Diagnostics最新文献

Background: The mental foramen is a key anatomical structure of the mandible through which the mental nerve and accompanying vessels emerge. Accurate knowledge of its location and morphology is essential for safe dental and surgical procedures in the anterior mandible. Objective: This study was conducted as a systematic review to summarize current evidence on the morphology, localization, and anatomical variants of the mental foramen and their clinical relevance. Methods: The PubMed and Google Scholar databases were searched for studies published between 2015 and 2025 in accordance with current systematic review guidelines. Cone-beam computed tomography (CBCT) studies and anthropological investigations assessing the position, dimensions, and anatomical variants of the mental foramen were included. Results: Thirty-five studies (30 CBCT-based and 5 anthropological) comprising a total of 6240 mandibles or patients were analyzed qualitatively. Considerable variability was observed in the horizontal and vertical position of the mental foramen in relation to mandibular borders and dental landmarks. Anatomical variations included differences in size and shape, the presence of unilateral or bilateral accessory mental foramina, and rare cases of unilateral or bilateral absence of the foramen. Conclusions: The synthesis of recent CBCT and anthropological data across diverse populations highlights clinically relevant patterns of variability. This study identifies key positional patterns and variants of the mental foramen, which can inform clinical planning and help reduce the risk of mental nerve injury.

Background: Vaginal infections often present with overlapping symptoms and involve single or multiple pathogens. However, the relationship between clinical symptoms and molecularly defined vaginal pathogen profiles, especially in multi-pathogen infections, remains poorly characterized in a routine care setting. This study exams the connection between vaginal symptoms and pathogen profiles among women with vaginitis in Northern Vietnam. Methods: We conducted a multicenter cross-sectional study of women with vaginitis at Bac Ninh CDC and Hanoi Obstetrics and Gynecology Hospital between December 2023 and December 2024. Baseline demographics and clinical symptoms were assessed by physicians. Vaginal swabs were collected for pH measurement and pathogen detection using multiplex real-time PCR. The correlation was analyzed using logistic regression in GraphPad Prism v10.1.1. Results: Among 289 symptomatic women, abnormal vaginal discharge and itching were the most common symptoms. Gardnerella vaginalis was the most commonly detected pathogen, occurring alone or in combination with Candida albicans, Mycoplasma hominis, and other genital pathogens. Multi-pathogen infection was associated with abnormal vaginal discharge (OR = 5.44), itching (OR = 2.13), and elevated vaginal pH (OR = 4.70). Women at the tertiary hospital showed greater symptom burden (OR = 1.75) and higher prevalence of multi-pathogen infections (OR = 9.75) than those attending the provincial CDC. Conclusions: Multiplex real-time PCR combined with simple clinical indicators (symptom clustering and vaginal pH) provides practical diagnostic value for identifying multi-pathogen infections in symptomatic women. This integrated approach may support more accurate etiologic diagnosis and guide rational testing strategies, particularly in resource-limited settings.

Caseous calcification of the mitral annulus (CCMA) is a liquefactive necrosis of mitral annular calcification (MAC). CCMA is rare and usually asymptomatic, has a benign course, and, when incidentally found, can be misdiagnosed as a thrombus, abscess, cardiac tumor or vegetation. Although rarely, CCMA may complicate with rupture, which can lead to ventricular-atrial fistulization, pseudoaneurysm, severe mitral regurgitation (with possible heart failure and atrial fibrillation) and systemic embolism of caseous material (with cerebral ischemic events). A significant increase in CCMA dimensions and an infectious involvement of liquefactive necrosis make CCMA prone to rupture. To date, only case reports and some case series have been published on CCMA, without focusing on the pathophysiological mechanisms responsible for rupture, nor recommendations for prevention and management. However, despite general concerns about surgical treatment of CCMA because of high perioperative risks, most published cases actually underwent successful cardiac surgery. In the present review, we conducted a systematic review of the studies published in the medical literature up to March 2025, reporting cases of CCMA and its complications, as identified through the PubMed database. We analyzed clinical and biological risk factors for CCMA rupture and its diagnostic criteria, focusing on imaging features differentiating mitral annular calcification from uncomplicated CCMA and ruptured CCMA. To this regard, we focused on the key role of multimodality imaging in the achievement of the correct diagnosis. Finally, we propose a management strategy for CCMA, with the aim to fill a gap in this field in the current literature.

