After more than three decades of research, breakdown at the dentin/adhesive (d/a) interface continues to plague the long-term clinical success of Class II composite restorations. A limitation of previous investigations of the interface is that the data was primarily qualitative in nature and that much of the data was collected using destructive techniques which provided a single measurement on one sample. Understanding the mechanisms and interfacial behavior of the d/a interface requires structure/property analyses that are complementary, synergistic and quantitative as well as qualitative. In most laboratories, these complementary analyses are performed on a heterogeneous mix of instrumentation which has been acquired from different manufacturers. Each instrument uses software unique to the vendor and frequently, the data must be stored in a proprietary format. Managing this mass of data, collected in several different formats, becomes a daunting, time-consuming task. Our inability to collect this data in a common format prohibits direct correlation of the complementary results; it impedes our ability to recognize trends in heterogeneous data. The overall hypothesis of this work is that spectral imaging data collected in vendor-specific formats using a variety of analytical techniques can be translated to a common computing platform that allows data sharing among research groups, visualization of relationships in heterogeneous data and the construction of new detailed, quantitative structure/property data images.
The specific aims will test the following hypotheses: 1) heterogeneous incompatible data from different analytical instrumentation, e.g., micro-Raman spectroscopy, FTIR spectroscopy and scanning acoustic microscopy, can be translated from vendor-specific formats to a common computing platform and managed without compromising the integrity of the original data; 2) the data convergence provided by this common computing platform will promote and facilitate visualization of relationships; 3) detailed, quantitative structure/property data images can be constructed using the data repository environment offered by this new, common computing platform and using this new imaging platform, the interrelationship between structure and properties can be understood. ? ?
