Advanced statistical techniques applied to comprehensive FTIR spectra on human colonic tissues

Classification of healthy and cancerous colon cells by Fourier transform infrared spectroscopy

Colorectal cancer is one of the most diagnosed types of cancer in developed countries. Current diagnostic methods are partly dependent on pathologist experience and laboratories instrumentation. In this study, we used Fourier Transform Infrared (FTIR) spectroscopy in transflection mode, combined with Principal Components Analysis followed by Linear Discriminant Analysis (PCA-LDA) and Partial Least Squares - Discriminant Analysis (PLS-DA), to build a classification algorithm to diagnose colon cancer in cell samples, based on absorption spectra measured in two spectral ranges of the mid-infrared spectrum. In particular, PCA technique highlights small biochemical differences between healthy and cancerous cells: these are related to the larger lipid content in the former compared with the latter and to the larger relative amount of protein and nucleic acid components in the cancerous cells compared with the healthy ones. Comparison of the classification accuracy of PCA-LDA and PLS-DA methods applied to FTIR spectra measured in the 1000-1800 cm-1 (low wavenumber range, LWR) and 2700-3700 cm-1 (high wavenumber range, HWR) remarks that both algorithms are able to classify hidden class FTIR spectra with excellent accuracy (100 %) in both spectral regions. This is a hopeful result for clinical translation of infrared spectroscopy: in fact, it makes reliable the predictions obtained using FTIR measurements carried out only in the HWR, in which the glass slides used in clinical laboratories are transparent to IR radiation.

Identification of remodeled collagen fibers in tumor stroma by FTIR Micro-spectroscopy: A new approach to recognize the colon carcinoma

The tumor stroma plays a pivotal role in colon cancer genesis and progression. It was observed that collagen fibers in the extracellular matrix (ECM) of cancer stroma, undergo a strong remodeling. These fibrous proteins result more aligned and compact than in physiological conditions, creating a microenvironment that favors cancer development. In this work, micro-FTIR spectroscopy was applied to investigate the chemical modifications in the tumor stroma. Using Fuzzy C-means clustering, mean spectra from diseased and normal stroma were compared and collagen was found to be responsible for the main differences between them. Specifically, the modified absorptions at 1203, 1238, 1284 cm^-1 and 1338 cm^-1 wavenumbers, were related to the amide III band and CH₂ bending of side chains. These signals are sensitive to the interactions between the α-chains in the triple helices of collagen structure. This provided robust chemical evidence that in cancer ECM, collagen fibers are more parallelized, stiff and ordered than in normal tissue. Principal Component Analysis (PCA) applied to the spectra from malignant and normal stroma confirmed these findings. Using LDA (Linear Discriminant Analysis) classification, the absorptions 1203, 1238, 1284 and 1338 cm^-1 were examined as spectral biomarkers, obtaining quite promising results. The use of a PCA-LDA prediction model on samples with moderate tumor degree further showed that the stroma chemical modifications are more indicative of malignancy compared to the epithelium. These preliminary findings have shown that micro-FTIR spectroscopy, focused on collagen signals, could become a promising tool for colon cancer diagnosis.

Empirical study on the effects of acquisition parameters for FTIR hyperspectral imaging of brain tissue

Fourier transform infrared (FTIR) spectroscopic imaging is a powerful technique for molecular imaging of pathologies associated with the nervous systems including multiple sclerosis research. However, there is no standard methodology or standardized protocol for FTIR imaging of tissue sections that maximize the ability to discriminate between the molecular, white and granular layers, which is essential in the investigation of the mechanism of demyelination process. Tissue sections are heterogeneous, complex and delicate, hence the parameters to generate high quality images in minimal time becomes essential in the modern clinical laboratory. This article presents an FTIR spectroscopic imaging study of post-mortem human brain tissue testing the effects of various measurement parameters and data analysis methods on image quality and acquisition time. Hyperspectral images acquired from the same region of a tissue using a range of the most common optical and collection parameters in different combinations were compared. These included magnification (4× and 15×), number of co-added scans (1, 4, 8, 16, 32, 64 and 128 scans) and spectral resolution (4, 8 and 16 cm-1). Images were compared in terms of acquisition time, signal-to-noise (S/N) ratio, and accuracy of the discrimination between three major tissue types in a section from the cerebellum (white matter, granular and molecular layers). In the latter case, unsupervised k-means cluster (KMC) analysis was employed to generate images from the hyperspectral images, which were compared to a reference image. The classification accuracy for tissue class discrimination was highest for the 4× magnifying objective, with 4 cm-1 spectral resolution and 128 co-added scans. The 15× magnifying objective gave the best accuracy for a spectral resolution of 4 cm-1 and 64 scans (96.3%), which was just above what was achieved using the 4× magnifying objective, with 4 cm-1 spectral resolution and 32 and 64 co-added scans (95.4 and 95.6%, respectively). These findings were correlated with a decrease in S/N ratio with increasing number of scans and was generally lower for the 15× objective. However, longer scan times were required using the 15× magnifying objective, which did not justify the very small improvement in the classification of tissue types.

