Diagnostics最新文献

Background and Study Aim: Colonoscopy holds the highest volume of all endoscopic procedures, allowing for large colonoscopy databases to serve as valuable datasets for quality assurance. We aimed to build a comprehensive colonoscopy database for quality assurance and the training of future AIs. Materials and Methods: As part of a cluster-randomized controlled trial, a designated, onsite medical student was used to acquire procedural and patient-specific data, ensuring a high level of data integrity. The following data were thereby collected for all colonoscopies: full colonoscopy vides, colonoscope position (XYZ-coordinates), intraprocedural timestamps, pathological report, endoscopist description, endoscopist planning, and patient-reported discomfort. Results: A total of 1447 patients were included from the 1st of February 2022 to the 21st of November 2023; 1191 colonoscopies were registered as completed, 88 were stopped due to inadequate bowel cleansing, and 41 were stopped due to patient discomfort. Of the 1191 completed colonoscopies, 601 contained polypectomies (50.4%), and 590 did not (49.6%). Comparing colonoscopies with polypectomies to those without the withdrawal time (caecum to extubating the scope) was significantly longer for all parts of the colon (p values < 0.001), except the transverse colon (p value = 0.92). The database was used to train an AI, automatically and objectively evaluating bowel preparation. Conclusions: We established the most thorough database in colonoscopy with previously inaccessible information, indicating that the transverse colon differs from the other parts of the colon in terms of withdrawal time for procedures with polypectomies. To further explore these findings and reach the full potential of the database, an AI evaluating bowel preparation was developed. Several research partners have been identified to collaborate in the development of future AIs.

Background/Objectives: Ultrasound (US) imaging plays a crucial role in the early detection and treatment of hepatocellular carcinoma (HCC). However, challenges such as speckle noise, low contrast, and diverse lesion morphology hinder its diagnostic accuracy. Methods: To address these issues, we propose CSM-FusionNet, a novel framework that integrates clustering, SoftMax-weighted Box Fusion (SM-WBF), and padding. Using raw US images from a leading hospital, Samsung Medical Center (SMC), we applied intensity adjustment, adaptive histogram equalization, low-pass, and high-pass filters to reduce noise and enhance resolution. Data augmentation generated ten images per one raw US image, allowing the training of 10 YOLOv8 networks. The mAP@0.5 of each network was used as SoftMax-derived weights in SM-WBF. Threshold-lowered bounding boxes were clustered using Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and outliers were managed within clusters. SM-WBF reduced redundant boxes, and padding enriched features, improving classification accuracy. Results: The accuracy improved from 82.48% to 97.58% with sensitivity reaching 100%. The framework increased lesion detection accuracy from 56.11% to 95.56% after clustering and SM-WBF. Conclusions: CSM-FusionNet demonstrates the potential to significantly improve diagnostic reliability in US-based lesion detection, aiding precise clinical decision-making.

Background/Objectives: The Geriatric Trauma Outcome Score (GTOS) is used to predict in-hospital mortality in geriatric patients with trauma. However, its applicability to elderly patients with multiple traumas and severe traumatic brain injury (TBI) remains poorly understood. This study aimed to evaluate the predictive accuracy of the GTOS in elderly patients with multiple traumas and TBI and assess its performance in patients with mild and severe TBI. Methods: We retrospectively analyzed 1283 geriatric multiple trauma patients (aged ≥ 65 years) treated at a regional trauma center from 2019 to 2023. Patients were stratified into mild (head Abbreviated Injury Scale [AIS] ≤ 3) and severe (head AIS ≥ 4) TBI groups. GTOS values were calculated for each patient, and predicted mortality was compared with in-hospital mortality. GTOS predictive accuracy was assessed by analyzing the receiver operating characteristic curve. Results: Patients had a median Injury Severity Score of 18 (interquartile range: 10-25); 33.3% of patients received red blood cell transfusions within 24 h. The overall in-hospital mortality rate was 17.9%; GTOS predicted a mortality rate of 17.6% ± 0.17. The GTOS accurately predicted the in-hospital mortality in the entire cohort, achieving an Area Under the Curve (AUC) of 0.798. Predictive accuracy diminished for patients with severe TBI (AUC = 0.657), underestimating actual mortality (39.5% vs. 28.8% predicted). Conclusions: While the GTOS remains a useful tool for predicting in-hospital mortality in elderly patients with multiple traumas, it consistently underestimates mortality risk in those with severe TBI. Therefore, applying the GTOS in this patient subgroup warrants careful consideration.

