Glioma biopsies Classification Using Raman Spectroscopy and Machine Learning Models on Fresh Tissue Samples

A ratiometric SERS strategy for the prediction of cancer cell proportion and guidance of glioma surgical resection

Rapid and accurate identification of tumor boundaries is critical for the cure of glioma, but it is difficult due to the invasive nature of glioma cells. This paper aimed to explore a rapid diagnostic strategy based on a label-free surface-enhanced Raman scattering (SERS) technique for the quantitative detection of glioma cell proportion intraoperatively. With silver nanoparticles as substrate, an in-depth SERS analysis was performed on simulated clinical samples containing normal brain tissue and different concentrations of patient-derived glioma cells. The results revealed two universal characteristic peaks of 655 and 717 cm-1, which strongly correlated with glioma cell proportion regardless of individual differences. Based on the intensity ratio of the two peaks, a ratiometric SERS strategy for the quantification of glioma cells was established by employing an artificial neuron network model and a polynomial regression model. Such a strategy accurately estimated the proportion of glioma cells in simulated clinical samples (R2 = 0.98) and frozen samples (R2 = 0.85). More importantly, it accurately facilitated the delineation of tumor margins in freshly obtained samples. Taken together, this SERS-based method ensured a rapid and more detailed identification of tumor margins during surgical resection, which could be beneficial for intraoperative decision-making and pathological evaluation.

Glioma Identification Based on Digital Multimodal Spectra Integrated With Deep Learning Feature Fusion Using a Miniature Raman Spectrometer

The miniature fiber Raman spectroscopy detection technology can reflect the properties of biomolecules through spectral characteristics and has the advantages of noninvasiveness, real-time, safety, label-free operation, and potential for early cancer diagnosis. This technology holds promise for developing portable, low-cost, intraoperative tumor detection instruments. Glioma is one of the most common malignant tumors of the central nervous system with rapid growth and a short disease course. However, the considerable heterogeneity of the glioma sample leads to substantial intraclass variance in collected spectra, coupled with the miniature Raman spectrometer's low signal-to-noise ratio. These factors diminish the accuracy of the brain glioma recognition model. To address this issue, a glioma identification method based on digital multimodal spectra integrated with deep learning features fusion (DMS-DLFF) using the miniature Raman spectrometer is proposed. Different from existing multimodal tumor detection methods employing multiple spectral instruments, DMS-DLFF enhances tumor identification accuracy without increasing hardware costs. The method mathematically decomposes the original spectra to Raman and fluorescence spectra, so as to augment the biospectral information. Then, the deep learning method is used to extract the feature information of the two kinds of spectra, respectively, and the digital multimodal spectral fusion is realized at the feature level. Moreover, a two-layer pattern recognition model is constructed based on the ensemble strategy, amalgamating the strengths of diverse classifiers. Meanwhile, the bagging strategy is introduced to improve support vector machine algorithms, one of the basic classifiers. Compared with traditional methodologies, DMS-DLFF operates at both the feature level and decision level, employing high-information-density feature vectors to train ensemble classification models for increasing overall recognition accuracy. This study collected 260 Raman spectra of glioma and 151 Raman spectra of normal brain tissue. The accuracy, sensitivity, and specificity were 91.9%, 96.7%, and 80.8%, respectively. The proposed method outperforms traditional algorithms in brain glioma detection, which helps doctors formulate precise surgical plans and thereby improve patient prognosis.

A Single-Cell Metabolic Profiling Characterizes Human Aging via SlipChip-SERS

Metabolic dysregulation is a key driver of cellular senescence, contributing to the progression of systemic aging. The heterogeneity of senescent cells and their metabolic shifts are complex and unexplored. A microfluidic SlipChip integrated with surface-enhanced Raman spectroscopy (SERS), termed SlipChip-SERS, is developed for single-cell metabolism analysis. This SlipChip-SERS enables compartmentalization of single cells, parallel delivery of saponin and nanoparticles to release intracellular metabolites and to realize SERS detection with simple slipping operations. Analysis of different cancer cell lines using SlipChip-SERS demonstrated its capability for sensitive and multiplexed metabolic profiling of individual cells. When applied to human primary fibroblasts of different ages, it identified 12 differential metabolites, with spermine validated as a potent inducer of cellular senescence. Prolonged exposure to spermine can induce a classic senescence phenotype, such as increased senescence-associated β-glactosidase activity, elevated expression of senescence-related genes and reduced LMNB1 levels. Additionally, the senescence-inducing capacity of spermine in HUVECs and WRL-68 cells is confirmed, and exogenous spermine treatment increased the accumulation and release of H2O2. Overall, a novel SlipChip-SERS system is developed for single-cell metabolic analysis, revealing spermine as a potential inducer of senescence across multiple cell types, which may offer new strategies for addressing ageing and ageing-related diseases.

