Pyramidic Clustering of Large-Scale Microarray Images

With ongoing research and development of imaging techniques such as those involved in brain MRIs, cDNA microarrays and satellite reconnaissance, the need for tools that can intelligently parse larger images is ever increasing. One group of such techniques often used is that of segmentation, an example of which is that of clustering algorithms. In order to deal with large data sets, current approaches require the data to be sampled or summarized before true analysis can take place. In this paper we propose a novel image analysis technique using pyramidic type grouping, namely copasetic clustering, which focuses on the problem of applying traditional clustering techniques to these large-scale image data sets with limited resources. A further benefit of the technique is the transparency of its intermediate clustering steps; when applied to spatial data sets this allows the capture and incorporation of contextual information to improve result accuracy. The algorithm achieves an ∼1--3 dB w-to-noise ratio when compared with the conventional techniques described.