Phasor Representation Approach for Rapid Exploratory Analysis of Large Infrared Spectroscopic Imaging Data Sets

Infrared (IR) spectroscopic imaging is potentially usefulfor digitalhistopathology as it provides spatially resolved molecular absorptionspectra, which can subsequently yield useful information by powerfulartificial intelligence methods. A typical analysis pipeline in usingIR imaging data for chemical pathology often involves iterative processesof segmentation, evaluation, and analysis that necessitate rapid dataexploration. Here, we present a fast, reliable, and intuitive methodbased on a phasor representation of spectra and discuss its uniqueapplicability for IR imaging data. We simulate different featuresextant in IR spectra and discuss their influence on the phasor waveforms;similarly, we undertake IR image analysis in the transform space tounderstand spectral similarity and variance. We demonstrate the potentialof phasor analysis for biomedical tissue imaging using a variety ofsamples, using fresh frozen surgical prostate resections and formalin-fixedparaffin-embedded breast cancer tissue microarray samples as modelsystems that span common histopathology practice. To demonstrate furthergeneralizability of this approach, we apply the method to data fromdifferent experimental conditions including standard (5.5 & mu;mx 5.5 & mu;m pixel size) and high-definition (1.1 & mu;mx 1.1 & mu;m pixel size) Fourier transform IR (FTIR) spectroscopicimaging using transmission and transflection modes. Quantitative segmentationresults from our approach are compared to previous studies, showinggood agreement and quick visualization. The presented method is rapid,easy to use, and highly capable of deciphering compositional differences,presenting a convenient tool for exploratory analysis of IR imagingdata.

Automated summary

Infrared (IR) spectroscopic imaging provides spatially resolved molecular absorption spectra, and this study introduces a fast and reliable method based on a phasor representation of spectra. The method is applicable for rapid data exploration and analysis of IR imaging data, and shows potential in biomedical tissue imaging.