Diagnostics最新文献

Background: There have been multiple studies on the use of Doppler ultrasound to define skin inflammation, but the visible vessels of healthy skin have yet to be described. Objective: This study aimed to evaluate the visible vessels of healthy skin using Doppler ultrasound. Methods: Prospective multicenter study using Doppler ultrasound to analyze healthy skin. The color percentage, flow velocity, and maximum vessel diameter were calculated. Results: 943 images from 152 patients were recorded. The most frequently used mode was color Doppler (40.6%), followed by power Doppler (30.4%). Visible vessels were detected in 18.23%; in positive Doppler images, color occupied less than 5%. The malar region exhibited the highest visible vessels. The 22 MHz probe detected smaller vessels with slower flows than the 18 MHz probe. Spectral Doppler showed peak systolic values of less than 10 cm/s and a vessel diameter of less than 1 mm. In most of the participating centers, the operators had less than 10 years of experience in performing skin ultrasound examinations. Sensitivity of the Doppler may vary according to the device. Conclusions: With the used ultrasound equipment, it was uncommon to visualize vessels in healthy skin. When seen, they covered less than 5% of the image with low flow and small size.

Abdominal imaging has undergone a significant transformation in recent years, driven by the rapid evolution of diagnostic technologies and their integration into clinical practice [...].

Background/Objectives: Early detection and diagnosis are important when treating gallbladder (GB) diseases. Poorer clinical outcomes and increased patient symptoms may result from any error or delay in diagnosis. Many signs and symptoms, especially those related to GB diseases with similar symptoms, may be unclear. Therefore, highly qualified medical professionals should interpret and understand ultrasound images. Considering that diagnosis via ultrasound imaging can be time- and labor-consuming, it may be challenging to finance and benefit from this service in remote locations. Methods: Today, artificial intelligence (AI) techniques ranging from machine learning (ML) to deep learning (DL), especially in large datasets, can help analysts using Content-Based Image Retrieval (CBIR) systems with the early diagnosis, treatment, and recognition of diseases, and then provide effective methods for a medical diagnosis. Results: The developed model is compared with two different textural and six different Convolutional Neural Network (CNN) models accepted in the literature-the developed model combines features obtained from three different pre-trained architectures for feature extraction. The cosine method was preferred as the similarity measurement metric. Conclusions: Our proposed CBIR model achieved successful results from six other different models. The AP value obtained in the proposed model is 0.94. This value shows that our CBIR-based model can be used to detect GB diseases.

Background and Objectives: We conducted a retrospective observational study to evaluate the impact of elevated blood glucose levels in patients with SARS-CoV-2 infection and a prior diagnosis of diabetes mellitus (DM) or newly diagnosed hyperglycemia. Materials and Methods: This study analyzed 6065 patients admitted to the COVID-19 departments of the "Marius Nasta" National Institute of Pulmonology in Bucharest, Romania, between 26 October 2020 and 5 January 2023. Of these, 813 patients (13.40%) were selected for analysis due to either a pre-existing diagnosis of DM or hyperglycemia at the time of hospital admission. Results: The erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels were elevated in patients with blood glucose levels exceeding 300 mg/dL. These elevations correlated with the presence of respiratory failure and increased mortality rates. Additionally, oxygen requirements were significantly higher at elevated blood glucose levels (p < 0.001), with a direct relationship between glycemia and oxygen demand. This was accompanied by lower oxygen saturation levels (p < 0.001). Maximum blood glucose levels were associated with the severity of respiratory failure (AUC 0.6, 95% CI: 0.56-0.63, p < 0.001). We identified cut-off values for blood glucose at admission (217.5 mg/dL) and maximum blood glucose during hospitalization (257.5 mg/dL), both of which were associated with disease severity and identified as risk factors for increased mortality. Conclusions: High blood glucose levels, both at admission and during hospitalization, were identified as risk factors for poor prognosis and increased mortality in patients with SARS-CoV-2 infection, regardless of whether the hyperglycemia was due to a prior diagnosis of DM or was newly developed during the hospital stay. These findings underscore the importance of glycemic control in the management of hospitalized COVID-19 patients.