Meta-Analysis of the Efficacy of Raman Spectroscopy and Machine-Learning-Based Identification of Glioma Tissue

Intraoperative Raman spectroscopy (RS) has been identified as a potential tool for surgeons to rapidly and noninvasively differentiate between diseased and normal tissue. Since the previous meta-analysis on the subject was published in 2016, improvements in both spectroscopy equipment and machine learning models used to process spectra may have led to an increase in RS efficacy. Therefore, we decided to conduct a meta-analysis to determine the efficacy of RS when differentiating between glioma tissue and normal brain tissue. Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines were followed while conducting this meta-analysis. A search was conducted on PubMed and Web of Science for prospective and retrospective studies published between 2016 and 2022 using intraoperative RS and standard histology methods to differentiate between glioma and normal brain tissue. Meta-analyses of log odds ratios, sensitivity, and specificity were conducted in JASP using the random-effects model with restricted maximum likelihood estimation. A total of 9 studies met our inclusion criteria, comprising 673 patients and 8319 Raman spectra. Meta-analysis of log diagnostic odds ratios revealed high heterogeneity (I2 = 79.83%) and yielded a back-transformed diagnostic odds ratio of 76.71 (95% confidence interval: 39.57-148.71). Finally, meta-analysis for sensitivity and specificity of RS for glioma tissue showed high heterogeneity (I2 = 99.37% and 98.21%, respectively) and yielded an overall sensitivity of 95.3% (95% confidence interval: 91.0%-99.6%) and an overall specificity of 71.2% (95% confidence interval: 54.8%-87.6%). Calculation of a summary receiver operating curve yielded an overall area under the curve of 0.9265. Raman spectroscopy represents a promising tool for surgeons to quickly and accurately differentiate between healthy brain tissue and glioma tissue.

Study of the relationship among biomarkers, cell and tissue of glioma through Raman spectroscopy

Glioma is the most common brain tumors with high mortality and recurrence rates. Currently, the diagnosis methods for glioma are mainly based on tissue level, cellular level and biomarker level. In this paper, the characteristics of biomarkers (γ-aminobutyric acid and matrix mtalloproteinses-2), U87MG glioma cell and tissue were studied based on Raman spectroscopy, respectively. The results showed that the γ-aminobutyric acid concentration exhibited a linear relation with the intensity of characteristic peaks in 800-1600 cm-1 region, whereas the spectral baseline increased with the increasing of sample concentration in 200-700 cm-1 region. The Raman characteristics of matrix mtalloproteinses-2 in 20-1800 cm-1 region was investigated. Especially, it is demonstrated that the matrix mtalloproteinses-2 showed sixteen low-wavenumber Raman peaks in the range of 20-300 cm-1. Moreover, the U87MG glioma cell showed seven different Raman characteristic peaks in 600-1800 cm-1 region. Compared with the normal tissue, the Raman intensity of tumor tissue showed apparent intensity differences in 300-1800 cm-1, where the intensity changes of these Raman peaks were related to the reducing of the lipid metabolic pathways, and increase of the RNA in tumor tissue region. Furthermore, it is found that the Raman spectra of U87MG glioma cell and tumor tissue had corresponding peaks in the Raman spectra of the liquid γ-aminobutyric acid and matrix mtalloproteinses-2. It is suggested that the γ-aminobutyric acid and matrix mtalloproteinses-2 contributed to the formation and growth of glioma cell and tissue. Thus, Raman spectroscopy not only can diagnose glioma at the biomarkers, cellular and tissue level, but also analyze the relationship among the three. Furthermore, the results provided a physical marker for the detection of glioma in clinically.

Accurate and rapid molecular subgrouping of high-grade glioma via deep learning-assisted label-free fiber-optic Raman spectroscopy

