Information Extraction Techniques in Hyperspectral Imaging Biomedical Applications

Abstract

Hyperspectral imaging (HSI) is a technology able to measure information about the spectral reflectance or transmission of light from the surface. The spectral data, usually within the ultraviolet and infrared regions of the electromagnetic spectrum, provide information about the interaction between light and different materials within the image. This fact enables the identification of different materials based on such spectral information. In recent years, this technology is being actively explored for clinical applications. One of the most relevant challenges in medical HSI is the information extraction, where image processing methods are used to extract useful information for disease detection and diagnosis. In this chapter, we provide an overview of the information extraction techniques for HSI. First, we introduce the background of HSI, and the main motivations of its usage for medical applications. Second, we present information extraction techniques based on both light propagation models within tissue and machine learning approaches. Then, we survey the usage of such information extraction techniques in HSI biomedical research applications. Finally, we discuss the main advantages and disadvantages of the most commonly used image processing approaches and the current challenges in HSI information extraction techniques in clinical applications.

Histological Hyperspectral Glioblastoma Dataset (HistologyHSI-GB)

“…purposes using various image processing methods. Although a wide variety of techniques are used in the literature to this end, it is difficult to compare the different approaches fairly, mainly due to the lack of publicly available datasets 34 .…”

Intra-operative brain tumor detection with deep learning-optimized hyperspectral imaging

“…The use of HSI in medicine has been growing in the last decade, [24][25][26] and it is establishing itself as a noninvasive, non-ionizing, label-free diagnostic tool. As such, it has seen significant applications in many medical domains, including the analysis of multiple types of cancer (in vivo and ex vivo), 27 computational pathology, 28 gastroenterology, 29 dermatology, 30 general surgery [31][32][33] (all references here are to recent review articles).…”

“…One of the most important aspects underlying all the above-mentioned studies is the quest for the optimal method to extract information from biomedical HSI data. 26 The simplest method is based on optical inverse modelling: if a physical model for the tissue spectra exists, observations can be used to directly infer its parameters; 34 e.g., the ratio of the spectral bands corresponding to oxygenated and de-oxygenated Hb can be used for cancer detection. 35 In most cases however, an accurate physical model is missing, and a data-driven machine learning approach is more suitable, either of the classical feature learning (FL) type, or based on deep learning (DL).…”