Background/Objectives: Accurate diagnosis and staging of periodontitis rely on clinical measurements and radiographic assessment of alveolar bone loss. Methods: Studies published between 1 January 2020 and 31 October 2025 were searched in the Web of Science and PubMed databases in accordance with the PRISMA guidelines. Original research articles that evaluated periodontal pathology on radiographic images using fractal analysis and/or artificial intelligence approaches, with clearly defined methodologies, were included. Due to methodological heterogeneity, a quantitative meta-analysis was not performed, and the findings were summarized using a narrative synthesis approach. Results: Of 346 records, 80 studies (9 fractal, 71 AI) met the inclusion criteria. Fractal analysis studies predominantly calculated the fractal dimension on panoramic or periapical radiographs using the box-counting method. In artificial intelligence studies, the task types mainly comprised classification, segmentation, detection, and hybrid approaches (multi-stage models or models combining multiple tasks). Panoramic and intraoral radiographs were the predominant imaging modalities. Performance metrics were reported across wide ranges (sensitivity 0.23-1.00; accuracy 0.506-1.00; specificity 0.41-0.99; F1 score 0.15-0.99; AUC 0.75-0.99), and in some studies, these metrics were only partially reported. Conclusions: Fractal analysis and artificial intelligence approaches offer objective and reproducible assessment of periodontal bone loss; however, methodological and reporting heterogeneity limit comparability and generalizability. Standardization of ROI definitions, datasets, study designs, and performance reporting is needed to improve clinical applicability. Future research should also explore hybrid models that combine the quantitative microstructural insights of fractal analysis with the automated detection capabilities of artificial intelligence to enhance diagnostic precision.

Background: Radiomic analyses have been extensively explored in oncologic imaging and more recently in neuroimaging. However, radiomic characterization of the crystalline lens using computed tomography has not yet been systematically investigated. Methods: In this retrospective study, semiautomatic segmentation of the eye lens on orbital CT was performed on 112 patients (mean age 48 ± 20 years, 38% female). After radiomics feature extraction, a Boruta feature selection approach based on the random forest algorithm was applied to select the most relevant radiomics features. Severity of white matter lesions were graded according to the Fazekas scale for each patient on axial non-contrast head CT. Results: In total, 17 important features were associated with age-related changes in the eye lens and three important radiomic features for the differentiation between patients with a Fazekas score > 1 and a control group. Significantly higher values were found in patients with a Fazekas score > 1 compared to the control group regarding all three features, "ClusterShade", "Skewness" and "DifferenceVariance" (p = 0.0006, 0.0023 and 0.0376, respectively), which are all measures of heterogeneity. No important radiomic features of the eye lens were confirmed between patients with and without hypertension. Conclusions: To the best of our knowledge, this is the first study to use CT-based radiomic analysis of the crystalline lens to detect differences among demographic or clinical groups with small vessel disease. The present results might help to expand the range of applications of radiomics regarding ophthalmic (patho-)physiology and suggest a possible new biomarker for systemic vascular diseases.

Background: Sodium/glucose cotransporter-2 inhibitors (SGLT2is), such as dapagliflozin and empagliflozin, are currently a standard therapy for heart failure (HF) patients. We report the real-world use of SGLT2is in a monocentric cohort of HF patients with reduced ejection fraction (HFrEF) and improved ejection fraction (HFimpEF), comparing patient characteristics and outcomes with those observed in large-scale randomized clinical trials (RCTs). Methods: We retrospectively analyzed a cohort of 370 stable patients with HFrEF or HFimpEF who initiated therapy with dapagliflozin or empagliflozin between June 2019 and November 2023. Baseline data, including medical history, concomitant diseases, therapy, laboratory tests, echocardiographic results and cardiopulmonary exercise tests (CPETs), were collected at the start of the therapy with SGLT2is. After a median period of 18 months, follow-up data on treatment adherence, adverse events, hospitalizations, and mortality were also reviewed. A comparison was made between patients taking dapagliflozin and those taking empagliflozin and then individual populations were compared with those from the trials. Results: Among 370 patients (81% HFrEF, 19% HFimpEF), 276 received dapagliflozin and 94 empagliflozin. Empagliflozin patients were older, had higher NYHA class and LVEF, and higher incidence of diabetes, while dapagliflozin users had greater use of sacubitril/valsartan and mineralocorticoid receptor antagonists. Both groups were older than the RCT cohorts. Dapagliflozin patients had LVEF comparable to DAPA-HF, while empagliflozin patients had higher LVEF than EMPEROR-Reduced. HF hospitalizations were more frequent in the real-world groups, but mortality was lower than in RCTs. The composite outcome of death and worsening HF was higher in the real-world dapagliflozin cohort vs. DAPA-HF but similar between the real-world empagliflozin cohort and EMPEROR-Reduced. Conclusions: In this real-world cohort, the use of empagliflozin was associated with cardio-nephro-metabolic comorbidities and dapagliflozin being prescribed more frequently for patients with isolated cardiac symptoms. While outcomes were generally favorable, they differed from those seen in RCTs, highlighting the importance of real-world data in understanding the practical application of these therapies.