Molecular genetics is highly related with prognosis of high-grade glioma. Accordingly, the latest WHO guideline recommends that molecular subgroups of the genes, including IDH, 1p/19q, MGMT, TERT, EGFR, Chromosome 7/10, CDKN2A/B, need to be detected to better classify glioma and guide surgery and treatment. Unfortunately, there is no preoperative or intraoperative technology available for accurate and comprehensive molecular subgrouping of glioma. Here, we develop a deep learning-assisted fiber-optic Raman diagnostic platform for accurate and rapid molecular subgrouping of high-grade glioma. Specifically, a total of 2,354 fingerprint Raman spectra was obtained from 743 tissue sites (astrocytoma: 151; oligodendroglioma: 150; glioblastoma (GBM): 442) of 44 high-grade glioma patients. The convolutional neural networks (ResNet) model was then established and optimized for molecular subgrouping. The mean area under receiver operating characteristic curves (AUC) for identifying the molecular subgroups of high-grade glioma reached 0.904, with mean sensitivity of 83.3%, mean specificity of 85.0%, mean accuracy of 83.3%, and mean time expense of 10.6 s. The diagnosis performance using ResNet model was shown to be superior to PCA-SVM and UMAP models, suggesting that high dimensional information from Raman spectra would be helpful. In addition, for the molecular subgroups of GBM, the mean AUC reached 0.932, with mean sensitivity of 87.8%, mean specificity of 83.6%, and mean accuracy of 84.1%. Furthermore, according to saliency maps, the specific Raman features corresponding to tumor-associated biomolecules (e.g. nucleic acid, tyrosine, tryptophan, cholesteryl ester, fatty acid, and collagen) were found to contribute to the accurate molecular subgrouping. Collectively, this study opens up new opportunities for accurate and rapid molecular subgrouping of high-grade glioma, which would assist optimal surgical resection and instant post-operative decision-making.

Optical Methods for Brain Tumor Detection: A Systematic Review

Background: In brain tumor surgery, maximal tumor resection is typically desired. This is complicated by infiltrative tumor cells which cannot be visually distinguished from healthy brain tissue. Optical methods are an emerging field that can potentially revolutionize brain tumor surgery through intraoperative differentiation between healthy and tumor tissues. Methods: This study aimed to systematically explore and summarize the existing literature on the use of Raman Spectroscopy (RS), Hyperspectral Imaging (HSI), Optical Coherence Tomography (OCT), and Diffuse Reflectance Spectroscopy (DRS) for brain tumor detection. MEDLINE, Embase, and Web of Science were searched for studies evaluating the accuracy of these systems for brain tumor detection. Outcome measures included accuracy, sensitivity, and specificity. Results: In total, 44 studies were included, covering a range of tumor types and technologies. Accuracy metrics in the studies ranged between 54 and 100% for RS, 69 and 99% for HSI, 82 and 99% for OCT, and 42 and 100% for DRS. Conclusions: This review provides insightful evidence on the use of optical methods in distinguishing tumor from healthy brain tissue.

A review of clinical use of surface-enhanced Raman scattering-based biosensing for glioma

Glioma is the most common malignant tumor of the nervous system in recent centuries, and the incidence rate of glioma is increasing year by year. Its invasive growth and malignant biological behaviors make it one of the most challenging malignant tumors. Maximizing the resection range (EOR) while minimizing the impact on normal brain tissue is crucial for patient prognosis. Changes in metabolites produced by tumor cells and their microenvironments might be important indicators. As a powerful spectroscopic technique, surface-enhanced Raman scattering (SERS) has many advantages, including ultra-high sensitivity, high specificity, and non-invasive features, which allow SERS technology to be widely applied in biomedicine, especially in the differential diagnosis of malignant tumor tissues. This review first introduced the clinical use of responsive SERS probes. Next, the sensing mechanisms of microenvironment-responsive SERS probes were summarized. Finally, the biomedical applications of these responsive SERS probes were listed in four sections, detecting tumor boundaries due to the changes of pH-responsive SERS probes, SERS probes to guide tumor resection, SERS for liquid biopsy to achieve early diagnosis of tumors, and the application of free-label SERS technology to detect fresh glioma specimens. Finally, the challenges and prospects of responsive SERS detections were summarized for clinical use.

FT-Raman spectra in combination with machine learning and multivariate analyses as a diagnostic tool in brain tumors

Brain tumors are one of the most dangerous, because the position of these are in the organ that governs all life processes. Moreover, a lot of brain tumor types were observed, but only one main diagnostic method was used - histopathology, for which preparation of sample was long. Consequently, a new, quicker diagnostic method is needed. In this paper, FT-Raman spectra of brain tissues were analyzed by Principal Component Analysis (PCA), Hierarchical Cluster Analysis (HCA), four different machine learning (ML) algorithms to show possibility of differentiating between glioblastoma G4 and meningiomas, as well as two different types of meningiomas (atypical and angiomatous). Obtained results showed that in meningiomas additional peak around 1503 cm-1 and higher level of amides was noticed in comparison with glioblastoma G4. In the case of meningiomas differentiation, in angiomatous meningiomas tissues lower level of lipids and polysaccharides were visible than in atypical meningiomas. Moreover, PCA analyses showed higher distinction between glioblastoma G4 and meningiomas in the FT-Raman range between 800 cm-1 and 1800 cm-1 and between two types of meningiomas in the range between 2700 cm-1 and 3000 cm-1. Decision trees showed, that the most important peaks to differentiate glioblastoma and meningiomas were at 1151 cm-1 and 2836 cm-1 while for angiomatous and atypical meningiomas - 1514 cm-1 and 2875 cm-1. Furthermore, the accuracy of obtained results for glioblastoma G4 and meningiomas was 88 %, while for meningiomas - 92 %. Consequently, obtained data showed possibility of using FT-Raman spectroscopy in diagnosis of different types of brain tumors.

