Tesi etd-09242025-142130
Link copiato negli appunti
Tipo di tesi
Dottorato
Autore
MAZZOTTA, ALESSANDRO DARIO
URN
etd-09242025-142130
Titolo
Advanced Training in Robotic-Assisted Surgery For Delicate Tissue Manipulation
Settore scientifico disciplinare
ING-IND/34
Corso di studi
Istituto di Biorobotica - PHD IN BIOROBOTICA
Relatori
relatore Prof.ssa MENCIASSI, ARIANNA
Parole chiave
- Global Evaluative Assessment of Robotic Skills
- human structures
- PVA
- Animal Model
- Translational Validation
- Preliminary Multicentric Validation
- Design and Development of the Simulator
Data inizio appello
27/04/2026;
Disponibilità
completa
Riassunto analitico
This PhD project investigates the development, validation, and translational impact of a high-fidelity, sensorized simulator for training in robotic-assisted surgery (RAS), with a focus on delicate tissue in particular vascular dissection. The research addresses a fundamental limitation of robotic platforms—the lack of haptic feedback—which increases the risk of intraoperative bleeding, excessive traction, and conversion to open surgery. This work was carried out by a surgeon who progressively acquired fundamental notions of biorobotics, bridging clinical expertise with engineering innovation.
The thesis is structured into three main phases:
Design and Development of the Simulator
A vascular model was engineered using silicone for vessels and polyvinyl alcohol (PVA) for connective tissue, replicating the anatomical and mechanical properties of human structures. A fabric-based stretch sensor was embedded into the vessel wall to measure deformation during dissection. Excessive traction triggered LED and acoustic alerts, providing trainees with real-time corrective feedback. The model reproduced the surgical steps of vascular exposure, loop passage, and stapler positioning, simulating a critical task in thoracic and hepatobiliary procedures.
Preliminary Multicentric Validation
Validation was performed at IRCAD (France) and Sapienza University of Rome (Italy), involving surgeons across multiple specialties and experience levels. Face and content validity confirmed the realism and educational utility of the simulator. Construct validity was demonstrated through significant differences in vessel deformation metrics between novices, fellows, and experts. Additionally, specialty-specific differences (e.g., between general and urologic surgeons) were detected, supporting the simulator’s potential for tailored training.
Translational Validation in an Animal Model
A randomized controlled trial compared novices trained on the simulator with untrained counterparts performing robotic vascular dissection in a porcine model. Trained participants achieved significantly higher GEARS (Global Evaluative Assessment of Robotic Skills) scores, particularly in depth perception and bimanual dexterity. This provided the first evidence that structured dry-lab training with a sensorized simulator can enhance in vivo surgical performance.
The thesis is structured into three main phases:
Design and Development of the Simulator
A vascular model was engineered using silicone for vessels and polyvinyl alcohol (PVA) for connective tissue, replicating the anatomical and mechanical properties of human structures. A fabric-based stretch sensor was embedded into the vessel wall to measure deformation during dissection. Excessive traction triggered LED and acoustic alerts, providing trainees with real-time corrective feedback. The model reproduced the surgical steps of vascular exposure, loop passage, and stapler positioning, simulating a critical task in thoracic and hepatobiliary procedures.
Preliminary Multicentric Validation
Validation was performed at IRCAD (France) and Sapienza University of Rome (Italy), involving surgeons across multiple specialties and experience levels. Face and content validity confirmed the realism and educational utility of the simulator. Construct validity was demonstrated through significant differences in vessel deformation metrics between novices, fellows, and experts. Additionally, specialty-specific differences (e.g., between general and urologic surgeons) were detected, supporting the simulator’s potential for tailored training.
Translational Validation in an Animal Model
A randomized controlled trial compared novices trained on the simulator with untrained counterparts performing robotic vascular dissection in a porcine model. Trained participants achieved significantly higher GEARS (Global Evaluative Assessment of Robotic Skills) scores, particularly in depth perception and bimanual dexterity. This provided the first evidence that structured dry-lab training with a sensorized simulator can enhance in vivo surgical performance.
File
| Nome file | Dimensione |
|---|---|
| Tesi_for...zotta.pdf | 2.00 Mb |
Contatta l'autore |
|