Background: Medical diagnosis for skin diseases, including leukemia, early skin cancer, benign neoplasms, and alternative disorders, becomes difficult because of external variations among groups of patients. A research goal is to create a fusion-level deep learning model that improves stability and skin disease classification performance. Methods: The model design merges three convolutional neural networks (CNNs): EfficientNet-B0, EfficientNet-B2, and ResNet50, which operate independently under distinct branches. The neural network model uses its capability to extract detailed features from multiple strong architectures to reach accurate results along with tight classification precision. A fusion mechanism completes its operation by transmitting extracted features to dense and dropout layers for generalization and reduced dimensionality. Analyses for this research utilized the 27,153-image Kaggle Skin Diseases Image Dataset, which distributed testing materials into training (80%), validation (10%), and testing (10%) portions for ten skin disorder classes. Results: Evaluation of the proposed model revealed 99.14% accuracy together with excellent precision, recall, and F1-score metrics. Conclusions: The proposed deep learning approach demonstrates strong potential as a starting point for dermatological diagnosis automation since it shows promise for clinical use in skin disease classification.

Background/Objectives: The emergence and spread of carbapenemase-producing Enterobacterales (CPE) represents a significant challenge prompting the need to optimize diagnostic tools to detect CPE carriers. The Xpert^® Carba-R assay (Cepheid, Sunnyvale, CA, USA), a Real-Time PCR-based test, can detect the bla_KPC, bla_VIM, bla_OXA-48, bla_NDM and bla_IMP carbapenemase genes directly from rectal swabs. This study assessed the performance of the Xpert^® Carba-R assay using FecalSwab™ (Copan, Brescia, Italy), a liquid-based collection device. Methods: The first part of the study aimed to establish the FecalSwab^TM volume which gave the most similar Ct values to those obtained by the Transystem™ double swab (Copan). The best volume was then used to assess the limit of detection (LoD) for each target and compare the accuracy of different FecalSwab^TM storage conditions (room temperature or 4 °C after 16 h compared to T₀). Results: The results indicated that using 200 µL of the FecalSwab™ medium provided reliable Ct values, with the lowest number of invalid samples compared to traditional methods. The average LoDs for different carbapenemases ranged from 4.7 × 10³ to 6.8 × 10³ CFU/mL. FecalSwab™ showed a better performance after 16 h at room temperature compared to storage at 4 °C. Conclusions: This study supports single sampling with the FecalSwab™ medium for both molecular and cultural methods, for the potential optimization of CPE screening.

Background: In the study of coronary artery disease, the mechanisms underlying atherosclerosis initiation and progression or regression remain incompletely understood. Our research conceptualized the cardiovascular system as an integrated network of pumps and pipes, advocating for a paradigm shift from static imaging of coronary stenosis to dynamic assessments of coronary flow. Further review of fluid mechanics highlighted the water hammer phenomenon as a compelling analog for processes in coronary arteries. Methods: In this review, the analytical methodology employed a comprehensive, multifaceted approach that incorporated a review of fluid mechanics principles, in vitro acoustic experimentation, frame-by-frame visual angiographic assessments of in vivo coronary flow, and an artificial intelligence (AI) protocol designed to analyze the water hammer phenomenon within an acoustic framework. In the analysis of coronary flow, the angiograms were selected from patients with unstable angina if they had previously undergone one or more coronary angiograms, allowing for a longitudinal comparison of dynamic flow and phenomena. Results: The acoustic investigations pinpointed pockets of contrast concentrations, which might correspond to compression and rarefaction zones. Compression antinodes were correlated to severe stenosis, due to rapid shifts from low-pressure diastolic flow to high-pressure systolic surges, resulting in intimal injury. Rarefaction antinodes were correlated with milder lesions, due to de-escalating transitions from high systolic pressure to lower diastolic pressure. The areas of nodes remained without lesions. Based on the locations of antinodes and nodes, a coronary acoustic action map was constructed, enabling the identification of existing lesions, forecasting the progression of current lesions, and predicting the development of future lesions. Conclusions: The results suggested that intimal injury was likely induced by acoustic retrograde pressure waves from the water hammer phenomenon and developed new lesions at specifically exact locations.