Spectral insights: Navigating the frontiers of biomedical and microbiological exploration with Raman spectroscopy

Raman spectroscopy (RS), a powerful analytical technique, has gained increasing recognition and utility in the fields of biomedical and biological research. Raman spectroscopic analyses find extensive application in the field of medicine and are employed for intricate research endeavors and diagnostic purposes. Consequently, it enjoys broad utilization within the realm of biological research, facilitating the identification of cellular classifications, metabolite profiling within the cellular milieu, and the assessment of pigment constituents within microalgae. This article also explores the multifaceted role of RS in these domains, highlighting its distinct advantages, acknowledging its limitations, and proposing strategies for enhancement.

Computational Assessment of Spectral Heterogeneity within Fresh Glioblastoma Tissue Using Raman Spectroscopy and Machine Learning Algorithms

Understanding and classifying inherent tumor heterogeneity is a multimodal approach, which can be undertaken at the genetic, biochemical, or morphological level, among others. Optical spectral methods such as Raman spectroscopy aim at rapid and non-destructive tissue analysis, where each spectrum generated reflects the individual molecular composition of an examined spot within a (heterogenous) tissue sample. Using a combination of supervised and unsupervised machine learning methods as well as a solid database of Raman spectra of native glioblastoma samples, we succeed not only in distinguishing explicit tumor areas-vital tumor tissue and necrotic tumor tissue can correctly be predicted with an accuracy of 76%-but also in determining and classifying different spectral entities within the histomorphologically distinct class of vital tumor tissue. Measurements of non-pathological, autoptic brain tissue hereby serve as a healthy control since their respective spectroscopic properties form an individual and reproducible cluster within the spectral heterogeneity of a vital tumor sample. The demonstrated decipherment of a spectral glioblastoma heterogeneity will be valuable, especially in the field of spectroscopically guided surgery to delineate tumor margins and to assist resection control.

Distinguishing IDH mutation status in gliomas using FTIR-ATR spectra of peripheral blood plasma indicating clear traces of protein amyloid aggregation

BACKGROUND

Glioma is a primary brain tumor and the assessment of its molecular profile in a minimally invasive manner is important in determining treatment strategies. Among the molecular abnormalities of gliomas, mutations in the isocitrate dehydrogenase (IDH) gene are strong predictors of treatment sensitivity and prognosis. In this study, we attempted to non-invasively diagnose glioma development and the presence of IDH mutations using multivariate analysis of the plasma mid-infrared absorption spectra for a comprehensive and sensitive view of changes in blood components associated with the disease and genetic mutations. These component changes are discussed in terms of absorption wavenumbers that contribute to differentiation.

METHODS

Plasma samples were collected at our institutes from 84 patients with glioma (13 oligodendrogliomas, 17 IDH-mutant astrocytoma, 7 IDH wild-type diffuse glioma, and 47 glioblastomas) before treatment initiation and 72 healthy participants. FTIR-ATR spectra were obtained for each plasma sample, and PLS discriminant analysis was performed using the absorbance of each wavenumber in the fingerprint region of biomolecules as the explanatory variable. This data was used to distinguish patients with glioma from healthy participants and diagnose the presence of IDH mutations.

RESULTS

The derived classification algorithm distinguished the patients with glioma from healthy participants with 83% accuracy (area under the curve (AUC) in receiver operating characteristic (ROC) = 0.908) and diagnosed the presence of IDH mutation with 75% accuracy (AUC = 0.752 in ROC) in cross-validation using 30% of the total test data. The characteristic changes in the absorption spectra suggest an increase in the ratio of β-sheet structures in the conformational composition of blood proteins of patients with glioma. Furthermore, these changes were more pronounced in patients with IDH-mutant gliomas.

CONCLUSIONS

The plasma infrared absorption spectra could be used to diagnose gliomas and the presence of IDH mutations in gliomas with a high degree of accuracy. The spectral shape of the protein absorption band showed that the ratio of β-sheet structures in blood proteins was significantly higher in patients with glioma than in healthy participants, and protein aggregation was a distinct feature in patients with glioma with IDH mutations.

