1,720,962 research outputs found
Automatic 4D mitral valve segmentation from transesophageal echocardiography: a semi-supervised learning approach
Full text link: Performing automatic and standardized 4D TEE segmentation and mitral valve analysis is challenging due to the limitations of echocardiography and the scarcity of manually annotated 4D images. This work proposes a semi-supervised training strategy using pseudo labelling for MV segmentation in 4D TEE; it employs a Teacher-Student framework to ensure reliable pseudo-label generation. 120 4D TEE recordings from 60 candidates for MV repair are used. The Teacher model, an ensemble of three convolutional neural networks, is trained on end-systole and end-diastole frames and is used to generate MV pseudo-segmentations on intermediate frames of the cardiac cycle. The pseudo-annotated frames augment the Student model's training set, improving segmentation accuracy and temporal consistency. The Student outperforms individual Teachers, achieving a Dice score of 0.82, an average surface distance of 0.37 mm, and a 95% Hausdorff distance of 1.72 mm for MV leaflets. The Student model demonstrates reliable frame-by-frame MV segmentation, accurately capturing leaflet morphology and dynamics throughout the cardiac cycle, with a significant reduction in inference time compared to the ensemble. This approach greatly reduces manual annotation workload and ensures reliable, repeatable, and time-efficient MV analysis. Our method holds strong potential to enhance the precision and efficiency of MV diagnostics and treatment planning in clinical settings
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Blood flow in cerebral aneurysms : comparison among 4D flow guided-CDF, FSI and 3D variational data assimilation
Full text linkLAUREA MAGISTRALELa rottura degli aneurismi intracranici è la principale causa di emorragia subaracnoidea, un’emergenza neurovascolare con elevata mortalità. Un’accurata analisi emodinamica può migliorare la valutazione del rischio di rottura degli aneurismi intracranici e la pianificazione del loro trattamento. In questo lavoro di tesi, viene proposto un approccio innovativo che combina la fluidodinamica computazionale e la risonanza magnetica 4D flow.
Un algoritmo di data assimilation (3DVar) è stato implementato, andando poi a valutare la capacità di quest’ultimo nel ricostruire il flusso sanguigno in un modello tridimensionale di un aneurisma cerebrale a partire dai dati forniti dalla risonanza magnetica 4D flow. Sullo stesso modello sono state eseguite una simulazione fluidodinamica a pareti rigide (CFD) e una a pareti deformabili (FSI), imponendo le misure di velocità fornite dalla risonanza magnetica 4D flow come condizioni al contorno.
L’algoritmo di 3DVar si è dimostrato efficiente nel ricostruire il flusso sanguigno all’interno dell’aneurisma risultando più accurato della simulazione CFD.
L’algoritmo qui implementato ha permesso quindi, di migliorare l’accuratezza della CFD nell’analisi emodinamica all’interno dell’aneurisma cerebrale sfruttando interamente le misure di velocità fornite dalla risonanza magnetica 4D flow, con un importante risvolto nel migliorare la valutazione del rischio di rottura degli aneurismi intracranici.Intracranial aneurysm (IA) rupture is the most common cause of subarachnoid hemorrhage (SAH), a life-threatening neurovascular emergency with significant mortality. Accurate hemodynamics analysis may improve IA risk assessment and treatment-planning. In this work, a novel approach combining computational fluid dynamics (CFD) and 4D flow magnetic resonance imaging (MRI) is proposed.
A 3D variational data assimilation (3DVar) approach was implemented and evaluated in reconstructing blood flow in an IA model from 4D flow measurements. Together 4D flow guided-CFD and FSI simulations were carried out on the same model.
The 3DVar approach proved to be efficient in reconstructing blood flow within the IA and more accurate than the CFD simulation.