Background/Objectives: Artificial intelligence (AI) has attracted great interest due to its applicability in many fields. The adoption of visual illustration techniques produced by AI in the field of graphic design has further expanded the field of use of this technology. This study focuses on foot anatomy illustrations generated by Adobe Firefly and Microsoft Designer Image Creator applications, evaluating them based on detail, clarity, anatomical realism, accuracy, and aesthetic appeal. Methods: The illustrations were created using text-based scripts, and five anatomists compared them to traditional illustrations from the Sobotta Atlas of Human Anatomy. Results: Fleiss' Kappa statistic was used to analyze consistency among expert evaluations. For the four figures generated by both AI applications, Fleiss' Kappa agreement was high. Adobe Firefly performed slightly better in illustrating phalanx and ankle bones, but its anatomical accuracy was lower for tarsal and metatarsal bones. Microsoft Designer Image Creator excelled in illustrating metatarsal bones, while its tarsal and phalanx illustrations were less anatomically accurate than Adobe Firefly and the atlas drawings. Both programs showed average realism in ankle structures, while the tarsal bones had low realism. Conclusions: Artificial intelligence applications within the scope of the study showed fast performance. Aesthetic appeal is dominant at first glance in the resulting drawings. In general, both applications have struggled to reflect anatomical reality.

Background/Objectives: Given the significant economic and social burden of osteoporosis, there is growing interest in developing an efficient alternative to the traditional dual-energy X-ray absorptiometry (DXA). Radiofrequency Echographic Multi Spectrometry (REMS) is an innovative, non-ionizing imaging technique that recently emerged as a viable tool to diagnose osteoporosis and estimate the fragility fracture risk. Nevertheless, its clinical use is still limited due to its novelty and continuing uncertainty of long-term performance. Methods: In order to evaluate the accuracy of the REMS, a systematic review of the English-language literature was conducted. Three databases were searched for relevant publications from 1 January 2015 until 1 December 2024 using the keyword combinations "(radiofrequency echographic multi spectrometry OR REMS) AND (dual-energy X-ray absorptiometry OR DXA)". The initial search yielded 602 candidate articles. After screening the titles and abstracts following the eligibility criteria, 17 publications remained for full-text evaluation. Results: The reviewed studies demonstrated strong diagnostic agreement between REMS and DXA. Additionally, REMS showed enhanced diagnostic capabilities in cases where lumbar bone mineral density measurements by DXA were impaired by artifacts such as vertebral fractures, deformities, osteoarthritis, or vascular calcifications. REMS exhibited excellent intra-operator repeatability and precision, comparable to or exceeding the reported performance of DXA. The fragility score (FS), a parameter reflecting bone quality and structural integrity, effectively discriminated between fractured and non-fractured patients. Moreover, REMS proved to be a radiation-free option for bone health monitoring in radiation-sensitive populations or patients requiring frequent imaging to assess fracture risk. Conclusions: This current study underscores the robustness of REMS as a reliable method for diagnosing and monitoring osteoporosis and evaluating bone fragility via the FS. It also identifies critical knowledge gaps and emphasizes the need for further prospective studies to validate and expand the clinical applications of REMS across diverse patient populations.

Background: Early and accurate detection of breast cancer is crucial for improving treatment outcomes and survival rates. To achieve this, innovative imaging technologies such as microwave radiometry (MWR)-which measures internal tissue temperature-combined with advanced diagnostic methods like deep learning are essential. Methods: To address this need, we propose a hierarchical self-contrastive model for analyzing MWR data, called Joint-MWR (J-MWR). J-MWR focuses on comparing temperature variations within an individual by analyzing corresponding sub-regions of the two breasts, rather than across different samples. This approach enables the detection of subtle thermal abnormalities that may indicate potential issues. Results: We evaluated J-MWR on a dataset of 4932 patients, demonstrating improvements over existing MWR-based neural networks and conventional contrastive learning methods. The model achieved a Matthews correlation coefficient of 0.74 ± 0.02, reflecting its robust performance. Conclusions: These results emphasize the potential of intra-subject temperature comparison and the use of deep learning to replicate traditional feature extraction techniques, thereby improving accuracy while maintaining high generalizability.