Background: Invasive coronary angiography is the gold standard for assessing in-stent restenosis (ISR) in patients with coronary artery disease. However, the predictive value of non-invasive Tissue Doppler Imaging (TDI) to evaluate patients with ISR has not been studied extensively. Methods: A total of 41 patients (19 with acute myocardial infarction and 22 with stable angina pectoris) who received percutaneous coronary intervention (PCI) were enrolled in the study. Time-to-peak velocities (TpV) of 12 non-apical segments of the left ventricle, by pulse wave TDI echocardiography, were obtained within two days prior to the PCI and six months later. Results: A 12-segmental mean TpV ≥ 279.6 ms at six months after PCI was able to detect ISR (odds ratio: 2.09, 95% CI 1.004-4.352, p = 0.049). Moreover, a significant decrease in the standard deviation of TpV was demonstrated in patients without ISR (85.8 ± 44.8 vs. 60.3 ± 31.7 ms, p = 0.001), but not in patients with ISR (97.7 ± 53.3 vs. 91.2 ± 52.6 ms, p = 0.57). Conclusions: Pulse-wave TDI echocardiography is a promising tool in the detection of ISR six months after PCI in patients with coronary artery disease.

Background/Objectives: Breast cancer is among the most frequently diagnosed cancers and leading cause of mortality worldwide. The accurate classification of breast cancer from the histology photographs is very important for the diagnosis and effective treatment planning. Methods: In this article, we propose a DenseNet121-based deep learning model for breast cancer detection and multi-class classification. The experiments were performed using whole-slide histopathology images collected from the BreakHis dataset. Results: The proposed method attained state-of-the-art performance with a 98.50% accuracy and an AUC of 0.98 for the binary classification. In multi-class classification, it obtained competitive results with 92.50% accuracy and an AUC of 0.94. Conclusions: The proposed model outperforms state-of-the-art methods in distinguishing between benign and malignant tumors as well as in classifying specific malignancy subtypes. This study highlights the potential of deep learning in breast cancer diagnosis and establishes the foundation for developing advanced diagnostic tools.

Background: The development and initial testing of an optomechatronic system for the reconstruction of three-dimensional (3D) images to identify abnormalities in breast tissue and assist in the diagnosis of breast cancer is presented. Methods: This system combines 3D reconstruction technology with diffuse optical mammography (DOM) to offer a detecting tool that complements and assists medical diagnosis. DOM analyzes tissue properties with light, detecting density and composition variations. Integrating 3D reconstruction enables detailed visualization for precise tumor localization and sizing, offering more information than traditional methods. This technological combination enables more accurate, earlier diagnoses and helps plan effective treatments by understanding the patient's anatomy and tumor location. Results: Using Chinese ink, it was possible to identify simulated abnormalities of 10, 15, and 20 mm in diameter in breast tissue phantoms from cosmetic surgery. Conclusions: Data can be processed using algorithms to generate three-dimensional images, providing a non-invasive and safe approach for detecting anomalies. Currently, the system is in a pilot testing phase using breast tissue phantoms, enabling the evaluation of its accuracy and functionality before application in clinical studies.

