Atomic Force Microscopy (AFM) has the unique capability of probing the nanoscale mechanical properties of biological systems that affect and are affected by the occurrence of many pathologies, including cancer. This capability has triggered growing interest in the translational process of AFM from physics laboratories to clinical practice. A factor still hindering the current use of AFM in diagnostics is related to the complexity of AFM data analysis, which is time-consuming and needs highly specialized personnel with a strong physical and mathematical background. In this work, we demonstrate an operator-independent neural-network approach for the analysis of surgically removed brain cancer tissues. This approach allowed us to distinguish - in a fully automated fashion - cancer from healthy tissues with high accuracy, also highlighting the presence and the location of infiltrating tumor cells.
Minelli, E., Ciasca, G., Sassun, T. E., Antonelli, M., Palmieri, V., Papi, M., Maulucci, G., Santoro, A., Giangaspero, F., Delfini, R., Campi, G., De Spirito, M., A fully-automated neural network analysis of AFM force-distance curves for cancer tissue diagnosis, <<APPLIED PHYSICS LETTERS>>, 2017; 111 (14): 1-5. [doi:10.1063/1.4996300] [http://hdl.handle.net/10807/122493]
A fully-automated neural network analysis of AFM force-distance curves for cancer tissue diagnosis
Minelli, EleonoraPrimo
Membro del Collaboration Group
;Ciasca, Gabriele
;Palmieri, ValentinaMembro del Collaboration Group
;Papi, MassimilianoMembro del Collaboration Group
;Maulucci, GiuseppeMembro del Collaboration Group
;De Spirito, MarcoUltimo
2017
Abstract
Atomic Force Microscopy (AFM) has the unique capability of probing the nanoscale mechanical properties of biological systems that affect and are affected by the occurrence of many pathologies, including cancer. This capability has triggered growing interest in the translational process of AFM from physics laboratories to clinical practice. A factor still hindering the current use of AFM in diagnostics is related to the complexity of AFM data analysis, which is time-consuming and needs highly specialized personnel with a strong physical and mathematical background. In this work, we demonstrate an operator-independent neural-network approach for the analysis of surgically removed brain cancer tissues. This approach allowed us to distinguish - in a fully automated fashion - cancer from healthy tissues with high accuracy, also highlighting the presence and the location of infiltrating tumor cells.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
