Work

Breast cancer histopathology using infrared spectroscopic imaging: The impact of instrumental configurations

Year: 2021

Type: article

Abstract: Digital analysis of cancer specimens using spectroscopic imaging coupled to machine learning is an emerging area that links spatially localized spectral signatures to tissue structure and disease. In this study, we examine the role of spatial-spectral tradeoffs in infrared spectroscopic imaging configurations for probing tumors and the associated microenvironment profiles at different levels of model complexity. We image breast tissue using standard and high-definition Fourier Transform Infrared (FT-IR) imaging and systematically examine the localization, spectral origins, and utility of data for classification. Results demonstrate that higher spatial detail provides high sensitivity and specificity of tissue segmentation, despite the increased subcellular variability. High definition imaging also allows accurate analysis of complex, multiclass models of breast tissue without compromising accuracy. A comparison of results also highlights the key differences in the data distributions and classification performance across modalities to better guide decision making for acquiring and analyzing IR imaging data for specific histopathological models.

Source: Clinical Spectroscopy

DOI: 10.1016/j.clispe.2021.100006

Authors: Shachi Mittal, Tomasz P. Wrobel, Michael Walsh, Andre Kajdacsy-Balla, Rohit Bhargava

Institutions: University of Illinois Urbana-Champaign, University of Wisconsin–Eau Claire, University of Illinois Chicago, Cancer Center at Illinois

Language: English

Topic: Spectroscopy Techniques in Biomedical and Chemical Research

Subfield: Biophysics

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $1,750