Discrimination of resected glioma tissues using surface enhanced Raman spectroscopy and Au@ZrO2 plasmonic nanosensor

Gliomas present one of the most prevalent malignant tumors related to the central nervous system. Surgical extraction is still a preferred route for glioma treatment. Nonetheless, neurosurgeons still have a considerable challenge to detect actual margins of the targeted glioma intraoperatively and correctly because of its great natural infiltration. Here we evaluated the possibility of using surface-enhanced Raman spectroscopy to analyze freshly resected brain tissues. The developed method is based on the application of Au@ZrO2 nanosensor. The plasmonic properties of the sensor were first tested on the analysis of Rhodamine 6G, where concentrations down to 10-7 mol/L can be successfully detected. We also compared the performance of the nanosensor with silver plasmonic nanoparticles, where similar results were obtained regarding the reduction of the fluorescence background and enhancement of the intensity of the measured analytical signal. However, application of silver nanospheres led to increased variations in spectral data due to its probable aggregation. Applied ZrO2@Au nanosensor thus dramatically lowers the fluorescence present in the Raman data, and considerably improves the quality of the measured signal. The developed method allows for rapid discrimination between the glioma's periphery and central parts, which could serve as a steppingstone toward highly precise neurosurgery.

Raman spectroscopic diagnosis of blast-induced traumatic brain injury in rats combined with machine learning

Blast-induced traumatic brain injury (bTBI) is a kind of nervous system disease, which results in a major health and economic problem to society. However, the rapid and label-free detection method with high sensitivity is still in great demand for the diagnosis of bTBI, especially for mild bTBI. In this paper, we report a new strategy for bTBI diagnosis through hippocampus and hypothalamus tissues based on Raman spectroscopy. The spectral characteristics of hippocampus and hypothalamus tissues of experimental bTBI in rats have been investigated for mild and moderate degrees at 3 h, 6 h, 24 h, 48 h, 72 h after blast exposure. The results show that the Raman spectra of mild and moderate bTBIs in 300-1700 cm-1 and 2800-3000 cm-1 regions exhibit significant differences at different time points compared with the control group. The main reason is the content change of proteins and lipids in hippocampus and hypothalamus tissues after bTBI. Moreover, four machine learning algorithms are used to automatically identify mild and moderate bTBIs at different time points (a total of 11 groups). The highest diagnostic accuracies are up to 95.3% and 88.5% based on Raman spectra of hippocampus and hypothalamus tissue, respectively. In addition, the classification performance of linear discriminant analysis classifier has been improved after data fusion. It is suggested that there has great potential as an alternative method for high-sensitive, rapid, label-free, economical diagnosis of bTBI.

Extended Analysis of Raman Spectra Using Artificial Intelligence Techniques for Colorectal Abnormality Classification

Raman spectroscopy (RS) techniques are attracting attention in the medical field as a promising tool for real-time biochemical analyses. The integration of artificial intelligence (AI) algorithms with RS has greatly enhanced its ability to accurately classify spectral data in vivo. This combination has opened up new possibilities for precise and efficient analysis in medical applications. In this study, healthy and cancerous specimens from 22 patients who underwent open colorectal surgery were collected. By using these spectral data, we investigate an optimal preprocessing pipeline for statistical analysis using AI techniques. This exploration entails proposing preprocessing methods and algorithms to enhance classification outcomes. The research encompasses a thorough ablation study comparing machine learning and deep learning algorithms toward the advancement of the clinical applicability of RS. The results indicate substantial accuracy improvements using techniques like baseline correction, L2 normalization, filtering, and PCA, yielding an overall accuracy enhancement of 15.8%. In comparing various algorithms, machine learning models, such as XGBoost and Random Forest, demonstrate effectiveness in classifying both normal and abnormal tissues. Similarly, deep learning models, such as 1D-Resnet and particularly the 1D-CNN model, exhibit superior performance in classifying abnormal cases. This research contributes valuable insights into the integration of AI in medical diagnostics and expands the potential of RS methods for achieving accurate malignancy classification.

In situ detection of human glioma based on tissue optical properties using diffuse reflectance spectroscopy

Safely maximizing brain cancer removal without injuring adjacent healthy tissue is crucial for optimal treatment outcomes. However, it is challenging to distinguish cancer from noncancer intraoperatively. This study aimed to explore the feasibility of diffuse reflectance spectroscopy (DRS) as a label-free and real-time detection technology for discrimination between brain cancer and noncancer tissues. Fifty-five fresh cancer and noncancer specimens from 19 brain surgeries were measured with DRS, and the results were compared with co-registered clinical standard histopathology. Tissue optical properties were quantitatively obtained from the diffuse reflectance spectra and compared among different types of brain tissues. A machine learning-based classifier was trained to differentiate cancerous versus noncancerous tissues. Our method could achieve a sensitivity of 93% and specificity of 95% for discriminating high-grade glioma from normal white matter. Our results showed that DRS has the potential to be used for label-free, real-time in vivo cancer detection during brain surgery.

