A Clinical Measuring Platform for Building the Bridge Across the Quantification of Pathological N-Cells in Medical Imaging for Studies of Disease

Abstract

In this paper, a clinical measuring platform for quantifying nucleus-cells (MPQ-N-cells) at combining a novel color region-based segmentation strategy is proposed to accelerate the discovery of diseases diagnostically in medical imaging. In the approach, average values of colors in an image are employed as similarity criteria to assign image voxels to regions using the minimum distance classifier in the color region growing process. Then, the binary image transformation and graphic contour line procedure are performed, followed by the operation of region area calculation to obtain the actual numbers of voxels within the segmented patterns of the N-cells quantitatively. The proposed approach of MPQ-N-cells is implemented on the heterogeneous medical image datasets related to Parkinson disease, oculopharyngeal muscular dystrophy (one type of protein conformational diseases) and glioblastoma cancer. Implementation results reveal that the proposed MPQ-N-cells approach is capable of quantifying a variety of pathological N-cells clinically with improved data visualization in heterogeneous datasets. This study has the potential to lead to more successful measurement of cell diagnostically and further to track changes of cell in medical imaging for a longitudinal study on supporting the studies of disease.