as a frequent site of microscopic tumor spread and a key determinant of circumferential resection margin (CRM) status, leading to the concept of standardized mesopancreatic excision (MPE). While its oncological relevance in PDAC is increasingly recognized, the role of the mesopancreas in PanNENs remains unclear. This study aimed to systematically evaluate mesopancreatic infiltration in PanNENs and to identify associated clinicopathological predictors. Methods: Consecutive patients undergoing oncological pancreatoduodenectomy, spleen-preserving distal pancreatectomy, or distal splenopancreatectomy for PanNENs and PanNECs were included. The mesopancreas was histopathologically examined for tumor infiltration within CRM assessment. Results: MP infiltration was detected in 60% of patients. It was associated with higher Ki-67 index, larger tumor size, lymph node involvement, venous invasion, and positive CRM status. A Ki-67 index ≥ 5% and tumor size ≥ 21.5 mm were identified as predictors of MP infiltration. Higher T stage predicted reduced overall survival (OS), whereas MP infiltration, lymphatic (L1) and venous (V1) invasion, and Ki-67 ≥ 5% were associated with impaired disease-free survival (DFS). Conclusion: Mesopancreatic infiltration is frequently present in PanNENs and correlates with aggressive tumor characteristics. Given its association with CRM positivity and reduced DFS, consideration of the mesopancreas in staging and surgical strategies appears oncologically justified. Larger studies are required to validate these findings. Full article

Background: Clinical target volume (CTV) delineation in glioblastoma remains debated, particularly in the era of modern chemoradiation and image-guided radiotherapy. Whether reduced CTV margins can preserve oncological outcomes without increasing marginal or out-of-field failures remains uncertain. We evaluated the association of the gross tumor volume (GTV)-to-CTV margin with survival, patterns of failure, and its interaction with O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status. Materials and Methods: We retrospectively analyzed a single-center cohort of patients with glioblastoma treated with conventionally fractionated chemoradiation (58–60 Gy in 29–33 fractions). Patients were categorized into two predefined margin groups: <1.5 cm and 1.5 cm. The primary endpoint was overall survival (OS); secondary endpoints included progression-free survival (PFS) and patterns of failure. Survival was assessed using Kaplan–Meier estimates and Cox regression, including an interaction term with MGMT status. Results: Among 102 eligible patients, 95 were included in the margin-based OS analysis. Reduced margins (<1.5 cm; applied range 1.0–1.4 cm) were not associated with worse OS, either overall or within MGMT subgroups. No significant differences were observed in PFS or recurrence patterns, with overlapping distributions and no increase in marginal or out-of-field recurrences. MGMT methylation and gross total resection were independently associated with improved survival, while no statistically significant interaction between margin and MGMT status was detected. Conclusions: In this retrospective exploratory cohort, reduced GTV-to-CTV margins were not associated with a clear signal of worse survival or less favorable recurrence patterns. These findings are consistent with the oncological adequacy of margins around 15 mm and justify cautious prospective evaluation of whether further reduction can be achieved safely, including formal assessment of toxicity, neurocognitive outcomes, and quality of life. Full article

Background: This study aims to systematically review the current literature on the application of radiomic features and artificial intelligence (AI) in the diagnosis and prognosis of common peripheral nerve-related conditions, including carpal tunnel syndrome (CTS), chronic inflammatory demyelinating polyneuropathy (CIDP), polyneuropathy, organomegaly, endocrinopathy, monoclonal gammopathy and skin abnormalities (POEMS) syndrome, and in distinguishing between benign and malignant tumors. Methods: A comprehensive literature search was conducted in PubMed and Google Scholar for studies published between January 2019 and September 2025. Inclusion criteria comprised studies that used radiomics or AI-based radiomics approaches with diagnostic or prognostic purposes in peripheral nerve disorders. Results: A total of 40 studies were identified, of which 17 met the inclusion criteria. Among these, 9 studies employed magnetic resonance imaging (MRI), including one combined with PET/CT, while 8 used ultrasound (US). Most studies were retrospective and limited by small sample sizes, lack of external validation, and predominance of single-center designs. Conclusions: Since a seminal study published in 2019, there has been increasing evidence supporting the role of radiomics and AI in improving the diagnosis and prognosis of peripheral nerve disorders, particularly using MRI and US. However, significant challenges remain, including variability in imaging data, segmentation complexity, and limited availability of validated datasets. Future advancements in imaging technologies and multidisciplinary collaboration are essential to enhance clinical applicability. Full article

