Automatic Identification of Paraffin Pixels on FTIR Images Acquired on FFPE Human Samples

The transfer of mid-infrared spectral histopathology to the clinic will be possible provided that its application in clinical practice is simple. Rapid analysis formalin-fixed paraffin-embedded (FFPE) tissue section thus a prerequisite. chemical dewaxing these samples before image acquisition used by majority studies contradiction with this principle. Fortunately, silico images acquired on FFPE using extended multiplicative signal correction (EMSC). However, removal pure paraffin pixels essential perform relevant classification spectra. So far, task was only if manual and subjective histogram analysis. In article, we propose new automatic multivariate methodology based optimized combinations EMSC regression coefficients validity indices KMeans clustering separate pixels. validation our method performed simulated infrared measuring Jaccard index between partitions model, values always over 0.90 for diverse baseline complexity signal-to-noise ratio. These encouraging results were also validated real comparing classical ones computing obtained same but after dewaxing, 0.84.