Automated Quantification of Tumor Viability in a Rabbit Liver Tumor Model after Chemoembolization Using Infrared Imaging

Abstract

The rabbit VX2 tumor is a fast-growing carcinoma model commonly used to study new therapeutic devices, such as catheter-based therapies for patients with inoperable hepatocellular carcinoma. The evaluation of tumor viability after such locoregional therapies is essential to directing hepatocellular carcinoma management. We used infrared microspectroscopy for the automatic characterization and quantification of the VX2 liver tumor viability after drug-eluting beads transarterial chemoembolization (DEB-TACE). The protocol consisted of K-means clustering followed by principal component analysis (PCA) and linear discriminant analysis (LDA). The K-means clustering was used to classify the spectra from the infrared images of control or treated tumors and to build a database of many tissue spectra. On the basis of this reference library, the PCA-LDA analysis was used to build a predictive model to identify and quantify automatically tumor viability on unknown tissue sections. For the DEB group, the LDA model determined that the surface of tumor necrosis represented 91.6% ± 8.9% (control group: 33.1% ± 19.6%; Mann-Whitney P = 0.0004) and the viable tumor 2.6% ± 4% (control group: 62.2% ± 15.2%; Mann-Whitney P = 0.0004). Tissue quantification measurements correlated well with tumor necrosis (r = 0.827, P < 0.0001) and viable tumor (r = 0.840, P < 0.0001). Infrared imaging and PCA-LDA analysis could be helpful for easily assessing tumor viability.

Lumican delays melanoma growth in mice and drives tumor molecular assembly as well as response to matrix-targeted TAX2 therapeutic peptide

“…In order to assess subtle changes that may occur in tumor molecular assembly depending on whether host animals do express lumican or not, an innovative method relying on unsupervised K -means clustering of FT-IR spectral images was then used 33 , 34 . Second derivative spectra obtained from s.c. melanoma allografts were partitioned into 9 clusters being representative of tissue biochemical properties 34 , 35 (Fig. 5a and b ).…”

“…With the objective to detect subtle intratumoral biological differences that may not be addressed through conventional histology approaches 33 , 34 , an unsupervised cluster analysis was applied to FT-IR microspectroscopy using the K -means method. For this purpose, we focused on spectra within the 900 to 1800 cm −1 IR absorption range that is considered to be the most informative region as far as the tissue biochemical features are concerned 35 , 76 .…”

Vibrational Analysis of Lung Tumor Cell Lines: Implementation of an Invasiveness Scale Based on the Cell Infrared Signatures

“…In this work, a double EMSC preprocessing proved its capacity for ensuring quality test and homogenizing the set of data in an automated manner. In paraffin-embedded tissue sections, a one-step EMSC is sufficient, on cells samples it was necessary to implement a more sophisticated preprocessing using a specific way to compute the target spectrum and a drastic selection of retained spectra; near 80% of the raw spectra were rejected. Here, the numerous preprocessing parameters were optimized in regards to the classification results, since preprocessing affects strongly the later statistical processing of the data.…”

“…In this work, a double EMSC preprocessing proved its capacity for ensuring quality test and homogenizing the set of data in an automated manner. In paraffin-embedded tissue sections, a one-step EMSC is sufficient, 31 on cells samples it was necessary to implement a more sophisticated preprocessing using a specific a The first and second column presents, respectively, each studied cell culture and the reference invasiveness chosen. The third column presented the mean predicted invasiveness.…”

Fourier Transform Infrared (FT-IR) and Laser Ablation Inductively Coupled Plasma–Mass Spectrometry (LA-ICP-MS) Imaging of Cerebral Ischemia: Combined Analysis of Rat Brain Thin Cuts Toward Improved Tissue Classification

“…Multivariate classifications such as partial least squares discriminant analysis (PLS-DA) or random decision forest (RDF) enable a spectroscopy-based determination of different types of tissue. 15,16 Classification with those methods can be improved when a fused spectral datacube, made up of data obtained using different techniques, is analyzed in contrast to the analysis of individual data sets. 17…”