Cardiovascular disease has a worldwide impact on quality of life impairment, and is the cause of an estimated 17.9 million deaths each year. A major subset of cardiovascular disease cases are composed of structural heart diseases, of which mitral regurgitation is a very common valvular disorder, affecting at least 10% of the general population. The gold standard approach to the treatment of mitral regurgitation is mitral valve clipping, an edge-to-edge technique where the surgeon sutures together the valve leaflets, performed percutaneously through the MitraClip delivery system. The procedure, however, is technically and mentally demanding for the operator, and its success rate and effectiveness is highly operator-dependant; in spite of that, by implementing in the surgical setup autonomous navigation and target tracking of the delivery catheter in the intracardiac space, accuracy, repeatability and safety of the procedure can be significantly improved. This work proposes an experimental setup for dynamic multi-sensor target tracking, as well as a sensor fusion algorithm based on the Kalman filter for multi-sensor measurement systems. The ARTERY project contemplates a probe for 3D transesophageal echocardiography and an Aurora electromagnetic tracking system in the presence of metallic distortion, while this work substitues the probe with a stereoscopic camera. The objective of these sensors is to track a target throughout a trajectory modeled after the MitraClip delivery catheter’s iter during a mitral valve repair procedure. The proposed solution achieves an accuracy of 3.1 mm RMSE, with a 64.3% RMSE reduction from the most accurate sensor employed in the setup.
Le malattie cardiovascolari hanno un impatto di scala mondiale sulla compromissione della qualità della vita, e sono la causa di circa 17,9 milioni di decessi ogni anno. Un importante sottogruppo dei casi di malattie cardiovascolari è costituito da malattie cardiache strutturali, tra cui il rigurgito mitralico è un disturbo valvolare molto comune che colpisce almeno il 10% della popolazione. L’approccio gold standard al trattamento del rigurgito mitralico è il clipping della valvola mitrale, una tecnica edge-to-edge in cui il chirurgo sutura insieme i lembi valvolari, eseguita per via percutanea attraverso il sistema MitraClip. La procedura, tuttavia, è tecnicamente e mentalmente impegnativa per l’operatore, e il suo tasso di successo ed efficacia dipende fortemente dall’operatore; introducendo però nel setup chirurgico navigazione e tracciamento autonomo del catetere di posizionamento nello spazio intracardiaco, è possibile migliorare notevolmente accuratezza, sicurezza e ripetibilità dell’operazione. Questo lavoro propone un apparato sperimentale per il tracciamento dinamico multisensore di target, nonché un algoritmo di sensor fusion basato sul filtro di Kalman per sistemi di misurazione multisensore. Il progetto ARTERY prevede una sonda per ecografia transesofagea 3D e un sistema di tracciamento elettromagnetico Aurora in presenza di distorsione metallica, mentre in questo lavoro la sonda è sostituita da una telecamera stereoscopica. La funzione di questi sensori è quella di tracciare un target attraverso una traiettoria modellata sull’iter del catetere di rilascio MitraClip durante una procedura di riparazione della valvola mitrale. La soluzione proposta raggiunge una precisione di 3,1 mm RMSE, con una riduzione del 64,3% RMSE rispetto al sensore più accurato utilizzato nel setup.
Experimental setup for multi-sensor target tracking with sensor fusion algorithm
Spinosi, Luca
2022/2023
Abstract
Cardiovascular disease has a worldwide impact on quality of life impairment, and is the cause of an estimated 17.9 million deaths each year. A major subset of cardiovascular disease cases are composed of structural heart diseases, of which mitral regurgitation is a very common valvular disorder, affecting at least 10% of the general population. The gold standard approach to the treatment of mitral regurgitation is mitral valve clipping, an edge-to-edge technique where the surgeon sutures together the valve leaflets, performed percutaneously through the MitraClip delivery system. The procedure, however, is technically and mentally demanding for the operator, and its success rate and effectiveness is highly operator-dependant; in spite of that, by implementing in the surgical setup autonomous navigation and target tracking of the delivery catheter in the intracardiac space, accuracy, repeatability and safety of the procedure can be significantly improved. This work proposes an experimental setup for dynamic multi-sensor target tracking, as well as a sensor fusion algorithm based on the Kalman filter for multi-sensor measurement systems. The ARTERY project contemplates a probe for 3D transesophageal echocardiography and an Aurora electromagnetic tracking system in the presence of metallic distortion, while this work substitues the probe with a stereoscopic camera. The objective of these sensors is to track a target throughout a trajectory modeled after the MitraClip delivery catheter’s iter during a mitral valve repair procedure. The proposed solution achieves an accuracy of 3.1 mm RMSE, with a 64.3% RMSE reduction from the most accurate sensor employed in the setup.File | Dimensione | Formato | |
---|---|---|---|
2024_04_Spinosi_Tesi_01.pdf
accessibile in internet per tutti
Descrizione: Elaborato di tesi
Dimensione
19.99 MB
Formato
Adobe PDF
|
19.99 MB | Adobe PDF | Visualizza/Apri |
2024_04_Spinosi_Executive Summary_02.pdf
accessibile in internet per tutti
Descrizione: Executive Summary
Dimensione
1.62 MB
Formato
Adobe PDF
|
1.62 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/219495