Diagnostics最新文献

Introduction: Cancer is a highly lethal disease with a significantly high mortality rate. One of the most commonly used methods for treatment is radiation therapy. However, cancer treatment using radiotherapy is a time-consuming process that requires significant manual work from planners and doctors. In radiation therapy treatment planning, determining the dose distribution for each of the regions of the patient's body is one of the most difficult and important tasks. Nowadays, artificial intelligence has shown promising results in improving the quality of disease treatment, particularly in cancer radiation therapy. Objectives: The main objective of this study is to build a high-performance deep learning model for predicting radiation therapy doses for cancer and to develop software to easily manipulate and use this model. Materials and Methods: In this paper, we propose a custom 3D convolutional neural network model with a U-Net-based architecture to automatically predict radiation doses during cancer radiation therapy from CT images. To ensure that the predicted doses do not have negative values, which are not valid for radiation doses, a rectified linear unit (ReLU) function is applied to the output to convert negative values to zero. Additionally, a proposed loss function based on a dose-volume histogram is used to train the model, ensuring that the predicted dose concentrations are highly meaningful in terms of radiation therapy. The model is developed using the OpenKBP challenge dataset, which consists of 200, 100, and 40 head and neck cancer patients for training, testing, and validation, respectively. Before the training phase, preprocessing and augmentation techniques, such as standardization, translation, and flipping, are applied to the training set. During the training phase, a cosine annealing scheduler is applied to update the learning rate. Results and Conclusions: Our model achieved strong performance, with a good DVH score (1.444 Gy) on the test dataset, compared to previous studies and state-of-the-art models. In addition, we developed software to display the dose maps predicted by the proposed model for each 2D slice in order to facilitate usage and observation. These results may help doctors in treating cancer with radiation therapy in terms of both time and effectiveness.

Marfan syndrome (MFS) is a genetic disorder affecting connective tissue, often leading to cardiovascular complications such as aortic aneurysms and mitral valve prolapse. Cardiovascular multimodality imaging plays a crucial role in the diagnosis, monitoring, and management of MFS patients. This review explores the advancements in echocardiography, cardiovascular magnetic resonance (CMR), cardiac computed tomography (CCT), and nuclear medicine techniques in MFS. Echocardiography remains the first-line tool, essential for assessing aortic root, mitral valve abnormalities, and cardiac function. CMR provides detailed anatomical and functional assessments without radiation exposure, making it ideal for long-term follow-up. CT offers high-resolution imaging of the aorta, crucial for surgical planning, despite its ionizing radiation. Emerging nuclear medicine techniques, though less common, show promise in evaluating myocardial involvement and inflammatory conditions. This review underscores the importance of a comprehensive imaging approach to improve outcomes and guide interventions in MFS patients. It also introduces novel aspects of multimodality approaches, emphasizing their impact on early detection and management of cardiovascular complications in MFS.

We appreciate the interest of Troisi and his colleagues [...].

Background: Italy, particularly the northern region of Lombardy, has experienced very high rates of COVID-19 cases and deaths. Several indicators, i.e., the number of new positive cases, deaths and hospitalizations, have been used to monitor virus spread, but all suffer from biases. The aim of this study was to evaluate an alternative data source from Emergency Medical Service (EMS) activities for COVID-19 monitoring. Methods: Calls to the emergency number (112) in Lombardy (years 2015-2022) were studied and their overlap with the COVID-19 pandemic, influenza and official mortality peaks were evaluated. Modeling it as a counting process, a specific cause contribution (i.e., COVID-19 symptoms, the "signal") was identified and enucleated from all other contributions (the "background"), and the latter was subtracted from the total observed number of calls using statistical methods for excess event estimation. Results: A total of 6,094,502 records were analyzed and filtered for respiratory and cardiological symptoms to identify potential COVID-19 patients, yielding 742,852 relevant records. Results show that EMS data mirrored the time series of cases or deaths in Lombardy, with good agreement also being found with seasonal flu outbreaks. Conclusions: This novel approach, combined with a machine learning predictive approach, could be a powerful public health tool to signal the start of disease outbreaks and monitor the spread of infectious diseases.

Background: Although the cure rates of classical Hodgkin Lymphoma (cHL) are as high as 90% using the current treatment protocols, the prognosis is poor for primary refractory patients. Thus, a biomarker that can predict patients with early progression at the time of diagnosis is an unmet clinical need. Endothelial activation and stress index (EASIX) and its variant modified EASIX (mEASIX) is a scoring system currently used for the prediction of prognosis in hematologic malignancies. This study aimed to investigate the prognostic value of the mEASIX score in newly diagnosed cHL patients. Methods: Data from 206 patients who underwent positron emission tomography (PET)-guided doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD) therapy for cHL between January 2007 and November 2023 were retrospectively analyzed. The prognostic value of the mEASIX score was evaluated using the receiver operating characteristic (ROC) analysis, Cox regression analysis, and the Kaplan-Meier method, and then compared with standard risk assessment methods. Results: The median age at diagnosis was 33 years, and the rate of patients in the advanced stage was 67%. ROC analysis determined an optimal mEASIX score cut-off of 17.28, categorizing patients into mEASIX^high (47%) and mEASIX^low (53%) groups. The 5-year progression-free survival (PFS) (60% vs. 84.3%) and overall survival (OS) (79.6% vs. 95.8%) were significantly lower in the mEASIX^high group (p < 0.001). Additionally, multivariate analysis showed that the independent variables affecting PFS included the nodular sclerosing subtype (HR: 0.4), bone marrow involvement (HR: 2.6), and elevated mEASIX (HR: 3.1). Independent variables, which had an effect on OS included elevated mEASIX (HR:3.8) and higher IPS-3 scores (HR:1.9). Furthermore, a higher mEASIX score (≥17.28) was identified as an independent variable indicating primary refractory disease (OR: 6.5). Conclusions: mEASIX is a powerful and easy-to-access marker for the detection of primary refractory disease and prognosis in newly diagnosed cHL cases.