Differentiation of Pectobacterium and Dickeya spp. phytopathogens using infrared spectroscopy and machine learning analysis

Pectobacterium and Dickeya spp. are soft rot Pectobacteriaceae that cause aggressive diseases on agricultural crops leading to substantial economic losses. The accurate, rapid and low-cost detection of these pathogenic bacteria are very important for controlling their spread, reducing the consequent financial loss and for producing uninfected potato seed tubers for future generations. Currently used methods for the identification of these bacterial pathogens at the strain level are based mainly on molecular techniques, which are expensive. We used an alternative method, infrared spectroscopy, to measure 24 strains of five species of Pectobacterium and Dickeya. Measurements were then analyzed using machine learning methods to differentiate among them at the genus, species and strain levels. Our results show that it is possible to differentiate among different bacterial pathogens with a success rate of ~99% at the genus and species levels and with a success rate of over 94% at the strain level.

Potential of infrared microscopy to differentiate between dementia with Lewy bodies and Alzheimer's diseases using peripheral blood samples and machine learning algorithms

SIGNIFICANCE

Accurate and objective identification of Alzheimer's disease (AD) and dementia with Lewy bodies (DLB) is of major clinical importance due to the current lack of low-cost and noninvasive diagnostic tools to differentiate between the two. Developing an approach for such identification can have a great impact in the field of dementia diseases as it would offer physicians a routine objective test to support their diagnoses. The problem is especially acute because these two dementias have some common symptoms and characteristics, which can lead to misdiagnosis of DLB as AD and vice versa, mainly at their early stages.

AIM

The aim is to evaluate the potential of mid-infrared (IR) spectroscopy in tandem with machine learning algorithms as a sensitive method to detect minor changes in the biochemical structures that accompany the development of AD and DLB based on a simple peripheral blood test, thus improving the diagnostic accuracy of differentiation between DLB and AD.

APPROACH

IR microspectroscopy was used to examine white blood cells and plasma isolated from 56 individuals: 26 controls, 20 AD patients, and 10 DLB patients. The measured spectra were analyzed via machine learning.

RESULTS

Our encouraging results show that it is possible to differentiate between dementia (AD and DLB) and controls with an ∼86  %   success rate and between DLB and AD patients with a success rate of better than 93%.

CONCLUSIONS

The success of this method makes it possible to suggest a new, simple, and powerful tool for the mental health professional, with the potential to improve the reliability and objectivity of diagnoses of both AD and DLB.

Early diagnosis of Alzheimer's disease using infrared spectroscopy of isolated blood samples followed by multivariate analyses

Alzheimer's disease (AD) is the most common cause of dementia, particularly in the elderly. The disease is characterized by cognitive decline that typically starts with insidious memory loss and progresses relentlessly to produce global impairment of all higher cortical functions. Due to better living conditions and health facilities in developed countries, which result in higher overall life spans, these countries report upward trends of AD among their populations. There are, however, no specific diagnostic tests for AD and clinical diagnosis is especially difficult in the earliest stages of the disease. Early diagnosis of AD is frequently subjective and is determined by physicians (generally neurologists, geriatricians, and psychiatrists) depending on their experience. Diagnosing AD requires both medical history and mental status testing. Having trouble with memory does not mean you have AD. AD has no current cure, but treatments for symptoms are available and research continues. In this study, we investigated the potential of infrared microscopy to differentiate between AD patients and controls, using Fourier transform infrared (FTIR) spectroscopy of isolated blood components. FTIR is known as a quick, safe, and minimally invasive method to investigate biological samples. For this goal, we measured infrared spectra from white blood cells (WBCs) and plasma taken from AD patients and controls, with the consent of the patients or their guardians. Applying multivariate analysis, principal component analysis (PCA) followed by linear discriminant analysis (LDA), it was possible to differentiate among the different types of mild, moderate, and severe AD, and the controls, with 85% accuracy when using the WBC spectra and about 77% when using the plasma spectra. When only the moderate and severe stages were included, an 83% accuracy was obtained using the WBC spectra and about 89% when using the plasma spectra.

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

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.

