Author: Silvia Calusi, Lucia Cavigli, Alberto Dalla Mora, Laura Di Sieno, Giacomo Insero, Riccardo Lisci, Livia Marrazzo, Cosimo Nardi, Stefania Pallotta, Andrea Profili, Fulvio Ratto, Giovanni Romano, Michaela Servi, Immacolata Vanore, Yary Volpe 👨🔬
Affiliation: Italian National Research Council IFAC-CNR, Institute of Applied Physics, Department of Physics, Politecnico di Milano, Department of Agricultural Food and Forestry System, University of Florence, Medical Physics Unit, Azienda Ospedaliero-Universitaria Careggi, Department of Industrial Engineering, University of Florence, Department of Experimental and Clinical Biomedical Sciences “Mario Serio”, University of Florence 🌍
Purpose: To develop a multi-purpose lung phantom prototype to replicate respiratory dynamics and morphological features observed in clinical radiological (CT and MR) imaging of lung parenchyma.
Methods: To create the lung-equivalent material, micron-sized hydrogel particle emulsions were combined with polydimethylsiloxane (PDMS) sponges featuring sub-millimeter pores. Material characterization was performed using CT and MR imaging to compare CT numbers and longitudinal relaxation times (T1) with those of human lung parenchyma, also assessing the material long-term reproducibility. Realistic respiratory dynamics were achieved through an ad-hoc developed chest phantom, scanned and analysed by 4D-CT.
Results: CT and 1.5 T MR imaging demonstrated that the PDMS sponges successfully replicate human lung CT number values (-850 to -650 HU) and T1 relaxation times (1100-1300 ms). CT number, T1 and volume of tested material resulted to be stable over a six-month period, showing data to be consistent within a standard deviation (relative standard deviation below 2%)
The lung-like sponge was incorporated into a miniaturized 3D-printed thorax phantom, which includes a motorized system to compress the sponge, mimicking diaphragmatic respiration and moving simulated lesion objects. Dosimetric evaluations are allowed due to a dedicated radiochromic film housing. Validation of the sponge compressibility through 4D-CT imaging revealed that CT number variation range across respiratory phases was up to sixfold greater in compressed sponge (-740 to -650 HU) compared to uncompressed one (-710 to -695 HU). The film axial plane remained stable throughout the compression process.
Conclusion: This study confirms the feasibility of using PDMS sponges to replicate lung characteristics and respiratory dynamics. The intrinsic inhomogeneity of the sponges allowed to recreate both healthy and pathological lung morphologies. The developed phantom met the study objectives, and future work will focus on developing a full-scale phantom with enhanced realism and an MR-compatible motion system to support theranostic applications, including MRI-guided radiotherapy.