We present a non-parametric, unsupervised method for localizing pathology-mimicking structures (spheroids) embedded in agarose-based phantoms using single-channel ultrasound data. The method aims to automate the detection of spheroids within unlabeled datasets, enabling subsequent signal characterization and potential inference of cell line-specific properties. This fully data-driven approach removes dependence on human supervision or reference signals, offering a scalable and reproducible solution for preclinical tissue assessment.
Quantile Thresholding and Unsupervised AI for Human Spheroids Detection in Tissue-Mimicking Phantoms
Benedetti, Ilaria;Auletta, Fabrizia;Barravecchia, Ivana;Angeloni, Debora;Oddo, Calogero Maria
2025-01-01
Abstract
We present a non-parametric, unsupervised method for localizing pathology-mimicking structures (spheroids) embedded in agarose-based phantoms using single-channel ultrasound data. The method aims to automate the detection of spheroids within unlabeled datasets, enabling subsequent signal characterization and potential inference of cell line-specific properties. This fully data-driven approach removes dependence on human supervision or reference signals, offering a scalable and reproducible solution for preclinical tissue assessment.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
Quantile_Thresholding_and_Unsupervised_AI_for_Human_Spheroids_Detection_in_Tissue-Mimicking_Phantoms.pdf
non disponibili
Tipologia:
Documento in Post-print/Accepted manuscript
Licenza:
Copyright dell'editore
Dimensione
1.32 MB
Formato
Adobe PDF
|
1.32 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
