BACKGROUND
Heterogeneous membranes or films are thin and soft structures with spatial variations in material property and thickness. Important heterogeneous membranes include bio tissues such as heart valves, eardrum membranes, skin, as well as artificial synthesized bio-compatible tissues, non-woven materials. Mechanical behavior of heterogeneous membranes is not well understood, mainly due to the difficulty in obtaining accurate and reliable material property data. Currently, the approach to obtaining mechanical properties of heterogenous soft membrane local areas relies on time-consuming and necessarily destructive sampling: cutting samples into small pieces and testing each piece separately. This method causes results with a low accuracy as the tissues’ properties change during the transfer from their specimen to a testing lab, and the loss of bio-context of the samples. These methodological shortcomings represent a serious roadblock to the advancement of research and development of heterogeneous bio-membranes and films, as well as related bio-medical research and engineering.
SUMMARY OF TECHNOLOGY
Researchers at OSU have developed an approach to characterizing heterogeneous soft membranes with a novel data processing software. The software uses machine learning techniques to train a neural network capable of real-time (~1ms) determination of high-resolution material properties using non-invasive full-field strain measurements of bio-tissue deformation. The proposed neural network is in part trained from finite element method based, computer-simulated full-field strain data, which can generate training data according to different mechanical property ranges (e.g., hard vs. soft materials). This novel approach requires less tissue preparation and preserves the integrity of the tissue sample. Experimental results show this methodology can determine the properties of up to 16x16 local areas and can be extended for greater numbers of local areas defined on a heterogenous membrane, such as various types of tissues.
POTENTIAL AREAS OF APPLICATION
MAIN ADVANTAGES
STAGE OF DEVELOPMENT