Selecting optimal features from Fourier transform infrared spectroscopy for discrete-frequency imaging

Abstract

Tissue histology utilizing chemical and immunohistochemical labels plays an important role in biomedicine and disease diagnosis. Recent research suggests that mid-infrared (IR) spectroscopic imaging may augment histology by providing quantitative molecular information. One of the major barriers to this approach is long acquisition time using Fourier-transform infrared (FTIR) spectroscopy. Recent advances in discrete frequency sources, particularly quantum cascade lasers (QCLs), may mitigate this problem by allowing selective sampling of the absorption spectrum. However, DFIR imaging only provides a significant advantage when the number of spectral samples is minimized, requiring a priori knowledge of important spectral features. In this paper, we demonstrate the use of a GPU-based genetic algorithm (GA) using linear discriminant analysis (LDA) for DFIR feature selection. Our proposed method relies on pre-acquired broadband FTIR images for feature selection. Based on user-selected criteria for classification accuracy, our algorithm provides a minimal set of features that can be used with DFIR in a time-frame more practical for clinical diagnosis.

Automated osteosclerosis grading of clinical biopsies using infrared spectroscopic imaging

“…Cell and tissue classification using FTIR imaging is a mature area of research, [25][26][27][28][29] however specific clinical applications are rarely seen. We propose a protocol for automated classification of trabecular bone and collagen and calculation of TBA.…”

“…Feature selection was performed with a previously published genetic algorithm leveraging linear discriminant analysis. 29 A Random Forest (RF) classifier 11 was trained on the selected features to differentiate between trabecular bone, collagen, stroma, and background pixels. The RF used 100 untruncated trees for classification.…”

“…Our previous FTIR studies show that a minimal number of optimal features are sufficient for classification of most of the histological classes. 29 Optimal feature selection corresponding to histological classes provides the potential for future discrete-frequency imaging methods.…”

Spatial correlation of native and engineered cartilage components at micron resolution

“…Current histological methods to evaluate distribution of proteins throughout tissue utilize dyes that can be sensitive to preparation techniques . In addition, histological dyes may exhibit nonspecific staining and may be challenging for quantification of the distribution of a specific component throughout the sample.…”

“…Current histological methods to evaluate distribution of proteins throughout tissue utilize dyes that can be sensitive to preparation techniques. 16 In addition, histological dyes may exhibit nonspecific staining and may be challenging for quantification of the distribution of a specific component throughout the sample. Fourier transform infrared imaging spectroscopy (FT-IRIS) is an advanced vibrational spectroscopy technique that combines an FT-IR spectrometer with an optical microscope with an array detector to produce spatially resolved images based on the IR spectra.…”