The planning of Transcatheter Pulmonary Valve Implantation (TPVI) is a complex task due to the highly variable anatomy and dynamic behavior of the Right Ventricular Outflow Tract (RVOT), particularly in patients with congenital heart diseases such as Tetralogy of Fallot (ToF). Current commercial tools rely on static imaging, failing to capture time-dependent RVOT deformations, which are crucial for procedural planning. Additionally, existing dynamic tracking methods focus on selected cross-sections rather than the 3D structure, limiting the accuracy of motion analysis. To address these limitations, a fully automated Four-Dimensional Computed Tomography (4D CT) workflow was developed. This workflow enables frame-by-frame registration of the right ventricular outflow tract (RVOT) throughout the cardiac cycle, allowing for a comprehensive biomechanical evaluation of the RVOT. The automated workflow included the following steps: i) segmentation of the RVOT, ii) creation of a RVOT mesh for each phase of the 4D CT sequence, iii) non-rigid registration of the meshes and three-dimensional tracking of the relevant geometric features, and iv) calculation of the local deformations of the RVOT throughout the cardiac cycle. The workflow was tested on a dataset of 34 patients, including 19 TPVI candidates, 11 Surgical Pulmonary Valve Replacement (SPVR) candidates, and 4 on a waiting list. Validation was conducted on a subset of 10 patients, assessing segmentation accuracy, registration performance, geometric parameters, and strain calculation. Cross-sectional area measurements agree with a semi-automatic reference, and registration errors remained within clinically acceptable limits. The results confirm that 4D CT-based tracking significantly improves RVOT evaluation compared to traditional approaches. By combining 4D imaging with morphometric and strain analysis, this study introduces a novel computational tool for TPVI planning. The proposed workflow enhances patient-specific RVOT evaluation, ultimately improving procedural outcomes and broadening its potential clinical applications beyond TPVI.
La pianificazione dell’impianto transcatetere della valvola polmonare (TPVI) è un processo complesso a causa della notevole variabilità anatomica e del comportamento dinamico del tratto di efflusso del ventricolo destro (RVOT), soprattutto nei pazienti con cardiopatie congenite come la tetralogia di Fallot (ToF). Gli strumenti attualmente disponibili si basano su immagini statiche, senza riuscire a cogliere le deformazioni dinamiche dell’RVOT, fondamentali per una corretta pianificazione dell’intervento. Inoltre, i metodi di tracking dinamico esistenti si concentrano su sezioni specifiche, senza ricostruire la deformazione volumetrica completa, limitando l’accuratezza dell’analisi del movimento. Per superare queste limitazioni, è stato sviluppato un metodo completamente automatizzato basato su immagini 4D di Tomografia Computerizzata (4D CT), che consente la registrazione dinamica dell’RVOT lungo l’intero ciclo cardiaco. Il processo integra segmentazione, estrazione di superfici triangolate, tecniche di registrazione non rigida, calcolo delle deformazioni e analisi dei parametri geometrici, garantendo una caratterizzazione più completa della biomeccanica dell’RVOT. Il metodo è stato testato su un dataset di 34 pazienti, costituito da 19 candidati TPVI, 11 candidati alla sostituzione chirurgica della valvola polmonare (SPVR) e 4 in lista d’attesa. La validazione è stata condotta su un sottoinsieme di 10 pazienti, valutando l’accuratezza della segmentazione, le prestazioni della registrazione, il calcolo dei parametri geometrici e delle deformazioni. Le misurazioni delle aree trasversali hanno mostrato un’elevata concordanza con un riferimento semi-automatico, mentre gli errori di registrazione sono rimasti entro limiti clinicamente accettabili. I risultati confermano che l’analisi basata su immagini 4D CT migliora significativamente la valutazione dell’RVOT rispetto agli approcci tradizionali. L’integrazione dell’imaging 4D con l’analisi morfometrica e delle deformazioni introduce un nuovo strumento computazionale per supportare la pianificazione della TPVI. Il metodo proposto consente una valutazione più accurata e personalizzata dell’RVOT, migliorando gli esiti procedurali e ampliando le sue potenziali applicazioni cliniche oltre la TPVI.
A 4D-CT based workflow for the 3D quantification of RVOT time-dependent anatomy and TPVI planning
ODINELLI, ELENA;MARIANI, FRANCESCA
2023/2024
Abstract
The planning of Transcatheter Pulmonary Valve Implantation (TPVI) is a complex task due to the highly variable anatomy and dynamic behavior of the Right Ventricular Outflow Tract (RVOT), particularly in patients with congenital heart diseases such as Tetralogy of Fallot (ToF). Current commercial tools rely on static imaging, failing to capture time-dependent RVOT deformations, which are crucial for procedural planning. Additionally, existing dynamic tracking methods focus on selected cross-sections rather than the 3D structure, limiting the accuracy of motion analysis. To address these limitations, a fully automated Four-Dimensional Computed Tomography (4D CT) workflow was developed. This workflow enables frame-by-frame registration of the right ventricular outflow tract (RVOT) throughout the cardiac cycle, allowing for a comprehensive biomechanical evaluation of the RVOT. The automated workflow included the following steps: i) segmentation of the RVOT, ii) creation of a RVOT mesh for each phase of the 4D CT sequence, iii) non-rigid registration of the meshes and three-dimensional tracking of the relevant geometric features, and iv) calculation of the local deformations of the RVOT throughout the cardiac cycle. The workflow was tested on a dataset of 34 patients, including 19 TPVI candidates, 11 Surgical Pulmonary Valve Replacement (SPVR) candidates, and 4 on a waiting list. Validation was conducted on a subset of 10 patients, assessing segmentation accuracy, registration performance, geometric parameters, and strain calculation. Cross-sectional area measurements agree with a semi-automatic reference, and registration errors remained within clinically acceptable limits. The results confirm that 4D CT-based tracking significantly improves RVOT evaluation compared to traditional approaches. By combining 4D imaging with morphometric and strain analysis, this study introduces a novel computational tool for TPVI planning. The proposed workflow enhances patient-specific RVOT evaluation, ultimately improving procedural outcomes and broadening its potential clinical applications beyond TPVI.File | Dimensione | Formato | |
---|---|---|---|
2025_04_MARIANI_ODINELLI_TESI.pdf
accessibile in internet per tutti
Dimensione
30.66 MB
Formato
Adobe PDF
|
30.66 MB | Adobe PDF | Visualizza/Apri |
2025_04_MARIANI_ODINELLI_EXECUTIVE_SUMMARY.pdf
accessibile in internet per tutti
Dimensione
774.36 kB
Formato
Adobe PDF
|
774.36 kB | 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/10589/235715