Spectral organ fingerprints for machine learning-based intraoperative tissue classification with hyperspectral imaging in a porcine model

Visual discrimination of tissue during surgery can be challenging since different tissues appear similar to the human eye. Hyperspectral imaging (HSI) removes this limitation by associating each pixel with high-dimensional spectral information. While previous work has shown its general potential to discriminate tissue, clinical translation has been limited due to the method's current lack of robustness and generalizability. Specifically, the scientific community is lacking a comprehensive spectral tissue atlas, and it is unknown whether variability in spectral reflectance is primarily explained by tissue type rather than the recorded individual or specific acquisition conditions. The contribution of this work is threefold: (1) Based on an annotated medical HSI data set (9059 images from 46 pigs), we present a tissue atlas featuring spectral fingerprints of 20 different porcine organs and tissue types. (2) Using the principle of mixed model analysis, we show that the greatest source of variability related to HSI images is the organ under observation. (3) We show that HSI-based fully-automatic tissue differentiation of 20 organ classes with deep neural networks is possible with high accuracy (> 95%). We conclude from our study that automatic tissue discrimination based on HSI data is feasible and could thus aid in intraoperative decisionmaking and pave the way for context-aware computer-assisted surgery systems and autonomous robotics.

Detection of Tumoral Epithelial Lesions Using Hyperspectral Imaging and Deep Learning

We propose a new method for the analysis and classification of HSI images. The method uses deep learning to interpret the molecular vibrational behaviour of healthy and tumoral human epithelial tissue, based on data gathered via SWIR (short-wave infrared) spectroscopy. We analyzed samples of Melanoma, Dysplastic Nevus and healthy skin. Preliminary results show that human epithelial tissue is sensitive to SWIR to the point of making possible the differentiation between healthy and tumor tissues. We conclude that HSI-SWIR can be used to build new methods for tumor classification.

Hyperspectral interventional imaging for enhanced tissue visualization and discrimination combining band selection methods

PURPOSE

Hyperspectral imaging is an emerging technology recently introduced in medical applications inasmuch as it provides a powerful tool for noninvasive tissue characterization. In this context, a new system was designed to be easily integrated in the operating room in order to detect anatomical tissues hardly noticed by the surgeon's naked eye.

METHOD

Our LCTF-based spectral imaging system is operative over visible, near- and middle-infrared spectral ranges (400-1700 nm). It is dedicated to enhance critical biological tissues such as the ureter and the facial nerve. We aim to find the best three relevant bands to create a RGB image to display during the intervention with maximal contrast between the target tissue and its surroundings. A comparative study is carried out between band selection methods and band transformation methods. Combined band selection methods are proposed. All methods are compared using different evaluation criteria.

RESULTS

Experimental results show that the proposed combined band selection methods provide the best performance with rich information, high tissue separability and short computational time. These methods yield a significant discrimination between biological tissues.

CONCLUSION

We developed a hyperspectral imaging system in order to enhance some biological tissue visualization. The proposed methods provided an acceptable trade-off between the evaluation criteria especially in SWIR spectral band that outperforms the naked eye's capacities.

The pituitary gland under infrared light – in search of a representative spectrum for homogeneous regions

This work focuses on obtaining unique representative FTIR spectrum characteristic for one type of cells architecture. Presented idea is based on using of HCA for data evaluation to search for uniform patterns within samples from the perspective of FTIR spectra.

Medical hyperspectral imaging: a review

Hyperspectral imaging (HSI) is an emerging imaging modality for medical applications, especially in disease diagnosis and image-guided surgery. HSI acquires a three-dimensional dataset called hypercube, with two spatial dimensions and one spectral dimension. Spatially resolved spectral imaging obtained by HSI provides diagnostic information about the tissue physiology, morphology, and composition. This review paper presents an overview of the literature on medical hyperspectral imaging technology and its applications. The aim of the survey is threefold: an introduction for those new to the field, an overview for those working in the field, and a reference for those searching for literature on a specific application.