Cardiovascular inflammation has recently emerged as a critical issue across various cardiovascular diseases. Various non-invasive imaging modalities are applied for assessing cardiovascular inflammation. Positron emission tomography (PET) using ¹⁸F-fluorodeoxyglucose (FDG) is a valuable non-invasive imaging tool for identifying active cardiovascular inflammation. It is utilized in evaluating conditions, such as cardiac sarcoidosis, endocarditis, vasculitis, and unstable atherosclerosis. Furthermore, management of cardiovascular complications after aggressive cancer therapy has increasingly been required in cancer patients. FDG PET is considered a suitable approach not only for the assessment of tumor responses to cancer therapy, but also for early and accurate detection of cardiovascular complications. This review highlights the clinical value of FDG PET under appropriate patient preparation. The future perspectives of new molecular imaging tools for assessing active cardiovascular inflammation have been described.

Desmosine and isodesmosine (DID) are elastin-specific crosslinking amino acids that play a critical role in maintaining the structural integrity of elastic fibers, and their levels in body fluids may serve as biomarkers for alveolar wall injury. To support this concept, we present studies demonstrating the use of DID to detect elastic fiber damage that reflects distention and the rupture of airspaces. The emergence of airspace enlargement may be modeled by a percolation network describing the effect of changing proportions of intact and weak elastic fibers on the transmission of mechanical forces in the lung. Following the unraveling and fragmentation of weakened elastic fibers, the release of DID may correlate with an increasing alveolar diameter and provide an endpoint for clinical trials of novel agents designed to treat pulmonary emphysema. The limitations of the DID measurements related to specificity and reproducibility are also addressed, particularly regarding sample source and analytical techniques. Standardizing protocols to isolate and quantify DID may increase the use of this biomarker for the early detection of alveolar wall injury, which permits timely therapeutic intervention.

Carbon monoxide (CO) poisoning is a significant public health issue, with diagnosis often complicated by non-specific symptoms and limited access to specialised tools. Early detection is vital for preventing long-term complications. The review examines diagnostic challenges, prognostic factors, management strategies, and future advancements in CO poisoning. It highlights the limitations of current diagnostic techniques such as blood carboxyhaemoglobin levels and pulse CO-oximetry, while exploring emerging methods for rapid detection. Prognosis is influenced by exposure severity and delayed treatment, which increases the risk of neurological damage. Hyperbaric oxygen therapy (HBOT) remains the primary treatment but is not always accessible. Advances in portable CO-oximeters and biomarkers offer potential for improved early diagnosis and monitoring. Addressing resource limitations and refining treatment protocols are crucial for better patient outcomes. Future research should focus on personalised management strategies and the integration of modern technologies to enhance care.

Deep vein thrombosis (DVT) of the lower extremities, as part of venous thromboembolism disorder, is the third leading cause of acute cardiovascular syndrome after heart attack and stroke. It can result in disability due to pulmonary embolism (PE) and post-thrombotic syndrome (PTS), particularly in cases where the thrombosis extends to the iliofemoral veins. Anticoagulation therapy is effective in preventing thrombus propagation and embolism but may not be sufficient for thrombus degradation and venous patency restoration. Up to 50% of patients with iliofemoral DVT develop PTS, mainly due to venous outflow obstruction or valvular incompetence. To date, the advent of new devices that enables rapid thrombus elimination and the restoration of deep venous patency, known as the "OPEN VEIN hypothesis", may prevent valvular damage and reflux, cutting down the rate of PTS. Similarly, chronic venous disease could be related to a stenosis or occlusion of a major vein that can restrict blood flow. In this setting, intravascular ultrasound (IVUS) is an essential tool for correct diagnostic and therapeutic planning in acute and chronic vein disease. Only angiography in vein disease can limit the procedure's efficacy, with a high rate of stenosis misdiagnosed; IVUS provides further imaging that complements traditional angiographic study, and its role is now established by different international guidelines. If compared to angiography, IVUS allows for the evaluation of major axial veins in a 360-degree ultrasound image of the lumen and of the vessel wall structure. At the same time, the precise location and size of the major lower extremity veins allow for the placement of the stent to be more straightforward with a precise dimension of the vein in all of its diameters; moreover, other abnormalities should be visualized as acute or chronic thrombus, fibrosis, or trabeculations. This review aims to provide an in-depth analysis of IVUS findings in acute and chronic lower extremity DVT, emphasizing its diagnostic and therapeutic implications.