Statistics imaging data

Dudewicz, E.J. Statistical Analysis of Magnetic Resonance Imaging Data in The Normal Brain, Part I Data, Screening, Normality, Discrimination, Variability" unpublished report, 1985. 

As an alternative, qualitative classification models are developed from a library composed of pure component spectra. It is considerably easier to generate an imaging data set of a pure component, and these data enjoy a certain statistical robusmess due to the large number of individual pixels representing the sample. The resulting models are then applied to the sample image data to produce score predictions that... 

These same analysis techniques can be applied to chemical imaging data. Additionally, because of the huge number of spectra contained within a chemical imaging data set, and the power of statistical sampling, the PLS algorithm can also be applied in what is called classification mode as described in Section 8.4.5. When the model is applied to data from the sample, each spectrum is scored relative to its membership to a particular class (i.e. degree of purity relative to a chemical component). Higher scores indicate more similarity to the pure component spectra. While these scores are not indicative of the absolute concentration of a chemical component, the relative abundance between the components is maintained, and can be calculated. If all sample components are accounted for, the scores for each component can be normalized to unity, and a statistical assessment of the relative abundance of the components made. 

These same analysis techniques can be applied to chemical imaging data. Additionally, because of the huge number of spectra contained within a chemical imaging data set, and the power of statistical sampling, the PLS algorithm can also be applied in what is called classification mode. In this case, the reference library used to establish the PLS model is... 

It is extremely useful to move beyond a subjective and qualitative analysis of the spatial distribution of sample components, and to begin to explore the quantitative information contained within chemical imaging data sets. One of the most powerful statistical representations of an image does not even maintain spatial information. A chemical image can be represented as a histogram, with intensity along the x-axis and the number of pixels with that intensity along the y-axis. This is a statistical and quantitative... 

Figure 2.6 The figure shows the different types of analyses that can be performed on chemical imaging data. The types of analyses that are performed can be grouped into three categories component abundance estimation, statistical analysis of component distribution, and morphological analysis of discrete particles. All three analyses are used to make inter- and intrasample comparisons, generating abundance and content uniformity estimates, sample heterogeneity and blend uniformity characterization, as well as domain statistics and domain size uniformity data.

Another type of classification is outlier selection or contamination identification. As an example, in Fig. 4.23(b), the butter is the desired material and bacteria the contamination. An arbitrary threshold for this image would be 0.02, in which all pixels >0.02 are considered suspect, and hopefully, because this is a food product, decontamination procedures are pursued. In these two examples of classification, only arbitrary thresholds have been defined and, as such, confidence in these classifications is lacking. This confidence can be achieved through statistical methods. Although this chapter is not the appropriate place for an involved discussion of application of statistics toward data analysis, we will give one example often used in chemometric classification. 

FIGURE 37.2 DMSO response of xerotic leg skin, (a) Dose dependent increase of DMSO induced cutaneous blood flow before on untreated skin, (b) After twice daily application of a water-in-oil emulsion for six weeks the response was markedly reduced. DMSO induced blood flow was measured with a laser Doppler imager. Data expressed as blood perfusion units. Statistical significance was determined using the paired r-test ( p >. 05,... 

A spectroscopic NIR imaging system, using a FPA detector, has been developed for remote and on-line measurements on a macroscopic scale. Multivariate statistical techniques are required to extract the important information from the multidimensional spectroscopic images. These techniques include PCA and linear discriminant analysis for supervised classification of spectroscopic image data (178). 

---