Medical hyperspectral imaging: a review

Hyperspectral imaging (HSI) is an emerging imaging modality for medical applications, especially in disease diagnosis and image-guided surgery. HSI acquires a three-dimensional dataset called hypercube, with two spatial dimensions and one spectral dimension. Spatially resolved spectral imaging obtained by HSI provides diagnostic information about the tissue physiology, morphology, and composition. This review paper presents an overview of the literature on medical hyperspectral imaging technology and its applications. The aim of the survey is threefold: an introduction for those new to the field, an overview for those working in the field, and a reference for those searching for literature on a specific application.

Progress in Fourier Transform Infrared Spectroscopic Imaging Applied to Venereal Cancer Diagnosis

Fourier transform infrared imaging spectroscopy is a powerful technique that provides molecular and spatial information at the single-cell level. We report on the progress of this technology in the field of cancer research, focusing on human cervical cancer because of the inherent difficulty in grading this type of cancer and as a model for venereal cancers in dogs. Using a suite of multivariate imaging processing techniques, we demonstrate the potential of this technique to identify histologic features in the normal epithelium and cervical intraepithelial neoplasia stages I and III. We highlight the advantages and detail the barriers that need to be overcome before implementation of this technology in the clinical environment.

Evidence of altered biochemical composition in the hearts of adult intrauterine growth-restricted rats

PURPOSE

Epidemiological studies clearly link intrauterine growth restriction with increased risk of cardiac disease in adulthood. The mechanisms leading to this increased risk are poorly understood; remodeling of the myocardium is implicated. The aim was to determine the effect of early life growth restriction on the biochemical composition of the left ventricular myocardium in adult rats.

METHODS

Wistar Kyoto dams were fed either a low protein diet (LPD; 8.7 % casein) or normal protein diet (NPD; 20 % casein) during pregnancy and lactation; from weaning, the offspring were fed normal rat chow. At 18 weeks of age, the biochemical composition of the hearts of NPD control (n = 9) and LPD intrauterine growth-restricted (n = 7) offspring was analyzed using Fourier Transform Infrared (FTIR) micro-spectroscopy.

RESULTS

Body weights at postnatal day 4 were significantly lower and remained lower throughout the experimental period in the LPD offspring compared to controls. FTIR analysis of the infrared absorption spectra across the whole "fingerprint" region (1,800-950 cm(-1)) demonstrated wider variation in absorbance intensity in the LPD group compared to controls. In particular, there were marked differences detected in the protein (1,540 cm(-1)), lipid (1,455 and 1,388 cm(-1)), proteoglycan (1,228 cm(-1)) and carbohydrate (1,038 cm(-1)) bands, indicating increased lipid, proteoglycan and carbohydrate content in the growth-restricted myocardium.

CONCLUSION

In conclusion, changes in the biochemical composition of the myocardium provide a likely mechanism for the increased vulnerability to cardiovascular disease in offspring that were growth restricted in early life.

Synchrotron Radiation Spectroscopic Techniques as Tools for the Medicinal Chemist: Microprobe X-Ray Fluorescence Imaging, X-Ray Absorption Spectroscopy, and Infrared Microspectroscopy

This review updates the recent advances and applications of three prominent synchrotron radiation techniques, microprobe X-ray fluorescence spectroscopy/imaging, X-ray absorption spectroscopy, and infrared microspectroscopy, and highlights how these tools are useful to the medicinal chemist. A brief description of the principles of the techniques is given with emphasis on the advantages of using synchrotron radiation-based instrumentation rather than instruments using typical laboratory radiation sources. This review focuses on several recent applications of these techniques to solve inorganic medicinal chemistry problems, focusing on studies of cellular uptake, distribution, and biotransformation of established and potential therapeutic agents. The importance of using these synchrotron-based techniques to assist the development of, or validate the chemistry behind, drug design is discussed.

