Erlangen 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
ST: Fachverband Strahlen- und Medizinphysik
ST 6: Poster Session
ST 6.2: Poster
Donnerstag, 19. März 2026, 14:15–15:45, Redoutensaal
Tracking Polymeric Residues in Silicone Implants and Adjacent Tissue via Raman Spectroscopy and CNN-Based Classification — •Hoang Thinh Nguyen1, Susanne Gramsall2, Rima Nuwayhid3, and Paul-Tiberiu Miclea1,2 — 1Martin Luther University Halle-Wittenberg, Institute of Physics, MicroMD Group — 2Fraunhofer Center for Silicon Photovoltaics, Halle (Saale), Germany — 3University Hospital Leipzig, D-04103 Leipzig, Germany
Silicone breast implants are widely used in reconstructive and aesthetic surgery. While silicone is generally considered biologically inert, concerns remain regarding long-term biocompatibility and potential microplastic release. Six samples (2 explanted implants, 4 surrounding tissue specimens) were analyzed using Raman spectroscopy combined with CNN-based spectral classification. Implant surfaces were examined directly, while tissue samples underwent KOH digestion and filtration onto silicon membrane filters (1 micrometer pore size). A Mask R-CNN model trained on about 50 reference spectra with an 8:2 split outperformed SVM and KNN. Raman mapping detected PVC on implant surfaces. In tissue, polymeric residues were found (primarily silicone), but also non-silicone polymers such as PS and PVC, near 960 cm-1 and 1600 cm-1. These results suggest migration of polymer components into surrounding tissue and support links to implant-related inflammatory responses.
Keywords: 1D-CNN; surgery; Silicone Implant; microplastic; inflammatory