An eXplainable Artificial Intelligence analysis of Raman spectra for thyroid cancer diagnosis

Raman spectroscopy shows great potential as a diagnostic tool for thyroid cancer due to its ability to detect biochemical changes during cancer development. This technique is particularly valuable because it is non-invasive and label/dye-free. Compared to molecular tests, Raman spectroscopy analyses can more effectively discriminate malignant features, thus reducing unnecessary surgeries. However, one major hurdle to using Raman spectroscopy as a diagnostic tool is the identification of significant patterns and peaks. In this study, we propose a Machine Learning procedure to discriminate healthy/benign versus malignant nodules that produces interpretable results. We collect Raman spectra obtained from histological samples, select a set of peaks with a data-driven and label independent approach and train the algorithms with the relative prominence of the peaks in the selected set. The performance of the considered models, quantified by area under the Receiver Operating Characteristic curve, exceeds 0.9. To enhance the interpretability of the results, we employ eXplainable Artificial Intelligence and compute the contribution of each feature to the prediction of each sample.

Raman Spectroscopy: A Tool for Molecular Fingerprinting of Brain Cancer

Brain cancer is one of those few cancers with very high mortality and low five-year survival rate. First and foremost reason for the woes is the difficulty in diagnosing and monitoring the progression of brain tumors both benign and malignant, noninvasively and in real time. This raises a need in this hour for a tool to diagnose the tumors in the earliest possible time frame. On the other hand, Raman spectroscopy which is well-known for its ability to precisely represent the molecular markers available in any sample given, including biological ones, with great sensitivity and specificity. This has led to a number of studies where Raman spectroscopy has been used in brain tumors in various ways. This review article highlights the fundamentals of Raman spectroscopy and its types including conventional Raman, SERS, SORS, SRS, CARS, etc. are used in brain tumors for diagnostics, monitoring, and even theragnostics, collating all the major works in the area. Also, the review explores how Raman spectroscopy can be even more effectively used in theragnostics and the clinical level which would make them a one-stop solution for all brain cancer needs in the future.

Dual Deep CNN for Tumor Brain Classification

Brain tumor (BT) is a serious issue and potentially deadly disease that receives much attention. However, early detection and identification of tumor type and location are crucial for effective treatment and saving lives. Manual diagnoses are time-consuming and depend on radiologist experts; the increasing number of new cases of brain tumors makes it difficult to process massive and large amounts of data rapidly, as time is a critical factor in patients' lives. Hence, artificial intelligence (AI) is vital for understanding disease and its various types. Several studies proposed different techniques for BT detection and classification. These studies are on machine learning (ML) and deep learning (DL). The ML-based method requires handcrafted or automatic feature extraction algorithms; however, DL becomes superior in self-learning and robust in classification and recognition tasks. This research focuses on classifying three types of tumors using MRI imaging: meningioma, glioma, and pituitary tumors. The proposed DCTN model depends on dual convolutional neural networks with VGG-16 architecture concatenated with custom CNN (convolutional neural networks) architecture. After conducting approximately 22 experiments with different architectures and models, our model reached 100% accuracy during training and 99% during testing. The proposed methodology obtained the highest possible improvement over existing research studies. The solution provides a revolution for healthcare providers that can be used as a different disease classification in the future and save human lives.

Data augmentation method based on the Gaussian kernel density for glioma diagnosis with Raman spectroscopy

Glioma is an intracranial malignant brain tumor with high infiltration. It is difficult to identify the glioma boundary. Raman spectroscopy can potentially detect this boundary accurately in situ and in vivo during surgery. However, when building a classification model for an in vitro experiment, fresh normal tissue is difficult to obtain. The number of normal tissues is far less than that of glioma tissues, which leads to a classification bias toward the majority class. In this study, a data augmentation algorithm GKIM based on the Gaussian kernel density is proposed for the data augmentation of normal tissue spectra. A weight coefficient calculation formula is proposed based on the Gaussian density instead of a fixed coefficient to synthesize new spectra, which increases sample diversity and improves the robustness of modeling. Additionally, the fuzzy nearest neighbor distance replaces the general fixed neighbor number K to select the original spectra for synthesis. It automatically determines the nearest spectra and adaptively synthesizes new spectra according to the characteristics of the input spectra. It effectively overcomes the problem of the newly generated sample distribution being too concentrated in specific spaces for the common data augmentation method. In this study, 769 Raman spectra of glioma and 136 Raman spectra of normal brain tissue corresponding to 205 and 37 cases, respectively, were collected. The Raman spectra of the normal tissue were extended to 600. The accuracy, sensitivity, and specificity were 91.67%, 91.67%, and 91.67%. The proposed method achieved better predictive performance than traditional algorithms for class imbalance.

