The high incidence of arrhythmic events during the first minutes of Acute Myocardial Ischemia (AMI) represents a severe clinical risk. Following coronary occlusion, abrupt metabolic alterations, such as hyperkalemia, hypoxia, and acidosis, convert the myocardium into a highly arrhythmogenic substrate, promoting the initiation and maintenance of reentry circuits. For patient-specific risk assessment, computational models represent a powerful tool; however, relying on simplified geometries or chronic scars has previously limited their predictive power in the acute phase. To overcome these limitations, the present study uses Myocardial Blood Flow (MBF) maps obtained during stress-CTP, where the induced hyperemic condition allows for a realistic reproduction of the patient-specific ischemia. On this basis, arrhythmic vulnerability was evaluated following the delivery of an ectopic beat on the epicardial surface, integrating the computational framework developed by Corda et al. (2025). By analyzing a dataset of 6 real anatomies and 14 derived virtual scenarios, arrhythmogenesis was quantified through the In silico Risk Index (IRI). This work pursues two main objectives. First, to evaluate the impact of transmural differentiation: by implementing an innovative preprocessing algorithm, the physiological electrical heterogeneity of the three ventricular layers (epicardium, mid-myocardium, and endocardium) was introduced into the model. Second, to develop a rapid predictive model based purely on the morphological parameters of the ischemic lesion. The extent of the ischemic zone and its spatial distribution, quantified through an original Asymmetry Index (AI), were identified as the key geometric determinants of arrhythmogenesis. Using the simulation-derived IRI as the ground-truth, a Risk Calibrated Index (RCI) was then calibrated. By combining these simple geometric parameters, the RCI allows for a direct estimation of arrhythmic susceptibility, providing a potential clinical risk stratification tool that bypasses the need for complex electrophysiological simulations.
L'elevata incidenza di eventi aritmici nei primi minuti di un'ischemia miocardica acuta rappresenta un grave fattore di rischio clinico. A seguito dell'occlusione coronarica, le rapide alterazioni metaboliche, quali iperkalemia, ipossia e acidosi, convertono il miocardio in un substrato aritmogenico, favorendo l'innesco e il mantenimento di circuiti di rientro. Per la valutazione paziente-specifica del rischio, i modelli computazionali offrono grandi potenzialità; tuttavia, il frequente impiego di geometrie semplificate o derivate da cicatrici croniche ne ha finora limitato il potere predittivo in fase acuta. Per superare tali criticità, il presente studio sfrutta mappe di perfusione (MBF) ottenute da tomografie stress-CTP, la cui condizione iperemica permette di riprodurre fedelmente la specifica estensione dell'ischemia individuale. Su questa base, è stata valutata la vulnerabilità aritmica a seguito di uno stimolo ectopico epicardico, integrando il framework computazionale di Corda et al. (2025). Analizzando sei anatomie reali e 14 scenari virtuali derivati, l'aritmogenesi è stata quantificata tramite un indice di rischio in silico (IRI). Il lavoro persegue due obiettivi principali. In primo luogo, valutare l'impatto della differenziazione transmurale: implementando un innovativo algoritmo di preprocessing, è stata introdotta nel modello l'eterogeneità elettrica dei tre strati della parete cardiaca (epicardio, mesomiocardio ed endocardio). In secondo luogo, sviluppare un modello predittivo basato sui parametri morfologici della lesione. L'estensione e la distribuzione spaziale della zona ischemica, quest'ultima quantificata tramite un originale indice di asimmetria (AI), sono state identificate come determinanti chiave dell'aritmogenesi. Utilizzando l'IRI come rischio aritmogenico di riferimento, è stato calibrato un indice di rischio predittivo (RCI). Combinando questi semplici parametri geometrici, l'RCI permette una stima diretta della suscettibilità aritmica, fornendo un potenziale strumento di stratificazione del rischio clinico che prescinde dall'esecuzione di complesse simulazioni elettrofisiologiche.
Risk prediction modeling of arrhythmogenesis in acute cardiac ischemic conditions
Ubiali, Caterina
2025/2026
Abstract
The high incidence of arrhythmic events during the first minutes of Acute Myocardial Ischemia (AMI) represents a severe clinical risk. Following coronary occlusion, abrupt metabolic alterations, such as hyperkalemia, hypoxia, and acidosis, convert the myocardium into a highly arrhythmogenic substrate, promoting the initiation and maintenance of reentry circuits. For patient-specific risk assessment, computational models represent a powerful tool; however, relying on simplified geometries or chronic scars has previously limited their predictive power in the acute phase. To overcome these limitations, the present study uses Myocardial Blood Flow (MBF) maps obtained during stress-CTP, where the induced hyperemic condition allows for a realistic reproduction of the patient-specific ischemia. On this basis, arrhythmic vulnerability was evaluated following the delivery of an ectopic beat on the epicardial surface, integrating the computational framework developed by Corda et al. (2025). By analyzing a dataset of 6 real anatomies and 14 derived virtual scenarios, arrhythmogenesis was quantified through the In silico Risk Index (IRI). This work pursues two main objectives. First, to evaluate the impact of transmural differentiation: by implementing an innovative preprocessing algorithm, the physiological electrical heterogeneity of the three ventricular layers (epicardium, mid-myocardium, and endocardium) was introduced into the model. Second, to develop a rapid predictive model based purely on the morphological parameters of the ischemic lesion. The extent of the ischemic zone and its spatial distribution, quantified through an original Asymmetry Index (AI), were identified as the key geometric determinants of arrhythmogenesis. Using the simulation-derived IRI as the ground-truth, a Risk Calibrated Index (RCI) was then calibrated. By combining these simple geometric parameters, the RCI allows for a direct estimation of arrhythmic susceptibility, providing a potential clinical risk stratification tool that bypasses the need for complex electrophysiological simulations.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_07_Ubiali_Tesi.pdf
accessibile in internet per tutti
Dimensione
28.98 MB
Formato
Adobe PDF
|
28.98 MB | Adobe PDF | Visualizza/Apri |
|
2026_07_Ubiali_Executive Summary.pdf
accessibile in internet per tutti
Dimensione
6.28 MB
Formato
Adobe PDF
|
6.28 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/10589/261247