An application of raman spectroscopy in combination with machine learning to determine gastric cancer spectroscopy marker

Detection of lung cancer through SERS analysis of serum

Early detection and postoperative assessment are crucial for improving overall survival among lung cancer patients. Here, we report a non-invasive technique that integrates Raman spectroscopy with machine learning for the detection of lung cancer. The study encompassed 88 postoperative lung cancer patients, 73 non-surgical lung cancer patients, and 68 healthy subjects. The primary aim was to explore variations in serum metabolism across these cohorts. Comparative analysis of average Raman spectra was conducted, while principal component analysis was employed for data visualization. Subsequently, the augmented dataset was used to train convolutional neural networks (CNN) and Resnet models, leading to the development of a diagnostic framework. The CNN model exhibited superior performance, as verified by the receiver operating characteristic curve. Notably, postoperative patients demonstrated an increased likelihood of recurrence, emphasizing the crucial need for continuous postoperative monitoring. In summary, the integration of Raman spectroscopy with CNN-based classification shows potential for early detection and postoperative assessment of lung cancer.

Enhanced PSO feature selection with Runge-Kutta and Gaussian sampling for precise gastric cancer recurrence prediction

Gastric cancer (GC), characterized by its inconspicuous initial symptoms and rapid invasiveness, presents a formidable challenge. Overlooking postoperative intervention opportunities may result in the dissemination of tumors to adjacent areas and distant organs, thereby substantially diminishing prospects for patient survival. Consequently, the prompt recognition and management of GC postoperative recurrence emerge as a matter of paramount urgency to mitigate the deleterious implications of the ailment. This study proposes an enhanced feature selection model, bRSPSO-FKNN, integrating boosted particle swarm optimization (RSPSO) with fuzzy k-nearest neighbor (FKNN), for predicting GC. It incorporates the Runge-Kutta search, for improved model accuracy, and Gaussian sampling, enhancing the search performance and helping to avoid locally optimal solutions. It outperforms the sophisticated variants of particle swarm optimization when evaluated in the CEC 2014 test suite. Furthermore, the bRSPSO-FKNN feature selection model was introduced for GC recurrence prediction analysis, achieving up to 82.082 % and 86.185 % accuracy and specificity, respectively. In summation, this model attains a notable level of precision, poised to ameliorate the early warning system for GC recurrence and, in turn, advance therapeutic options for afflicted patients.

Development and Validation of a Novel Machine Learning Model to Predict the Survival of Patients with Gastrointestinal Neuroendocrine Neoplasms

INTRODUCTION

Well-calibrated models for personalized prognostication of patients with gastrointestinal neuroendocrine neoplasms (GINENs) are limited. This study aimed to develop and validate a machine-learning model to predict the survival of patients with GINENs.

METHODS

Oblique random survival forest (ORSF) model, Cox proportional hazard risk model, Cox model with least absolute shrinkage and selection operator penalization, CoxBoost, Survival Gradient Boosting Machine, Extreme Gradient Boosting survival regression, DeepHit, DeepSurv, DNNSurv, logistic-hazard model, and PC-hazard model were compared. We further tuned hyperparameters and selected variables for the best-performing ORSF. Then, the final ORSF model was validated.

RESULTS

A total of 43,444 patients with GINENs were included. The median (interquartile range) survival time was 53 (19-102) months. The ORSF model performed best, in which age, histology, M stage, tumor size, primary tumor site, sex, tumor number, surgery, lymph nodes removed, N stage, race, and grade were ranked as important variables. However, chemotherapy and radiotherapy were not necessary for the ORSF model. The ORSF model had an overall C index of 0.86 (95% confidence interval, 0.85-0.87). The area under the receiver operation curves at 1, 3, 5, and 10 years were 0.91, 0.89, 0.87, and 0.80, respectively. The decision curve analysis showed superior clinical usefulness of the ORSF model than the American Joint Committee on Cancer Stage. A nomogram and an online tool were given.

CONCLUSION

The machine learning ORSF model could precisely predict the survival of patients with GINENs, with the ability to identify patients at high risk for death and probably guide clinical practice.

Rapid diagnosis of systemic lupus erythematosus by Raman spectroscopy combined with spiking neural network

Multiple organs are affected by the autoimmune inflammatory connective tissue disease known as systemic lupus erythematosus (SLE). If not diagnosed and treated in a timely manner, it can lead to nephritis and damage to the blood system in severe cases, resulting in the patient's death. Therefore, correct and timely diagnosis and treatment are essential for patients. In this study, a framework based on neural network algorithm and Raman spectroscopy technique was established to diagnose SLE patients. Firstly, we pre-processed the obtained Raman data by three methods: baseline correction, smoothing processing and normalization methods, before using it as input for the model, and then ANN, ResNet and SNN classification models were established. The respective classification accuracies for SLE patients were 89.61%, 85.71%, and 95.65% for the three models, with corresponding AUC values of 0.8772, 0.8100, and 0.9555. The results of the experimental indicate that SNN possesses a good classification effect, and the number of model parameters is only 525,826, which is 414,221 less than that of ResNet model. Since the network only uses 0 and 1 to transmit information, and only has basic operations such as summation, compared with the second-generation artificial neural network, which simplifies the product operation of floating point numbers into multiple addition operations, the network has low energy consumption and is suitable for embedding portable Raman spectrometer for clinical diagnosis. This research highlights the significant potential for quick and precise SLE patient discrimination offered by Raman spectroscopy in conjunction with spiking neural networks.

