Salivary ATR-FTIR Spectroscopy Coupled with Support Vector Machine Classification for Screening of Type 2 Diabetes Mellitus

Lipids balance as a spectroscopy marker of diabetes. Analysis of FTIR spectra by 2D correlation and machine learning analyses

The number of people suffering from type 2 diabetes has rapidly increased. Taking into account, that elevated intracellular lipid concentrations, as well as their metabolism, are correlated with diminished insulin sensitivity, in this study we would like to show lipids spectroscopy markers of diabetes. For this purpose, serum collected from rats (animal model of diabetes) was analyzed using Fourier Transformed Infrared-Attenuated Total Reflection (FTIR-ATR) spectroscopy. Analyzed spectra showed that rats with diabetes presented higher concentration of phospholipids and cholesterol in comparison with non-diabetic rats. Moreover, the analysis of second (IInd) derivative spectra showed no structural changes in lipids. Machine learning methods showed higher accuracy for IInd derivative spectra (from 65 % to 89 %) than for absorbance FTIR spectra (53-65 %). Moreover, it was possible to identify significant wavelength intervals from IInd derivative spectra using random forest-based feature selection algorithm, which further increased the accuracy of the classification (up to 92 % for phospholipid region). Moreover decision tree based on the selected features showed, that peaks at 1016 cm-1 and 2936 cm-1 can be good candidates of lipids marker of diabetes.

Non-invasive diagnostic test for lung cancer using biospectroscopy and variable selection techniques in saliva samples

Lung cancer is one of the most commonly occurring malignant tumours worldwide. Although some reference methods such as X-ray, computed tomography or bronchoscope are widely used for clinical diagnosis of lung cancer, there is still a need to develop new methods for early detection of lung cancer. Especially needed are approaches that might be non-invasive and fast with high analytical precision and statistically reliable. Herein, we developed a swab "dip" test in saliva whereby swabs were analysed using attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy harnessed to principal component analysis-quadratic discriminant analysis (QDA) and variable selection techniques employing successive projections algorithm (SPA) and genetic algorithm (GA) for feature selection/extraction combined with QDA. A total of 1944 saliva samples (56 designated as lung-cancer positive and 1888 designed as controls) were obtained in a lung cancer-screening programme being undertaken in North-West England. GA-QDA models achieved, for the test set, sensitivity and specificity values of 100.0% and 99.1%, respectively. Three wavenumbers (1422 cm-1, 1546 cm-1 and 1578 cm-1) were identified using the GA-QDA model to distinguish between lung cancer and controls, including ring C-C stretching, CN adenine, Amide II [δ(NH), ν(CN)] and νs(COO-) (polysaccharides, pectin). These findings highlight the potential of using biospectroscopy associated with multivariate classification algorithms to discriminate between benign saliva samples and those with underlying lung cancer.

A hybrid learning approach to better classify exhaled breath's infrared spectra: A noninvasive optical diagnosis for socially significant diseases

Early diagnosis is crucial for effective treatment of socially significant diseases, such as type 1 diabetes mellitus (T1DM), pneumonia, and asthma. This study employs a diagnostic method based on infrared laser spectroscopy of human exhaled breath. The experimental setup comprises a quantum cascade laser, which emits in a pulsed mode with a peak power of up to 150 mW in the spectral range of 5.3-12.8 μm (780-1890 cm-1), and a Herriott multipass gas cell with a specific optical path length of 76 m. Using this setup, spectra of exhaled breath in the mid-infrared range were obtained from 165 volunteers, including healthy individuals, patients with T1DM, asthma, and pneumonia. The study proposes a hybrid approach for classifying these spectra, utilizing a variational autoencoder for dimensionality reduction and a support vector machine method for classification. The results demonstrate that the proposed hybrid approach outperforms other machine learning method combinations.

A Study of the Association between Primary Oral Pathologies (Dental Caries and Periodontal Diseases) Using Synchrotron Molecular FTIR Spectroscopy in View of the Patient's Personalized Clinical Picture (Demographics and Anamnesis)

In this exploratory study, we searched for associations between the two most common diseases of the oral cavity-dental caries and periodontal diseases-taking into account additional factors, such as personalized clinical pictures (the individual risk factors of the patient), based on the method of a multivariate data analysis of the molecular changes in the composition of human gingival crevicular fluid (GCF). For this purpose, a set of synchrotron Fourier-transform infrared spectroscopy (FTIR) spectra of gingival crevicular fluid samples from patients with different demographics, levels of dental caries development and periodontal diseases, and the presence/absence of concomitant chronic diseases were obtained and analyzed. Using a set of techniques (v-, F-, Chi-square tests; a principal component analysis (PCA); and the hierarchical clustering of principal components (HCPCs)) implemented in the R package FactoMineR allowed us to assess the relationship between the principal components (PCs) and characteristics of the respondents. By identifying the features (vibrational modes in the FTIR spectra) that contribute most to the differentiation of the spectral dataset, and by taking into account the interrelationships between the patients' characteristics, we were able to match specific biological markers (specific molecular groups) to the two factors of interest-two types of oral pathologies. The results obtained show that the observed changes in the quantitative and qualitative composition of the modes in the infrared (IR) spectra of the GCF samples from patients with different dental caries developments and periodontal diseases present confirm the difficulty of identifying patient-specific spectral information. At the same time, different periodontal pathologies are more closely associated with other characteristics of the patients than the level of their caries development. The multivariate analysis performed on the spectral dataset indicates the need to take into account not only the co-occurrence of oral diseases, but also some other factors. The lack of this consideration (typical in lots of studies in this area) may lead to misinterpretations and consequently to a loss of data when searching for biological markers of certain oral diseases.