The 3DVar approach allowed therefore to improve the CFD accuracy in the hemodynamic analysis within the IA exploiting entirely the 4D flow measurements, with an important implication in improving the IA risk assessment
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
Deep learning-driven segmentation of echocardiographic images for intraprocedural support in percutaneous mitral valve repair
No full textDOTTORATOIl rigurgito mitralico (RM) è una condizione patologica diffusa che spesso richiede un intervento chirugico. Sebbene la chirurgia tradizionale a torace aperto sia stata il trattamento principale per la RM, le procedure percutanee minimamente invasive, come la riparazione transcatetere edge-to-edge (TEER), sono emerse come valide alternative, in particolare per i pazienti non idonei alla chirurgia a causa dell'età o di comorbidità. Nonostante i loro vantaggi, le procedure percutanee sono altamente complesse, richiedendo una guida per immagini precisa, un ampio addestramento e una notevole esperienza dell'operatore. L'ecocardiografia, in particolare l'ecocardiografia transesofagea (TEE) in modalità 2D e 3D, svolge un ruolo cruciale in questi interventi. Tuttavia, le sfide intrinseche dell'imaging ecocardiografico, tra cui lo scarso contrasto, la dipendenza dall'operatore e la suscettibilità agli artefatti, rappresentano ostacoli significativi per una guida procedurale efficace.\\
Questa tesi esplora il potenziale dell'intelligenza artificiale (IA) e del deep learning (DL) per migliorare l'imaging ecocardiografico nel trattamento percutaneo della valvola mitrale (VM). L'obiettivo principale di questo lavoro è lo sviluppo di metodi basati su DL per migliorare la segmentazione, l'interpretazione e l'analisi delle immagini ecocardiografiche, facilitando la guida procedurale e il processo decisionale nella cardiologia interventistica. I principali contributi di questa tesi includono: (1) un nuovo sistema basato su una rete neurale convoluzionale (CNN) per la segmentazione completamente automatizzata e multi-classe dell'anulus mitralico e dei lembi valvolari a partire da immagini TEE 3D, ottenendo un'elevata accuratezza e ripetibilità della segmentazione e consentendo l'estrazione e la quantificazione automatica delle caratteristiche anatomiche; (2) un framework di apprendimento semi-supervisionato che utilizza un approccio Teacher-Student per la segmentazione dinamica 4D della VM, riducendo significativamente la necessità di annotazioni manuali e migliorando l'accuratezza della segmentazione nel ciclo cardiaco; (3) una strategia di adattamento auto-supervisionato per consentire una segmentazione della VM indipendente dal fornitore, affrontando la variabilità tra diverse piattaforme ecocardiografiche; e (4) un sistema automatizzato per il rilevamento e l'analisi della configurazione del MitraClip, migliorando la visualizzazione in tempo reale e l'efficienza procedurale.\\
I metodi proposti hanno dimostrato prestazioni eccellenti nella segmentazione della VM, nel tracciamento dinamico e nel rilevamento dei dispositivi, fornendo miglioramenti clinicamente significativi in termini di accuratezza, efficienza e sicurezza della TEER. Questa ricerca rappresenta un passo importante verso l'automazione degli interventi strutturali cardiaci percutanei, aprendo la strada a piattaforme robotiche assistite dall'IA che si basano sull'ecocardiografia per la navigazione procedurale. Riducendo la dipendenza dalla fluoroscopia, migliorando l'efficienza degli operatori e ottimizzando gli esiti procedurali, i progressi presentati in questa tesi contribuiscono all'obiettivo più ampio di ampliare l'accesso ai trattamenti percutanei salvavita per le malattie della valvola mitrale.Mitral regurgitation (MR) is a prevalent condition that often necessitates intervention. While traditional open-chest surgery has been the primary treatment for MR, minimally invasive percutaneous procedures such as transcatheter edge-to-edge repair (TEER) have emerged as viable alternatives, particularly for patients ineligible for surgery due to age or comorbidities. Despite their advantages, percutaneous procedures require precise imaging guidance, extensive training, and significant operator expertise, making them highly demanding interventions. Echocardiography, particularly transesophageal echocardiography (TEE) in both 2D and 3D modes, plays a crucial role in these interventions. However, the inherent challenges of echocardiographic imaging, including low contrast, operator dependency, and susceptibility to artifacts, pose significant barriers to effective procedural guidance.\\
This thesis explores the potential of artificial intelligence (AI) and deep learning (DL) to enhance echocardiographic imaging for percutaneous mitral valve (MV) treatment. The primary focus of this work is the development of DL-based methods to improve the segmentation, interpretation, and analysis of echocardiographic images, enhancing procedural guidance and decision-making in interventional cardiology. Key contributions of this thesis include: (1) a novel convolutional neural network (CNN)-based pipeline for fully automated, multi-class segmentation of the MV annulus and leaflets from 3D TEE, achieving high segmentation accuracy and repeatability, and enabling automatic anatomical feature extraction and quantification; (2) a semi-supervised learning framework utilizing a Teacher-Student approach for dynamic 4D MV segmentation, significantly reducing the need for manual annotations while improving segmentation accuracy across the cardiac cycle; (3) a self-supervised domain adaptation strategy to enable vendor-agnostic MV segmentation, addressing cross-domain variability in echocardiographic imaging; and (4) an automated pipeline for MitraClip detection and configuration analysis, improving real-time visualization and procedural efficiency. \\
The proposed methods have demonstrated outstanding performance in MV segmentation, dynamic tracking, and device detection, providing clinically significant enhancements in TEER accuracy, efficiency, and safety. This research represents a significant step toward the automation of percutaneous structural heart interventions, paving the way for AI-assisted robotic platforms that rely on echocardiography for procedural navigation. By reducing reliance on fluoroscopy, enhancing operator efficiency, and improving procedural outcomes, the advancements presented in this thesis contribute to the broader goal of expanding access to life-saving percutaneous treatments for MV diseases.DIPARTIMENTO DI ELETTRONICA, INFORMAZIONE E BIOINGEGNERIA36REDAELLI, ALBERTO CESARE LUIGIDELLACA', RAFFAEL
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