A dense SERS substrate of the AgNPs@GO compound film for detecting homocysteine molecules

This study focuses on the development of a highly sensitive surface-enhanced Raman scattering (SERS) sensor for detecting homocysteine (Hcy) molecules. The Hcy sensor was created by depositing silver nanoparticles (AgNPs) onto the surface of graphene oxide (GO) film to form a dense AgNPs@GO composite film. The AgNPs on the composite film interacted with sulfur atoms (S) of Hcy molecules to form Ag-S bonds, which boosted the chemisorption of Hcy molecules and enabled them to be specifically recognized. The SERS sensor exhibited a maximum enhancement factor of up to 1.1 × 104, with a reliable linear response range from 1 to 60 ng mL-1. The limit of detection (LOD) for Hcy molecules was as low as 1.1 × 10-9 M. Moreover, Hcy molecules were successfully distinguished in a mixed solution of γ-aminobutyric acid and Hcy molecules. In this study, a simple preparation process of SERS substrate and a novel detection method for Hcy molecules provided a new pathway for the rapid and effective detection of Hcy molecules in the food and biomedicine fields.

Detection of primary myelofibrosis in blood serum via Raman spectroscopy assisted by machine learning approaches; correlation with clinical diagnosis

Primary myelofibrosis (PM) is one of the myeloproliferative neoplasm, where stem cell-derived clonal neoplasms was noticed. Diagnosis of this disease is based on: physical examination, peripheral blood findings, bone marrow morphology, cytogenetics, and molecular markers. However, the molecular marker of PM, which is a mutation in the JAK2V617F gene, was observed also in other myeloproliferative neoplasms such as polycythemia vera and essential thrombocythemia. Therefore, there is a need to find methods that provide a marker unique to PM and allow for higher accuracy of PM diagnosis and consequently the treatment of the disease. Continuing, in this study, we used Raman spectroscopy, Principal Components Analysis (PCA), and Partial Least Squares (PLS) analysis as helpful diagnostic tools for PM. Consequently, we used serum collected from PM patients, which were classified using clinical parameters of PM such as the dynamic international prognostic scoring system (DIPSS) for primary myelofibrosis plus score, the JAK2V617F mutation, spleen size, bone marrow reticulin fibrosis degree and use of hydroxyurea drug features. Raman spectra showed higher amounts of C-H, C-C and C-C/C-N and amide II and lower amounts of amide I and vibrations of CH3 groups in PM patients than in healthy ones. Furthermore, shifts of amides II and I vibrations in PM patients were noticed. Machine learning methods were used to analyze Raman regions: (i) 800 cm-1 and 1800 cm-1, (ii) 1600 cm-1-1700 cm-1, and (iii) 2700 cm-1-3000 cm-1 showed 100 % accuracy, sensitivity, and specificity. Differences in the spectral dynamic showed that differences in the amide II and amide I regions were the most significant in distinguishing between PM and healthy subjects. Importantly, until now, the efficacy of Raman spectroscopy has not been established in clinical diagnostics of PM disease using the correlation between Raman spectra and PM clinical prognostic scoring. Continuing, our results showed the correlation between Raman signals and bone marrow fibrosis, as well as JAKV617F. Consequently, the results revealed that Raman spectroscopy has a high potential for use in medical laboratory diagnostics to quantify multiple biomarkers simultaneously, especially in the selected Raman regions.

Application of the Fuzzy Approach for Evaluating and Selecting Relevant Objects, Features, and Their Ranges

Relevant attribute selection in machine learning is a key aspect aimed at simplifying the problem, reducing its dimensionality, and consequently accelerating computation. This paper proposes new algorithms for selecting relevant features and evaluating and selecting a subset of relevant objects in a dataset. Both algorithms are mainly based on the use of a fuzzy approach. The research presented here yielded preliminary results of a new approach to the problem of selecting relevant attributes and objects and selecting appropriate ranges of their values. Detailed results obtained on the Sonar dataset show the positive effects of this approach. Moreover, the observed results may suggest the effectiveness of the proposed method in terms of identifying a subset of truly relevant attributes from among those identified by traditional feature selection methods.