Using Infrared Spectroscopy and Multivariate Analysis to Detect Antibiotics' Resistant Escherichia coli Bacteria

Bacterial pathogens are one of the primary causes of human morbidity worldwide. Historically, antibiotics have been highly effective against most bacterial pathogens; however, the increasing resistance of bacteria to a broad spectrum of commonly used antibiotics has become a global health-care problem. Early and rapid determination of bacterial susceptibility to antibiotics has become essential in many clinical settings and, sometimes, can save lives. Currently classical procedures require at least 48 h for determining bacterial susceptibility, which can constitute a life-threatening delay for effective treatment. Infrared (IR) microscopy is a rapid and inexpensive technique, which has been used successfully for the detection and identification of various biological samples; nonetheless, its true potential in routine clinical diagnosis has not yet been established. In this study, we evaluated the potential of this technique for rapid identification of bacterial susceptibility to specific antibiotics based on the IR spectra of the bacteria. IR spectroscopy was conducted on bacterial colonies, obtained after 24 h culture from patients' samples. An IR microscope was utilized, and a computational classification method was developed to analyze the IR spectra by novel pattern-recognition and statistical tools, to determine E. coli susceptibility within a few minutes to different antibiotics, gentamicin, ceftazidime, nitrofurantoin, nalidixic acid, ofloxacin. Our results show that it was possible to classify the tested bacteria into sensitive and resistant types, with success rates as high as 85% for a number of examined antibiotics. These promising results open the potential of this technique for faster determination of bacterial susceptibility to certain antibiotics.

Verification of the effectiveness of the Fourier transform infrared spectroscopy computational model for colorectal cancer

Colorectal cancer is one of the most common cancers. Its formation is influenced by genetic and environmental factors. Despite the continuous development of diagnostic tools and cancer therapies, there are no methods that allow a real-time estimation of treatment efficiency. This method can be a vibrational spectroscopy. The resulting infrared spectrum (FTIR) of the tissue gives us information about the chemical composition and the content of the individual components. We have noticed that tumor tissues, healthy and after chemotherapy tissues, have different vibrational spectra. It was also shown that spectra acquired from normal (benign) tissues were similar to those derived from tissues post-chemotherapy. The similarity was greater, when the effectiveness of chemotherapy, confirmed by medical documentation, was better. Therefore, we decided to use the physical model proposed in our earlier paper to verify its correctness and to show whether a particular type of chemotherapy was effective or not. Comparison of the results obtained from the physical model with patients data have been found as close to the physical condition.

Detection of antibiotic resistant Escherichia Coli bacteria using infrared microscopy and advanced multivariate analysis

DeterminingE. colibacteria susceptibility by analyzing their FTIR spectra using multivariate analysis.

Use of FTIR spectroscopy and PCA-LDC analysis to identify cancerous lesions within the human colon

Colorectal cancer constitutes 33% of all cancer morbidity, so the research of the new methods for colorectal cancer diagnosis and chemotherapy monitoring is gaining its momentum. Diagnostic instruments are being sought, which enable the detection of single malignant cells based on the analysis of tissue material potentially reusable at further stages of diagnostic management. The most common approach to tissue specimen processing is paraffin-embedding. Yet, paraffin may cause background noise in spectroscopic measurements with the wavenumber ranging between 900cm^-1 and 3500cm^-1. However, the study by Depciuch et al. (2016) proved that appropriate specimen processing and paraffin-embedding technique as well as a strict measurement methodology may eliminate paraffin vibrations. As a result, spectroscopic measurements may become a reliable and precise method for the diagnosis and treatment monitoring in patients with colorectal cancer as long as the high standards of specimen processing are maintained. Chemotherapy is the main medical treatment in colorectal cancer. Unfortunately, the absence of tools which enable monitoring its efficacy leads to the partial response or non-response frequently seen in affected patients. Hence, diagnostic instruments are also being sought capable of monitoring treatment efficacy so as to enable early changes of chemotherapy regimen thus increasing the chance of cure. The paper aims at comparing the results of FTIR (Fourier Transform Infrared) spectroscopy in several types of colon tissue: healthy colon, cancerous colon, post-chemotherapy colon and healthy surgical margin of colon cancer sample. The obtained FTIR spectra along with the Principal Component Analysis-Linear Discriminant Analysis (PCA-LDC) as well as bandwidth analysis of the primary amide region revealed some differences between the spectra of healthy tissues as compared to cancerous tissues (pre- or post-chemotherapy). Apart from confirming that FTIR spectroscopy is a good source of information on the composition of analysed samples, this fact supports its application as a tool to facilitate understanding the pathophysiology of various conditions and to monitor efficacy of chemotherapy in cancer patients.