Shedding new light on the molecular architecture of oocytes using a combination of synchrotron Fourier transform-infrared and Raman spectroscopic mapping

Synchrotron Fourier transform-infrared (FT-IR) and Raman microspectroscopy were applied to investigate changes in the molecular architecture of mouse oocytes and demonstrate the overall morphology of the maturing oocyte. Here we show that differences were identified between immature mouse oocytes at the germinal vesicle (GV) and mature metaphase II (MII) stage when using this technology, without the introduction of any extrinsic markers, labels, or dyes. GV mouse oocytes were found to have a small, centrally located lipid deposit and another larger polar deposit of similar composition. MII oocytes have very large, centrally located lipid deposits. Each lipid deposit for both cell types contains an inner and outer lipid environment that differs in composition. To assess interoocyte variability, line scans were recorded across the diameter of the oocytes and compared from three independent trials (GV, n = 91; MII, n = 172), and the data were analyzed with principal component analysis (PCA). The average spectra and PCA loading plots show distinct and reproducible changes in the CH stretching region that can be used as molecular maturation markers. The method paves the way for developing an independent assay to assess oocyte status during maturation providing new insights into lipid distribution at the single cell level.

FTIR-based spectroscopic analysis in the identification of clinically aggressive prostate cancer

Fourier transform infrared (FTIR) spectroscopy is a vibrational spectroscopic technique that uses infrared radiation to vibrate molecular bonds within the sample that absorbs it. As different samples contain different molecular bonds or different configurations of molecular bonds, FTIR allows us to obtain chemical information on molecules within the sample. Fourier transform infrared microspectroscopy in conjunction with a principal component-discriminant function analysis (PC-DFA) algorithm was applied to the grading of prostate cancer (CaP) tissue specimens. The PC-DFA algorithm is used alongside the established diagnostic measures of Gleason grading and the tumour/node/metastasis system. Principal component-discriminant function analysis improved the sensitivity and specificity of a three-band Gleason score criterion diagnosis previously reported by attaining an overall sensitivity of 92.3% and specificity of 99.4%. For the first time, we present the use of a two-band criterion showing an association of FTIR-based spectral characteristics with clinically aggressive behaviour in CaP manifest as local and/or distal spread. This paper shows the potential for the use of spectroscopic analysis for the evaluation of the biopotential of CaP in an accurate and reproducible manner.

Focal plane array infrared imaging: a new way to analyse leaf tissue

* Here, a new approach to macromolecular imaging of leaf tissue using a multichannel focal plane array (FPA) infrared detector was compared with the proven method of infrared mapping with a synchrotron source, using transverse sections of leaves from a species of Eucalyptus. * A new histological method was developed, ideally suited to infrared spectroscopic analysis of leaf tissue. Spatial resolution and the signal-to-noise ratio of the FPA imaging and synchrotron mapping methods were compared. * An area of tissue 350 microm(2) required approx. 8 h to map using the synchrotron technique and approx. 2 min to image using the FPA. The two methods produced similar infrared images, which differentiated all tissue types in the leaves according to their macromolecular chemistry. * The synchrotron and FPA methods produced similar results, with the synchrotron method having superior signal-to-noise ratio and potentially better spatial resolution, whereas the FPA method had the advantage in terms of data acquisition time, expense and ease of use. FPA imaging offers a convenient, laboratory-based approach to microscopic chemical imaging of leaves.

A Fourier transform infrared microspectroscopic imaging investigation into an animal model exhibiting glioblastoma multiforme

Glioblastoma multiforme (GBM) is a highly malignant human brain tumour for which no cure is available at present. Numerous clinical studies as well as animal experiments are under way with the goal being to understand tumour biology and develop potential therapeutic approaches. C6 cell glioma in the adult rat is a frequently used and well accepted animal model for the malignant human glial tumour. By combining standard analytical methods such as histology and immunohistochemistry with Fourier Transform Infrared (FTIR) microspectroscopic imaging and multivariate statistical approaches, we are developing a novel approach to tumour diagnosis which allows us to obtain information about the structure and composition of tumour tissues that could not be obtained easily with either method alone. We have used a "Stingray" FTIR imaging spectrometer to analyse and compare the compositions of coronal brain tissue sections of a tumour-bearing animal and those from a healthy animal. We have found that the tumour tissue has a characteristic chemical signature, which distinguishes it from tumour-free brain tissue. The physical-chemical differences, determined by image and spectral comparison are consistent with changes in total protein absorbance, phosphodiester absorbance and physical dispersive artefacts. The results indicate that FTIR imaging analysis could become a valuable analytic method in brain tumour research and possibly in the diagnosis of human brain tumours.