Publications by authors named "Canu S"

Sarcoidosis is a rare, systemic inflammatory disease whose diagnosis and management can pose a challenge for clinicians and specialists. Scientific knowledge on the molecular pathways that drive its development is still lacking, with no standardized therapies available and insufficient strategies to predict patient outcome. In recent years, oxidative stress has been highlighted as an important factor in the pathogenesis of sarcoidosis, involving several enzymes and molecules in the mechanism of the disease.

View Article and Find Full Text PDF

Background: Haemoptysis is a challenging symptom that can be associated with potentially life-threatening medical conditions. Follow-up is key in these patients to promptly detect new or misdiagnosed pathologic findings. Few prospective studies have evaluated long-term prognostic outcomes in patients with haemoptysis.

View Article and Find Full Text PDF

Magnetic Resonance Imaging (MRI) is a widely used imaging technique to assess brain tumor. Accurately segmenting brain tumor from MR images is the key to clinical diagnostics and treatment planning. In addition, multi-modal MR images can provide complementary information for accurate brain tumor segmentation.

View Article and Find Full Text PDF
Article Synopsis
  • The COVID-19 pandemic significantly impacted global public health, highlighting the need for effective diagnosis tools like Computed Tomography (CT).
  • A U-Net based segmentation network with an attention mechanism is proposed to improve the identification of COVID-19 in CT images by enhancing important features for better accuracy.
  • The method shows promising results, achieving rapid segmentation in just 0.29 seconds per CT slice, with notable performance metrics including an 83.1% Dice Score.
View Article and Find Full Text PDF
Article Synopsis
  • The paper introduces a deep learning-based 3D brain tumor segmentation network that utilizes multi-sequence MRI datasets in a three-stage process.
  • The first stage involves using a 3D U-Net to generate context constraints for tumor regions, followed by fusing multi-sequence MRIs with an attention mechanism for individual tumor segmentation.
  • A new loss function addresses multiple class segmentation, and a second 3D U-Net refines the predictions, achieving promising results in metrics like dice score and hausdorff distance on the BraTS 2017 dataset.
View Article and Find Full Text PDF

Background: Objective data on chemosensitive disorders during COVID-19 are lacking in the Literature.

Methods: Multicenter cohort study that involved four Italian hospitals. Three hundred and forty-five COVID-19 patients underwent objective chemosensitive evaluation.

View Article and Find Full Text PDF

Structuring raw medical documents with ontology mapping is now the next step for medical intelligence. Deep learning models take as input mathematically embedded information, such as encoded texts. To do so, word embedding methods can represent every word from a text as a fixed-length vector.

View Article and Find Full Text PDF

Background: Word embedding technologies, a set of language modeling and feature learning techniques in natural language processing (NLP), are now used in a wide range of applications. However, no formal evaluation and comparison have been made on the ability of each of the 3 current most famous unsupervised implementations (Word2Vec, GloVe, and FastText) to keep track of the semantic similarities existing between words, when trained on the same dataset.

Objective: The aim of this study was to compare embedding methods trained on a corpus of French health-related documents produced in a professional context.

View Article and Find Full Text PDF

Background: In fibrotic interstitial lung diseases, exertional breathlessness is strongly linked to health-related quality of life (HRQOL). Breathlessness is often associated with oxygen desaturation, but few data about the use of ambulatory oxygen in patients with fibrotic interstitial lung disease are available. We aimed to assess the effects of ambulatory oxygen on HRQOL in patients with interstitial lung disease with isolated exertional hypoxia.

View Article and Find Full Text PDF

Bluetongue (BT) is a vector-borne disease transmitted by species of Culicoides midges (Diptera: Ceratopogonidae). Many studies have contributed to clarifying various aspects of its aetiology, epidemiology and vector dynamic; however, BT remains a disease of epidemiological and economic importance that affects ruminants worldwide. Since 2000, the Sardinia region has been the most affected area of the Mediterranean basin.

View Article and Find Full Text PDF

Desmoplastic small round cell tumor of the pleura is a rare malignancy, with only a few cases reported in the scientific literature. The aim of the present review is to discuss the demographic, pathological, clinical, and therapeutic features of this rare tumor. English-language articles published since 1989, when the first case of desmoplastic small round cell tumor of the pleura was described, were retrieved, and fifteen cases included in fourteen articles were revised.

View Article and Find Full Text PDF

Introduction: Bronchial localization of Mucoepidermoid carcinoma (MEC) is rare. The precise nature of these neoplasms is not yet clear and little is known on the histogenesis and pathogenesis of the disease. Here we present a case of a bronchial MEC with a detailed pathological, immunohistochemical, and molecular analysis.