Acute aortic syndrome (AAS) encompasses a range of life-threatening conditions, including classical dissection, intramural hematoma, and penetrating aortic ulcer. Each of these conditions presents distinct clinical characteristics and carries the potential to progress to rupture. Because AAS can be asymptomatic or present with diverse symptoms, its diagnosis requires clinical evaluation, risk scoring, and biomarkers such as D-dimer (DD), C-reactive protein (CRP), homocysteine, natriuretic peptides (BNP), and imaging modalities like computed tomography (CT), magnetic resonance imaging (MRI), and echocardiography. While this review primarily focuses on widely used and clinically accessible biomarkers and imaging techniques, it also discusses alternative biomarkers proposed for diagnostic use. Although CT remains the gold standard for diagnosis, biomarkers facilitate rapid risk stratification, complementing imaging techniques. Emerging technologies, such as metabolomics, are reshaping diagnostic algorithms. Despite advances in diagnostic methods, challenges such as misdiagnosis and missed diagnoses persist. Ongoing research into novel biomarkers and innovative imaging techniques holds promise for improving diagnostic accuracy and patient outcomes.

Obesity has emerged as a global epidemic with far-reaching health complications, including its role as an independent risk factor for chronic kidney disease (CKD). Increasing evidence suggests that obesity contributes to CKD through multiple mechanisms, including chronic inflammation, hemodynamic alterations, insulin resistance, and lipid accumulation. These processes can culminate in histopathological changes collectively referred to as obesity-related glomerulopathy (ORG). This review aims to provide a comprehensive overview of the current knowledge regarding the prevalence, clinical manifestations, and pathophysiology of ORG. Furthermore, we emphasize the importance of identifying key biomarkers that facilitate the early detection of ORG. Finally, we explore emerging therapeutic strategies that offer promise in mitigating this growing global health crisis.

Background/Objectives: This study presents a novel approach, based on a combination of radiomic feature extraction, data resampling techniques, and machine learning algorithms, for the detection of degraded bone structures in Dual X-ray Absorptiometry (DXA) images. This comprehensive approach, which addresses the critical aspects of the problem, distinguishes this work from previous studies, improving the performance achieved by the most similar studies. The primary aim is to provide clinicians with an accessible tool for quality bone assessment, which is currently limited. Methods: A dataset of 1531 spine DXA images was automatically segmented and labelled based on Trabecular Bone Score (TBS) values. Radiomic features were extracted using Pyradiomics, and various resampling techniques were employed to address class imbalance. Three machine learning classifiers (Logistic Regression, Support Vector Machine (SVM), and XGBoost) were trained and evaluated using standard performance metrics. Results: The SVM classifier outperformed the other classifiers. The highest F-score of 97.5% was achieved using the Grey Level Dependence Matrix and Grey Level Run Length Matrix feature combination with SMOTEENN resampling, which proved to be the most effective resampling technique, while the undersampling method yielded the lowest performance. Conclusions: This research demonstrates the potential of radiomic texture features, resampling techniques, and machine learning methods for classifying DXA images into healthy or degraded bone structures, which potentially leads to improved clinical diagnosis and treatment.

Background and Clinical Significance: Spiradenoma is a rare benign skin adnexal tumor with unknown incidence and prevalence, typically affecting young to middle-aged adults without a sexual predilection. Case Presentation: A 59-year-old woman presented with a palpable lesion in the suprapubic region that had been there for 20 years and had become enlarged over the past 2 months. Physical examination revealed a firm, non-tender, subcutaneous mass, approximately 2 cm in size, in the right pubic region. Ultrasound revealed a hypoechoic, heterogeneous lesion with a well-defined border, measuring 2.37 × 0.94 × 1.67 cm, without hypervascularity. Therefore, the patient underwent excision of the subcutaneous tumor. The pathology report confirmed the diagnosis of spiradenoma of the pubis. Histochemistry showed that the inner luminal cells were positive for CK7, and the outer basaloid cells were positive for p63. CD56 and CD117 were focally positive. Conclusions: With an accurate diagnosis and appropriate surgical excision, the prognosis for spiradenoma is generally excellent. However, a long-term follow-up is advisable.

Background: This study aims to evaluate the impact of various weighting factors (WFs) on the quality of weighted average (WA) dual-energy computed tomography (DECT) non-contrast brain images and to determine the optimal WF value. Because they simulate standard CT images, 0.4-WA reconstructions are routinely used. Methods: In the initial phase of the research, quantitative and qualitative analyses of WA DECT images of an anthropomorphic head phantom, utilizing WFs ranging from 0 to 1 in 0.1 increments, were conducted. Based on the phantom study findings, WFs of 0.4, 0.6, and 0.8 were chosen for patient analyses, which were identically carried out on 85 patients who underwent non-contrast head DECT. Three radiologists performed subjective phantom and patient analyses. Results: Quantitative phantom image analysis revealed the best gray-to-white matter contrast-to-noise ratio (CNR) at the highest WFs and minimal noise artifacts at the lowest WF values. However, the WA reconstructions were deemed non-diagnostic by all three readers. Two readers found 0.6-WA patient reconstructions significantly superior to 0.4-WA images (p < 0.001), while reader 1 found them to be equally good (p = 0.871). All readers agreed that 0.8-WA images exhibited the lowest image quality. Conclusions: In conclusion, 0.6-WA reconstructions demonstrated superior image quality over 0.4-WA and are recommended for routine non-contrast brain DECT.