Renal Graft Fibrosis and Inflammation Quantification by an Automated Fourier-Transform Infrared Imaging Technique

Renal interstitial fibrosis and interstitial active inflammation are the main histologic features of renal allograft biopsy specimens. Fibrosis is currently assessed by semiquantitative subjective analysis, and color image analysis has been developed to improve the reliability and repeatability of this evaluation. However, these techniques fail to distinguish fibrosis from constitutive collagen or active inflammation. We developed an automatic, reproducible Fourier-transform infrared (FTIR) imaging-based technique for simultaneous quantification of fibrosis and inflammation in renal allograft biopsy specimens. We generated and validated a classification model using 49 renal biopsy specimens and subsequently tested the robustness of this classification algorithm on 166 renal grafts. Finally, we explored the clinical relevance of fibrosis quantification using FTIR imaging by comparing results with renal function at 3 months after transplantation (M3) and the variation of renal function between M3 and M12. We showed excellent robustness for fibrosis and inflammation classification, with >90% of renal biopsy specimens adequately classified by FTIR imaging. Finally, fibrosis quantification by FTIR imaging correlated with renal function at M3, and the variation in fibrosis between M3 and M12 correlated well with the variation in renal function over the same period. This study shows that FTIR-based analysis of renal graft biopsy specimens is a reproducible and reliable label-free technique for quantifying fibrosis and active inflammation. This technique seems to be more relevant than digital image analysis and promising for both research studies and routine clinical practice.

Detection of cancer using advanced computerized analysis of infrared spectra of peripheral blood

We have developed a novel approach for detection of cancer based on biochemical analysis of peripheral blood plasma using Fourier transform infrared spectroscopy. This approach has proven to be quick, safe, minimal invasive, and effective. Our approach recognizes any signs of solid tumor presence, regardless of location in the body or cancer type by measuring a spectrum that gives information regarding the total molecular composition and structure of the peripheral blood samples. The analysis includes clinically relevant preprocessing and feature extraction with principal component analysis, and uses Fisher's linear discriminant analysis to classify between cancer patients and healthy controls. We evaluated our method with leave-one-out cross validation and were able to establish sensitivity of 93.33%, specificity of 87.8%, and overall accuracy of 90.7%. Using our method for cancer detection should result in fewer unnecessary invasive procedures and yield fast detection of solid tumors.

Chemometrics of differentially expressed proteins from colorectal cancer patients

AIM: To evaluate the usefulness of differentially expressed proteins from colorectal cancer (CRC) tissues for differentiating cancer and normal tissues.

METHODS: A Proteomic approach was used to identify the differentially expressed proteins between CRC and normal tissues. The proteins were extracted using Tris buffer and thiourea lysis buffer (TLB) for extraction of aqueous soluble and membrane-associated proteins, respectively. Chemometrics, namely principal component analysis (PCA) and linear discriminant analysis (LDA), were used to assess the usefulness of these proteins for identifying the cancerous state of tissues.

RESULTS: Differentially expressed proteins identified were 37 aqueous soluble proteins in Tris extracts and 24 membrane-associated proteins in TLB extracts. Based on the protein spots intensity on 2D-gel images, PCA by applying an eigenvalue > 1 was successfully used to reduce the number of principal components (PCs) into 12 and seven PCs for Tris and TLB extracts, respectively, and subsequently six PCs, respectively from both the extracts were used for LDA. The LDA classification for Tris extract showed 82.7% of original samples were correctly classified, whereas 82.7% were correctly classified for the cross-validated samples. The LDA for TLB extract showed that 78.8% of original samples and 71.2% of the cross-validated samples were correctly classified.

CONCLUSION: The classification of CRC tissues by PCA and LDA provided a promising distinction between normal and cancer types. These methods can possibly be used for identification of potential biomarkers among the differentially expressed proteins identified.

Spectral signatures of colonic malignancies in the mid-infrared region: from basic research to clinical applicability

The process of carcinogenesis in the colon progresses through several overlapping stages, making the evaluation process challenging, as well as subjective. Owing to the complexity of colonic tissues and the search for a technique that is rapid and foolproof for precise grading and evaluation of biopsies, many spectroscopic techniques have been evaluated in the past few decades for their efficiency and clinical compatibility. Fourier-transform infrared spectroscopy, being quantitative and objective, has the capacity for automation and relevance to cancer diagnosis. This article highlights investigations on the application of Fourier-transform infrared spectroscopy (particularly microscopy) in colon cancer diagnosis and parallel developments in data analysis techniques for the characterization of spectral signatures of malignant tissues in the colon.