Background: Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most prevalent chronic liver disease worldwide and a major contributor to the global cardiometabolic burden. Early identification of patients at risk of metabolic dysfunction-associated steatohepatitis (MASH) and advanced fibrosis is essential to prevent liver-related and cardiovascular complications. In Spain, the burden of MASLD is increasing, yet information on routine clinical management by gastroenterologists remains limited. Methods: A nationwide cross-sectional online survey was conducted among members of the Spanish Society of Digestive Diseases (SEPD). The questionnaire explored five domains: MASLD knowledge, use of non-invasive biomarkers and imaging, awareness and implementation of clinical guidelines, cardiometabolic and alcohol-related risk assessment, and coordination with primary care. Results: A total of 429 specialists responded, 33.1% reported more than 20 years of practice and most worked in public hospitals, including 29.2% in large tertiary centers. Awareness of the MASLD definition was high, and 91.2% identified fibrosis as the main prognostic determinant. Non-invasive fibrosis biomarkers were widely used, whereas steatosis biomarkers were less frequently applied. Elastography was available to 96.1%. Guideline knowledge was reported by 80.4%, although implementation was lower. Cardiovascular risk evaluation varied: 75.1% reported systematic screening. Alcohol consumption was usually assessed. Coordination with primary care was limited: 91.1% expressed concerns regarding physicians’ familiarity with MASLD classification, and only 31.1% reported shared protocols. Conclusions: Spanish gastroenterologists show high awareness of MASLD and broad access to non-invasive diagnostic tools. However, important gaps remain in cardiovascular and alcohol risk assessment, guideline implementation, and coordination with primary care. Full article

The field of nanoparticle-based biotechnology has undergone substantial advancement, characterized by progress in targeted drug delivery systems, the development of innovative diagnostic and imaging platforms, the expanded adoption of environmentally sustainable (“green”) synthesis approaches, and an increasing emphasis on the integration of emerging technologies such as artificial intelligence and nanorobotics. Conventional nanoparticle synthesis often involves toxic reducing agents; however, recent advances promote eco-friendly green synthesis methods utilizing biological systems such as bacteria, fungi, algae, yeast, plants, and actinomycetes. These biological approaches are safe, sustainable, cost-effective, and capable of producing highly stable Nanoparticles (NPs). The interaction of nanomaterials with biological systems is crucial for developing intracellular and subcellular drug delivery technologies with minimal toxicity, governed by nano–bio interface mechanisms such as cellular translocation, surface wrapping, embedding, and internal attachment. Key factors influencing NP behavior include morphology, size, surface area, surface charge, and ligand chemistry. Magnetic nanoparticles, particularly iron-based forms, exhibit unique superparamagnetic properties that are strongly influenced by particle size, as explained by the Néel relaxation mechanism, in which thermal energy induces flipping of magnetic moments. Nanoparticles demonstrate diverse modes of action, including antimicrobial activity, reactive oxygen species (ROS)-induced cytotoxicity, genotoxicity, and plant growth promotion. NP performance and biological effects are strongly dependent on their size, shape, dosage, and concentration. This critical review article aims to elucidate evolution, classification, preparation methods, and multifaceted applications of nanoparticles Full article

Deep learning (DL) has transformed cancer imaging by enabling automated tumour detection, classification, and risk prediction. Despite impressive diagnostic performance, limited explainability and poor uncertainty calibration continue to restrict clinical integration. This review is guided by five research questions that examine the challenges, impact, and translational implications of explainable artificial intelligence (XAI) in oncology imaging. We identify key barriers to trust, including dataset bias, shortcut learning, opacity of convolutional neural networks, and workflow misalignment. Evidence suggests that explainable models can increase clinician confidence, reduce false positives, and improve collaborative decision-making when explanations are faithful, semantically meaningful, and uncertainty aware. We evaluate architectural strategies that embed interpretability such as concept-bottleneck models, prototype-based learning, and attention regularization along with post hoc techniques. Beyond performance metrics, we examine how interpretable AI aligns with clinical reasoning processes and analyse regulatory, ethical, and medico-legal considerations influencing deployment. The findings indicate that explainability alone is insufficient, durable trust requires epistemic alignment, prospective validation, lifecycle governance, and equity-focused evaluation. By reframing explainability as a structural design principle rather than a supplementary feature, this review outlines a pathway toward accountable and clinically dependable AI systems in oncology. Full article