View Article and Find Full Text PDF

Background: Tracheobronchopathia Osteochondroplastica (TBPO) is an uncommon and benign condition characterized by osseous or metaplastic cartilaginous 1-3 mm nodules in the submucosa of the tracheo-bronchial tree. Posterior membranous wall of trachea is typically spared. Ecchondrosis and exostosis nodules can cause chronic inflammation and mucosal metaplasia, stiffness and airway obstruction.

View Article and Find Full Text PDF
Learning SVM in Kreĭn Spaces.

IEEE Trans Pattern Anal Mach Intell

June 2016

This paper presents a theoretical foundation for an SVM solver in Kreĭn spaces. Up to now, all methods are based either on the matrix correction, or on non-convex minimization, or on feature-space embedding. Here we justify and evaluate a solution that uses the original (indefinite) similarity measure, in the original Kreĭn space.

View Article and Find Full Text PDF
Article Synopsis
  • Building an accurate training database in supervised classification is challenging due to uncertainty in data, especially in medical imaging where radiologists may lack histological ground truth.
  • The paper introduces a new method called probabilistic SVM (P-SVM) that combines both class labels and probability estimates to improve classification accuracy and probability predictions in uncertain data scenarios.
  • P-SVM outperforms traditional SVM and fuzzy-SVM in tests, demonstrating effective application for computer-aided decision systems in diagnosing prostate cancer using multiparametric MR imaging.
View Article and Find Full Text PDF

Tuberculosis remains the major cause of morbidity and mortality by a single infectious agent, particularly in developing countries. In recent years, we have witnessed the emergence of uncommon radiographic patterns of chest tuberculosis. Lymphadenitis is the most common extrapulmonary tuberculosis (TB) manifestation which, in developed countries, occurs more frequently in childhood, but also among adult immigrants from endemic countries and in HIV-infected people.

View Article and Find Full Text PDF

Recently, there has been much interest around multitask learning (MTL) problem with the constraints that tasks should share a common sparsity profile. Such a problem can be addressed through a regularization framework where the regularizer induces a joint-sparsity pattern between task decision functions. We follow this principled framework and focus on l(p)-l(q) (with 0 ≤ p ≤ 1 and 1 ≤ q ≤ 2) mixed norms as sparsity-inducing penalties.

View Article and Find Full Text PDF

To investigate whether there was a subpopulation of repeat breeders (cows or heifers that returned to oestrus after three inseminations) that were less fertile after a fourth artificial insemination (AI) with or without additional embryo transfer, and to estimate the efficacy of AI plus embryo transfer to overcome repeat breeding problems, a two-part investigation was carried out. Part 1 involved 85 repeat breeders and 85 controls subjected to AI alone. In part 2, 128 repeat breeders received AI on day 0 plus an embryo transfer seven days later, while controls received embryo transfer alone on day 7.

View Article and Find Full Text PDF

Evidence from both animal and human studies suggests a role for dopamine in the therapeutic effect of antidepressant drugs. Consistently, dopamine receptor antagonists antagonize the effect of antidepressant drugs in different experimental models of depression. Neurosteroids, and in particular allopregnanolone, seem to be involved both in the pathophysiology of depression and in the mechanism of action of antidepressant drugs, and their role seems to be particularly important in the understanding of mood disturbances related to the different phases of the reproductive life in women.

View Article and Find Full Text PDF

Sexual dimorphism, detectable in vascular smooth muscle cells freshly isolated from aorta of male and female rats, is associated with a different susceptibility to radiation-induced apoptosis. In this work we investigated the mechanism underlying this difference and discovered that, in comparison with cells from male rats, cells from female rats show adhesion-associated resistance to apoptosis, the so called anoikis resistance. This is apparently due to a more adhering phenotype, characterized by a well organized actin microfilament cytoskeleton and to an increased phosphorylated focal adhesion kinase, and, more importantly, to a higher propensity to undergo survival by autophagy.

View Article and Find Full Text PDF

Cell death processes have been widely investigated in recent years in order to elucidate the different pathways involved in the complex machinery implicated in determining cell fate. Different forms of cell death have been described: Apart from the classical form of death known as necrosis, a well-characterized traumatic injury of the cell, several additional forms of cell death have been identified. Of these, apoptosis has been characterized in the greatest detail.

View Article and Find Full Text PDF

Vascular smooth muscle cells (VSMC) have been isolated from male and female rat aorta and studied to assess their susceptibility to ultraviolet radiation-induced oxidative stress. Interestingly, a gender difference, in terms of reactive oxygen species production, was detected in both basal and irradiated VSMC. Namely, VSMC from male rats were more susceptible to radiation-induced stress and easier underwent apoptosis in comparison to cells from female rats.

View Article and Find Full Text PDF