Non-invasive prenatal diagnosis (NIPD) aims to obtain fetal genetic information while eliminating the risks associated with invasive procedures such as amniocentesis and villocentesis. Among the proposed strategies, the isolation of fetal nucleated erythroblasts (fNRBC) from maternal blood is a promising approach, but technically complex given their extreme rarity. In the field of separation technologies, gravitational field-flow fractionation (GrFFF) stands out as a simple, economical, biocompatible alternative that does not require the use of antibodies and is already patented for NIPD applications. The aim of this work was to develop a Digital Twin of the GrFFF system capable of reproducing and predicting its behavior, supporting the optimization of erythroblast isolation. The GrFFF fluidic device was used to perform experiments with cord blood, analyzing both homogeneous cell populations and complex samples treated with density gradient protocols. Furthermore, the use of rigid microspheres comparable in size to cells allowed the evaluation of the deformability contribution in the separation process. The experimental fractograms obtained formed the baseline for the calibration and validation of the Digital Twin, implemented in Python, integrating physical, stochastic modeling and experimental data. An interactive graphical user interface was developed to translate the computational model into a tool to support experimental decisions. The Digital Twin has demonstrated predictive capabilities under realistic conditions, reproducing the morphology of fractograms and trends as flow rate and composition vary. Finally, it has made it possible to define the optimal operating parameters to maximize erythroblast recovery, clarifying the role of cell deformability in separation mechanisms.
La diagnosi prenatale non invasiva (NIPD) mira a ottenere informazioni genetiche fetali eliminando i rischi associati a procedure invasive quali amniocentesi e villocentesi. Tra le strategie proposte, l’isolamento di eritroblasti nucleati fetali (fNRBC) dal sangue materno rappresenta un approccio promettente, ma tecnicamente complesso data la loro estrema rarità. Nell’ambito delle tecnologie di separazione, il frazionamento in campo flusso gravitazionale (Gravitational Field-Flow Fractionation, GrFFF) si distingue come alternativa semplice, economica, biocompatibile e priva dell’utilizzo di anticorpi, già oggetto di brevetto per applicazioni NIPD. Lo scopo di questo lavoro è stato sviluppare un Digital Twin del sistema GrFFF in grado di riprodurne e predirne il comportamento, supportando l’ottimizzazione dell’isolamento degli eritroblasti. Il dispositivo fluidico GrFFF è stato impiegato per eseguire esperimenti con sangue cor- donale, analizzando sia popolazioni cellulari omogenee, sia campioni complessi. Inoltre, l'uso di microsfere rigide di dimensioni comparabili a quelle delle cellule ha permesso di valutare il contributo della deformabilità nel processo separativo. I frattogrammi sperimentali ottenuti hanno costituito la base per la calibrazione e validazione del Digital Twin, implementato in Python, integrando modellazione fisica, stocastica e dati sperimentali. Un’interfaccia grafica interattiva è stata sviluppata per tradurre il modello computazionale in uno strumento di supporto alle decisioni sperimentali. Il Digital Twin ha dimostrato capacità predittiva in condizioni realistiche, riproducendo la morfologia dei frattogrammi e le tendenze al variare della portata e della composizione. Ha infine permesso di definire i parametri operativi ottimali per massimizzare il recupero degli eritroblasti, chiarendo il ruolo della deformabilità cellulare nei meccanismi di separazione.
Experimental characterization and digital twin development for non invasive prenatal diagnosis
Touijar, Sarah;Rastelli, Federica
2025/2026
Abstract
Non-invasive prenatal diagnosis (NIPD) aims to obtain fetal genetic information while eliminating the risks associated with invasive procedures such as amniocentesis and villocentesis. Among the proposed strategies, the isolation of fetal nucleated erythroblasts (fNRBC) from maternal blood is a promising approach, but technically complex given their extreme rarity. In the field of separation technologies, gravitational field-flow fractionation (GrFFF) stands out as a simple, economical, biocompatible alternative that does not require the use of antibodies and is already patented for NIPD applications. The aim of this work was to develop a Digital Twin of the GrFFF system capable of reproducing and predicting its behavior, supporting the optimization of erythroblast isolation. The GrFFF fluidic device was used to perform experiments with cord blood, analyzing both homogeneous cell populations and complex samples treated with density gradient protocols. Furthermore, the use of rigid microspheres comparable in size to cells allowed the evaluation of the deformability contribution in the separation process. The experimental fractograms obtained formed the baseline for the calibration and validation of the Digital Twin, implemented in Python, integrating physical, stochastic modeling and experimental data. An interactive graphical user interface was developed to translate the computational model into a tool to support experimental decisions. The Digital Twin has demonstrated predictive capabilities under realistic conditions, reproducing the morphology of fractograms and trends as flow rate and composition vary. Finally, it has made it possible to define the optimal operating parameters to maximize erythroblast recovery, clarifying the role of cell deformability in separation mechanisms.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_03_Rastelli_Touijar_Executive_Summary.pdf
non accessibile
Descrizione: Executive summary
Dimensione
2.68 MB
Formato
Adobe PDF
|
2.68 MB | Adobe PDF | Visualizza/Apri |
|
2026_03_Rastelli_Touijar_Tesi.pdf
non accessibile
Descrizione: Thesis
Dimensione
23.07 MB
Formato
Adobe PDF
|
23.07 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/10589/253173