Background: Magnetic Particle Imaging (MPI) is an emerging biomedical imaging modality that detects superparamagnetic iron oxide nanoparticles (SPIONs), providing high contrast, sensitivity, and quantification capabilities without ionizing radiation, making it particularly suitable for cancer diagnostics. Considerable engineering efforts are underway to translate MPI technology to clinical settings. Most of these MPI scanners feature a cylindrical bore geometry similar to that of other clinical imaging modalities, which limits their potential application primarily to head scanning. Methods: We have developed a single-sided MPI scanner designed to expand the modality’s applicability to other regions of the human body through a unique hardware design developed in our previous work. Imaging experiments were performed on an anatomical breast phantom containing implanted SPION point sources placed at anatomically plausible locations for breast tumors. These point sources served as artificial tumors for evaluating the system’s suitability for breast imaging applications. Results: The scanner successfully detected and clearly resolved the implanted SPION tumors in two orthogonal imaging planes. Tumor positioning was independently validated by ultrasound imaging, confirming MPI’s accurate localization. In addition, sensitivity measurements demonstrated a detection limit of 4.0 $\mathsf{\mu }\mathrm{g}$ of iron, below the estimated 4.8 $\mathsf{\mu }\mathrm{g}$ sensitivity threshold required for breast tumor detection with electronic depth scanning up to 3.5 cm deep. Conclusions: Together, these results demonstrate the capability of a single-sided MPI geometry for breast imaging applications. Imaging an anatomical breast-shaped volume presents significant challenges for MPI due to the size and accessibility constraints of conventional hardware. The results presented highlight the advantages of this approach and support its potential to extend MPI from small-animal imaging to clinically relevant applications. Full article

There are numerous benefits associated with creating digital copies of anatomical structures, which can be used during patient diagnosis. Such models can be used not only for visualization, but also in order to assess the condition of the patient. As advances in both medical imaging and 3D graphics are made, it is necessary to determine areas of application of the known reconstruction algorithms. Specifically, it is crucial to find advantages and disadvantages of known approaches to mesh generation, depending on the properties of the object and compare the quality of their results. In order to provide reliable ground-truth data, three 3D models with features resembling those identified in anatomical structures have been created. Based on these meshes, sets of CT-like DICOM images have been generated. Five different reconstruction approaches were proposed: using 3D occupancy information directly, two ways of obtaining point clouds and two methods that utilize Signed Distance Field. A neural network architecture for the SDF upsampling has also been presented. The obtained results justify the popularity of the Marching Cubes algorithm, as it produced accurate reconstructions most reliably. However, for certain scenarios, promising alternatives have been found. The presented outcomes make it clear that the approach to reconstruction must be tailored to the specific problem. Full article

Frequency response analysis (FRA) is recognized as an effective method in power transformer winding fault diagnosis. However, the traditional numerical index methods focus on the overall features of FRA curves, making it difficult to capture subtle deformations in transformer windings. Similarly, existing digital image processing methods rely on a single feature or a simple feature combination without adaptive fusion. These methods ignore differences in the data distributions of features, leading to feature mismatch, the loss of sensitive fault information, and lower diagnostic accuracy. To solve this problem, a novel adaptive multiple-image-feature fusion method for transformer winding fault diagnosis is proposed. First, a multi-dimensional feature space combining image pixel matrix similarity, morphological features, and image texture features is built to decode the difference in fault of FRA images. Second, the multiple kernel learning (MKL) framework is used to dynamically adjust the fusion weights of different kernels to make features compatible and remove redundant information. Finally, comparative and ablation experiments show that the proposed method outperforms the traditional methods in identifying different types and levels of faults. The method achieves over 99% accuracy in fault type identification across SVM, KNN, and RF classifiers. For radial deformation (RD) severity prediction, the accuracy of the proposed model is 93.37% with SVM and 94.85% with KNN, outperforming the full-feature concatenation method. These results confirm the method’s robustness and diagnostic precision. Full article