FTIR- based serum structure analysis in molecular diagnostics of essential thrombocythemia disease

Essential thrombocythemia (ET) reflects the transformation of a multipotent hematopoietic stem cell, but its molecular pathogenesis remains obscure. Nevertheless, tyrosine kinase, especially Janus kinase 2 (JAK2), has been implicated in myeloproliferative disorders other than chronic myeloid leukaemia. FTIR analysis was performed on the blood serum of 86 patients and 45 healthy volunteers as control with FTIR spectra-based machine learning methods and chemometrics. Thus, the study aimed to determine biomolecular changes and separation of ET and healthy control groups illustration by applying chemometrics and ML techniques to spectral data. The FTIR-based results showed that in ET disease with JAK2 mutation, there are alterations in functional groups associated with lipids, proteins and nucleic acids significantly. Moreover, in ET patients the lower amount of proteins with simultaneously higher amount of lipids was noted in comparison with the control one. Furthermore, the SVM-DA model showed 100% accuracy in calibration sets in both spectral regions and 100.0% and 96.43% accuracy in prediction sets for the 800-1800 cm^-1 and 2700-3000 cm^-1 spectral regions, respectively. While changes in the dynamic spectra showed that CH₂ bending, amide II and CO vibrations could be used as a spectroscopy marker of ET. Finally, it was found a positive correlation between FTIR peaks and first bone marrow fibrosis degree, as well as the absence of JAK2 V617F mutation. The findings of this study contribute to a better understanding of the molecular pathogenesis of ET and identifying biomolecular changes and may have implications for early diagnosis and treatment of this disease.

Label-Free Quantification of Nanoencapsulated Piperonyl Esters in Cosmetic Hydrogels Using Raman Spectroscopy

Raman spectroscopy is a well-established technique for the molecular characterisation of samples and does not require extensive pre-analytical processing for complex cosmetic products. As an illustration of its potential, this study investigates the quantitative performance of Raman spectroscopy coupled with partial least squares regression (PLSR) for the analysis of Alginate nanoencapsulated Piperonyl Esters (ANC-PE) incorporated into a hydrogel. A total of 96 ANC-PE samples covering a 0.4% w/w-8.3% w/w PE concentration range have been prepared and analysed. Despite the complex formulation of the sample, the spectral features of the PE can be detected and used to quantify the concentrations. Using a leave-K-out cross-validation approach, samples were divided into a training set (n = 64) and a test set, samples that were previously unknown to the PLSR model (n = 32). The root mean square error of cross-validation (RMSECV) and prediction (RMSEP) was evaluated to be 0.142% (w/w PE) and 0.148% (w/w PE), respectively. The accuracy of the prediction model was further evaluated by the percent relative error calculated from the predicted concentration compared to the true value, yielding values of 3.58% for the training set and 3.67% for the test set. The outcome of the analysis demonstrated the analytical power of Raman to obtain label-free, non-destructive quantification of the active cosmetic ingredient, presently PE, in complex formulations, holding promise for future analytical quality control (AQC) applications in the cosmetics industry with rapid and consumable-free analysis.

Biochemical differentiation between cancerous and normal human colorectal tissues by micro-Raman spectroscopy

Human colorectal tissues obtained by ten cancer patients have been examined by multiple micro-Raman spectroscopic measurements in the 500-3200 cm^-1 range under 785 nm excitation. Distinct spectral profiles are recorded from different spots on the samples: a predominant 'typical' profile of colorectal tissue, as well as those from tissue topologies with high lipid, blood or collagen content. Principal component analysis identified several Raman bands of amino acids, proteins and lipids which allow the efficient discrimination of normal from cancer tissues, the first presenting plurality of Raman spectral profiles while the last showing off quite uniform spectroscopic characteristics. Tree-based machine learning experiment was further applied on all data as well as on filtered data keeping only those spectra which characterize the largely inseparable data clusters of 'typical' and 'collagen-rich' spectra. This purposive sampling evidences statistically the most significant spectroscopic features regarding the correct identification of cancer tissues and allows matching spectroscopic results with the biochemical changes induced in the malignant tissues.

Raman spectroscopy-based biomarker screening by studying the fingerprint and lipid characteristics of Polycythemıa Vera cases blood serum

This study aimed to develop a novel approach for diagnosing Polycythemia Vera (PV), a stem cell-derived neoplasm of the myeloid lineage. The approach utilized Raman spectroscopy coupled with multivariate analysis to analyze blood serum samples collected from PV patients. The results showed that PV serum exhibited lower protein and lipid levels and structural changes in the functional groups that comprise proteins and lipids. The study also demonstrated differences in lipid biosynthesis and protein levels in PV serum. Using the Partial Least Square Discriminant Analysis (PLS-DA) model, Raman-based multivariate analysis achieved high accuracy rates of 96.49% and 93.04% in the training sets and 93.10% and 89.66% in the test sets for the 800-1800 cm-1 and 2700-3000 cm-1 ranges, respectively. The area under the curve (AUC) values of the test datasets were calculated as 0.92 and 0.89 in the 800-1800 cm-1 and 2700-3000 cm-1 spectral regions, respectively, demonstrating the effectiveness of the PLS-DA models for the diagnosis of PV. This study highlights the potential of Raman spectroscopy-based analysis in the early and accurate diagnosis of PV, enabling the application of effective treatment strategies.