Background: Cardiovascular magnetic resonance (CMR) is the clinical gold standard for assessing cardiac anatomy and function. However, the manual segmentation of cardiac structures and myocardial infarction (MI) is time-consuming, prone to inter-observer variability, and often depends on contrast-enhanced imaging. Although deep learning (DL) has enabled substantial automation, challenges remain in generalizability, particularly for MI detection from non-contrast cine CMR. Objective: This study proposes a comprehensive DL-based framework for automatic segmentation of cardiac structures and myocardial infarction using contrast-free cine CMR. Methods: The framework integrates multiple convolutional neural network (CNN) architectures for cardiac structure segmentation with an attention-based deep learning model for MI localization. Post-processing refinement using stacked autoencoders and active contour modeling is applied to improve anatomical consistency. Segmentation performance is evaluated using overlap-based and boundary-based metrics, including the Dice Similarity Coefficient (DSC), Mean Contour Distance (MCD), and Hausdorff Distance (HD). Results: The best-performing model achieved Dice scores of 0.93 ± 0.05 for the left ventricular (LV) cavity, 0.89 ± 0.04 for the LV myocardium, and 0.91 ± 0.06 for the right ventricular (RV) cavity, with consistently low boundary errors across all structures. Myocardial infarction segmentation achieved a Dice score of 0.80 ± 0.02 with high recall, demonstrating reliable infarct localization without the use of contrast agents. Conclusions: By enabling accurate cardiac structure and myocardial infarction segmentation from contrast-free cine CMR, the proposed framework supports broader clinical applicability, particularly for patients with contraindications to gadolinium-based contrast agents and in emergency or resource-limited settings. This approach facilitates scalable, contrast-independent cardiac assessment.

Background/Objectives: Fundus imaging provides a detailed view of the interior surface of the eye and plays a crucial role in the early diagnosis of retinal diseases. However, automated interpretation of fundus images remains challenging due to variations in illumination, noise, and structural complexity. Methods: A novel hybrid model that integrates the Intrinsic Mode Function (IMF) filter, derived from Empirical Mode Decomposition (EMD), with a Light Convolutional Neural Network (LightCNN) for enhanced fundus image classification was proposed. The IMF filter effectively decomposes the input signal into intrinsic components, isolating high-frequency noise and preserving critical retinal patterns. These refined components are subsequently processed by the LightCNN architecture, which offers lightweight yet highly discriminative feature extraction and classification capabilities. Results: Experimental results on DIARETDB fundus datasets demonstrate that the proposed IMF + LightCNN model achieves 99.4% accuracy, 99.1% precision, 98.87% recall, and a 98.31 F1-score, significantly outperforming conventional CNN and ResNet-based models. Conclusions: Integrating advanced signal processing with lightweight deep learning improves both diagnostic accuracy and computational efficiency. This hybrid framework establishes a promising pathway for reliable and real-time clinical screening of retinal diseases.

Background/Objectives: Early tracheostomy seems favorable in prolonged ventilated patients after surgery. Hence, predicting tracheostomy after cardiac surgery is essential. Recently proposed prediction models aim to support this decision-making process, but their diagnostic validity across other patient populations remains uncertain. Methods: A retrospective single-center study was performed at a university hospital. The patient sample included consecutive patients between 2010 and 2020 who underwent cardiac surgery. Patients who underwent tracheostomy after cardiac surgery were assigned to the intervention group. Control group patients, who had not undergone tracheostomy, were randomly assigned to the group. An existing model was evaluated by receiver operating characteristics curve analysis. Four sets of risk features were chosen depending on results from regression analysis, lasso regularization, random forest or clinical domain knowledge. Newly developed models were created using machine learning methods: random forest, naïve Bayes, nearest neighbor and deep learning. Multiple models were trained with either feature set and then assessed using confusion matrices on an independent test set. Results: A total of 4744 patients were included in this study. One-hundred and eighteen patients were included in the tracheostomy group. Diagnostic accuracy of the existing model showed insufficient discrimination (area under the curve (AUC) = 0.57). Likewise, newly developed models also showed overall poor diagnostic discrimination across all feature sets and algorithms. Conclusions: This study shows the diagnostic limitations of retrospective clinical data for the diagnostic prediction of tracheostomy, thereby informing the design of future prospective diagnostic studies. Training new models should not rely on retrospective data alone. Instead, prospective data collection and integration of physiological or imaging-based diagnostics could likely contribute to the development of a good classifier.

Background/Objectives: Virtual histological staining offers a rapid, cost-effective alternative to physical reprocessing but faces challenges related to spatial misalignment and staining heterogeneity between Hematoxylin and Eosin (H&E) and Masson's Trichrome (MT) domains. This study develops a robust framework for H&E-to-MT virtual staining to enable accurate fibrosis assessment without additional tissue consumption. Methods: We propose a transformer-based generative adversarial network (TbGAN) supported by a multi-stage alignment pipeline (SIFT (scale-invariant feature transform) coarse alignment, ORB/homography patch registration, and B-spline free-form deformation) and a weighted fusion mechanism combining four configuration outputs (O/10/3, O/3/10, R/10/3, and R/3/10). The framework was validated on 27 whole-slide images (>100,000 aligned patches) through 24 independent experiments. Results: The fused approach achieved state-of-the-art performance: MI = 0.9815 ± 0.0934, SSIM = 0.7474 ± 0.0597, NCC = 0.9320 ± 0.0220, and CS = 0.9946 ± 0.0014. Statistical analysis confirmed enhanced stability through narrower interquartile ranges, fewer outliers, and tighter 95% confidence intervals compared to individual configurations. Qualitative assessment demonstrated preserved collagen morphology critical for fibrosis staging. Conclusions: Our framework provides a reliable, IRB-compliant solution for virtual MT staining that maintains high structural fidelity suitable for diagnostic support. It enables resource-efficient fibrosis quantification and supports integration into clinical digital pathology workflows without patient-specific recalibration.