Comparison of micro-CT imaging and histology for approximal caries detection.

Histological sectioning is a generally accepted in vitro validation method for caries detection techniques. However, it requires cumbersome sample preparation and induces irreversible sample destruction. Micro-Computer Tomography (micro-CT) allows non-destructive imaging of tooth structure. The aim of this study was to compare the performance of histological sectioning and micro-CT imaging in detecting approximal carious lesions. Unlike previous studies, evaluation is objectified by comparing visual appearance of exactly corresponding anatomical regions. Sixty extracted human teeth were scanned with a desktop micro CT system. Axial histological slices were prepared and photographed. Sample preparation, combined with dedicated image processing, ensured selection of identical anatomical regions on radiographic and histological images. Evaluation of the presence and extent of carious lesions was performed by four dentists using custom-designed software. Each section was scored independently (histo or micro CT). Scores of approximal surfaces were retained for further analysis. Spearman's correlation coefficients (0.738 to 0.829, p < 0.0001) showed a good agreement between signs of carious lesions in the identical region obtained with both methods. Bland-Altman plots showed that 90.76% of the data points were within the limits of agreement. Micro-CT imaging was shown to provide an interesting alternative to histological sectioning as detection method for carious lesions.

2D image registration using focused mutual information for application in dentistry.

Spatial alignment of image data is a common task in computer vision and medical imaging. This should preferentially be done with minimal intervention of an operator. Similarity measures with origin in the information theory such as mutual information (MI) have proven to be robust registration criteria for this purpose. Intra-oral radiographs can be considered images of piecewise rigid objects. Teeth and jaws are rigid but can be displaced with respect to each other. Therefore MI criteria combined with affine deformations tend to fail, when teeth and jaws move with respect to each other between image acquisitions. In this paper, we consider a focused weighing of pixels in the reference image. The resulting criterion, focused mutual information (FMI) is an adequate tool for the registration of rigid parts of a scene. We also show that the use of FMI is more robust for the subtraction of lateral radiographs of teeth, than MI confined to a region of interest. Furthermore, the criterion allows the follow-up of small carious lesions when upper and lower jaw moved between the acquisition of test and reference image.

A probe for immittance spectroscopy based on the parallel electrode technique.

This paper presents the construction of a probe for immittance spectroscopy. The probe is based on the four-electrode technique. The current application electrodes are long parallel strips and have been partitioned into three segments to measure the immittance of relatively small volumes. The probe constant has been calculated assuming anisotropy of the materials examined. The experiments, based on the measurements of KCl solutions and numerical simulations (finite element method--FEM), have shown a high degree of consistency with the theoretical value. Experimental in vivo measurements on swine gluteal tissue were performed and the conductivity and permittivity reconstructed from these measurements are presented.

Genetic and antigenic heterogeneity among feline calicivirus isolates from distinct disease manifestations.

The capsid protein genes of five feline calicivirus (FCV) isolates associated with different disease manifestations were cloned and sequenced. The viruses represented two recent isolates from cats with chronic stomatitis, one recent isolate from a cat with acute stomatitis, one recent isolate each from a cat with acute respiratory symptoms and the classical limping syndrome strain FCV-2280. The amino acid sequences were compared with eight other published sequences and analyzed for their relationships. Phylogenetic analysis of the complete capsid protein sequences or of known antigenic regions of that protein (hypervariable regions A and E) did not group the isolates of different disease manifestations in distinct subclusters. Monoclonal antibodies (MAbs) generated against either a chronic stomatitis isolate or a recent isolate associated with respiratory symptoms were tested against a panel of 11 recent isolates and four "classical' FCV strains, covering all known disease associations. With those MAbs no obvious clustering with respect to disease manifestation could be seen. Four specific sera prepared in rabbits against our prototype isolates also failed to cluster those isolates according to the disease manifestations when examined in neutralization tests. From these antigenic and genetic analyses of the capsid protein the hypothesis of the existence of biotypes of FCV responsible for distinct disease manifestations could not be confirmed.