A Handheld Visible Resonance Raman Analyzer Used in Intraoperative Detection of Human Glioma

There is still a lack of reliable intraoperative tools for glioma diagnosis and to guide the maximal safe resection of glioma. We report continuing work on the optical biopsy method to detect glioma grades and assess glioma boundaries intraoperatively using the VRR-LRRTM Raman analyzer, which is based on the visible resonance Raman spectroscopy (VRR) technique. A total of 2220 VRR spectra were collected during surgeries from 63 unprocessed fresh glioma tissues using the VRR-LRRTM Raman analyzer. After the VRR spectral analysis, we found differences in the native molecules in the fingerprint region and in the high-wavenumber region, and differences between normal (control) and different grades of glioma tissues. A principal component analysis–support vector machine (PCA-SVM) machine learning method was used to distinguish glioma tissues from normal tissues and different glioma grades. The accuracy in identifying glioma from normal tissue was over 80%, compared with the gold standard of histopathology reports of glioma. The VRR-LRRTM Raman analyzer may be a new label-free, real-time optical molecular pathology tool aiding in the intraoperative detection of glioma and identification of tumor boundaries, thus helping to guide maximal safe glioma removal and adjacent healthy tissue preservation.

Glycosylation spectral signatures for glioma grade discrimination using Raman spectroscopy

BACKGROUND

Gliomas are the most common brain tumours with the high-grade glioblastoma representing the most aggressive and lethal form. Currently, there is a lack of specific glioma biomarkers that would aid tumour subtyping and minimally invasive early diagnosis. Aberrant glycosylation is an important post-translational modification in cancer and is implicated in glioma progression. Raman spectroscopy (RS), a vibrational spectroscopic label-free technique, has already shown promise in cancer diagnostics.

METHODS

RS was combined with machine learning to discriminate glioma grades. Raman spectral signatures of glycosylation patterns were used in serum samples and fixed tissue biopsy samples, as well as in single cells and spheroids.

RESULTS

Glioma grades in fixed tissue patient samples and serum were discriminated with high accuracy. Discrimination between higher malignant glioma grades (III and IV) was achieved with high accuracy in tissue, serum, and cellular models using single cells and spheroids. Biomolecular changes were assigned to alterations in glycosylation corroborated by analysing glycan standards and other changes such as carotenoid antioxidant content.

CONCLUSION

RS combined with machine learning could pave the way for more objective and less invasive grading of glioma patients, serving as a useful tool to facilitate glioma diagnosis and delineate biomolecular glioma progression changes.

Raman Spectroscopy as a Tool to Study the Pathophysiology of Brain Diseases

The Raman phenomenon is based on the spontaneous inelastic scattering of light, which depends on the molecular characteristics of the dispersant. Therefore, Raman spectroscopy and imaging allow us to obtain direct information, in a label-free manner, from the chemical composition of the sample. Since it is well established that the development of many brain diseases is associated with biochemical alterations of the affected tissue, Raman spectroscopy and imaging have emerged as promising tools for the diagnosis of ailments. A combination of Raman spectroscopy and/or imaging with tagged molecules could also help in drug delivery and tracing for treatment of brain diseases. In this review, we first describe the basics of the Raman phenomenon and spectroscopy. Then, we delve into the Raman spectroscopy and imaging modes and the Raman-compatible tags. Finally, we center on the application of Raman in the study, diagnosis, and treatment of brain diseases, by focusing on traumatic brain injury and ischemia, neurodegenerative disorders, and brain cancer.

Discovering Glioma Tissue through Its Biomarkers' Detection in Blood by Raman Spectroscopy and Machine Learning

The most commonly occurring malignant brain tumors are gliomas, and among them is glioblastoma multiforme. The main idea of the paper is to estimate dependency between glioma tissue and blood serum biomarkers using Raman spectroscopy. We used the most common model of human glioma when continuous cell lines, such as U87, derived from primary human tumor cells, are transplanted intracranially into the mouse brain. We studied the separability of the experimental and control groups by machine learning methods and discovered the most informative Raman spectral bands. During the glioblastoma development, an increase in the contribution of lactate, tryptophan, fatty acids, and lipids in dried blood serum Raman spectra were observed. This overlaps with analogous results of glioma tissues from direct Raman spectroscopy studies. A non-linear relationship between specific Raman spectral lines and tumor size was discovered. Therefore, the analysis of blood serum can track the change in the state of brain tissues during the glioma development.