Infrared spectroscopy for fast screening of diabetes and periodontitis

SIGNIFICANCE

FT-IR is an important and emerging tool, providing information related to the biochemical composition of biofluids. It is important to demonstrate that there is an efficacy in separating healthy and diseased groups, helping to establish FT-IR uses as fast screening tool.

AIM

Via saliva diagnosis evaluate the accuracy of FT-IR associate with machine learning model for classification among healthy (control group), diabetic (D) and periodontitis (P) patients and the association of both diseases (DP).

APPROACH

Eighty patients diagnosed with diabetes and periodontitis through conventional methods were recruited and allocated in one of the four groups. Saliva samples were collected from participants of each group (n = 20) and were processed using Bruker Alpha II spectrometer in a FT-IR spectral fingerprint region between 600 and-1800 cm-1, followed by data preprocessing and analysis using machine learning tools.

RESULTS

Various FTI-R peaks were detectable and attributed to specific vibrational modes, which were classified based on confusion matrices showed in paired groups. The highest true positive rates (TPR) appeared between groups C vs D (93.5 % ± 2.7 %), groups C vs. DP (89.2 % ± 4.1 %), and groups D and P (90.4 % ± 3.2 %). However, P vs DP presented higher TPR for DP (84.1 % ±3.1 %) while D vs. DP the highest rate for DP was 81.7 % ± 4.3 %. Analyzing all groups together, the TPR decreased.

CONCLUSION

The system used is portable and robust and can be widely used in clinical environments and hospitals as a new diagnostic technique. Studies in our groups are being conducted to solidify and expand data analysis methods with friendly language for healthcare professionals. It was possible to classify healthy patients in a range of 78-93 % of accuracy. Range over 80 % of accuracy between periodontitis and diabetes were observed. A general classification model with lower TPR instead of a pairwise classification would only have advantages in scenarios where no prior patient information is available regarding diabetes and periodontitis status.

CSSLdb: Discovery of cancer-specific synthetic lethal interactions based on machine learning and statistic inference

Synthetic lethality (SL) occurs when the inactivation of two genes results in cell death while the inactivation of either gene alone is non-lethal. SL-based therapy has become a promising anti-cancer treatment option with the increasing researches and applications in clinical practice, while the specific therapeutic opportunities for various cancers have not yet been comprehensively investigated. Herein, we described a computational approach based on machine learning and statistical inference to discover the cancer-specific synthetic lethal interactions. First, Random Forest and One-Class SVM were used to perform cancer unbiased prediction of synthetic lethality. Then, two strategies, including mutual exclusivity and differential expression, were used to screen cancer-specific synthetic lethal interactions, resulting in 14,582 SL gene pairs in 33 cancer types. Finally, we developed a freely available database of CSSLdb (Cancer Specific Synthetic Lethality Database, http://www.tmliang.cn/CSSL/) to present cancer-specific synthetic lethal genetic interactions, which would enrich the relevant research and contribute to underlying therapy strategies based on synthetic lethality.

Translating Biospectroscopy Techniques to Clinical Settings: A New Paradigm in Point-of-Care Screening and/or Diagnostics

As healthcare tools increasingly move towards a more digital and computational format, there is an increasing need for sensor-based technologies that allow for rapid screening and/or diagnostics [...].

Prediction of type 1 diabetes with machine learning algorithms based on FTIR spectral data in peripheral blood mononuclear cells

The incidence of autoimmunity is increasing, to ensure timely and comprehensive treatment, there must be a diagnostic method or markers that would be available to the general public. Fourier-transform infrared spectroscopy (FTIR) is a relatively inexpensive and accurate method for determining metabolic fingerprint. The metabolism, molecular composition and function of blood cells vary according to individual physiological and pathological conditions. Thus, by obtaining autoimmune disease-specific metabolic fingerprint markers in peripheral blood mononuclear cells (PBMC) and subsequently using machine learning algorithms, it might be possible to create a tool that will allow the diagnosis of autoimmune diseases. In this preliminary study, it was found that the peak shift at 1545 cm-1 could be considered specific for autoimmune disease type 1 diabetes (T1D), while the shifts at 1070 and 1417 cm-1 could be more attributed to the autoimmune condition per se. The prediction of T1D, despite the small number of participants in the study, showed an inverse AUC = 0.33 ± 0.096, n = 15, indicating a stable trend in the prediction of T1D based on FTIR metabolic fingerprint data in the PBMC.