Raman spectroscopy: A prospective intraoperative visualization technique for gliomas

The infiltrative growth and malignant biological behavior of glioma make it one of the most challenging malignant tumors in the brain, and how to maximize the extent of resection (EOR) while minimizing the impact on normal brain tissue is the pursuit of neurosurgeons. The current intraoperative visualization assistance techniques applied in clinical practice suffer from low specificity, slow detection speed and low accuracy, while Raman spectroscopy (RS) is a novel spectroscopy technique gradually developed and applied to clinical practice in recent years, which has the advantages of being non-destructive, rapid and accurate at the same time, allowing excellent intraoperative identification of gliomas. In the present work, the latest research on Raman spectroscopy in glioma is summarized to explore the prospect of Raman spectroscopy in glioma surgery.

Innovations in the Diagnosis and Surgical Management of Low-Grade Gliomas

Low-grade gliomas are a broad category of tumors that can manifest at different stages of life. As a group, their prognosis has historically been considered to be favorable, and surgery is a mainstay of treatment. Advances in the molecular characterization of individual lesions has led to newer classification systems, a better understanding of the biological behavior of different neoplasms, and the identification of previously unrecognized entities. New prospective genetic and molecular data will help delineate better treatment paradigms and will continue to change the taxonomy of central nervous system tumors in the coming years. Advances in the field of radiomics will help predict the molecular profile of a particular tumor through noninvasive testing. Similarly, more precise methods of intraoperative tumor tissue analysis will aid surgical planning. Improved surgical outcomes propelled by novel surgical techniques and intraoperative adjuncts and emerging forms of medical treatment in the field of immunotherapy have enriched the management of these lesions. We review the contemporary management and innovations in the treatment of low-grade gliomas.

Differentiation of glioblastoma tissues using spontaneous Raman scattering with dimensionality reduction and data classification

The neurosurgery of intracranial tumors is often complicated by the difficulty of distinguishing tumor center, infiltration area, and normal tissue. The current standard for intraoperative navigation is fluorescent diagnostics with a fluorescent agent. This approach can be further enhanced by measuring the Raman spectrum of the tissue, which would provide additional information on its composition even in the absence of fluorescence. However, for the Raman spectra to be immediately helpful for a neurosurgeon, they must be additionally processed. In this work, we analyzed the Raman spectra of human brain glioblastoma multiforme tissue samples obtained during the surgery and investigated several approaches to dimensionality reduction and data classificatin to distinguish different types of tissues. In our study two approaches to Raman spectra dimensionality reduction were approbated and as a result we formulated new technique combining both of them: feature filtering based on the selection of those shifts which correspond to the biochemical components providing the statistically significant differences between groups of examined tissues (center of glioblastoma multiforme, tissues from infiltration area and normally appeared white matter) and principal component analysis. We applied the support vector machine to classify tissues after dimensionality reduction of registered Raman spectra. The accuracy of the classification of malignant tissues (tumor edge and center) and normal ones using the principal component analysis alone was 83% with sensitivity of 96% and specificity of 44%. With a combined technique of dimensionality reduction we obtained 83% accuracy with 77% sensitivity and 92% specificity of tumor tissues classification.

Intraoperative tissue classification methods in orthopedic and neurological surgeries: A systematic review

Accurate tissue differentiation during orthopedic and neurological surgeries is critical, given that such surgeries involve operations on or in the vicinity of vital neurovascular structures and erroneous surgical maneuvers can lead to surgical complications. By now, the number of emerging technologies tackling the problem of intraoperative tissue classification methods is increasing. Therefore, this systematic review paper intends to give a general overview of existing technologies. The review was done based on the PRISMA principle and two databases: PubMed and IEEE Xplore. The screening process resulted in 60 full-text papers. The general characteristics of the methodology from extracted papers included data processing pipeline, machine learning methods if applicable, types of tissues that can be identified with them, phantom used to conduct the experiment, and evaluation results. This paper can be useful in identifying the problems in the current status of the state-of-the-art intraoperative tissue classification methods and designing new enhanced techniques.

Raman spectroscopy and machine learning for the classification of esophageal squamous carcinoma

Early diagnosis of esophageal squamous cell carcinoma (ESCC), a common malignant tumor with a low overall survival rate due to metastasis and recurrence, is critical for effective treatment and improved prognosis. Raman spectroscopy, an advanced detection technology for esophageal cancer, was developed to improve diagnosis sensitivity, specificity, and accuracy. This study proposed a novel, effective, and noninvasive Raman spectroscopy technique to differentiate and classify ESCC cell lines. Seven ESCC cell lines and tissues of an ESCC patient with staging of T3N1M0 and T3N2M0 at low and high differentiation levels were investigated through Raman spectroscopy. Raman spectral data analysis was performed with four machine learning algorithms, namely principal components analysis (PCA)- linear discriminant analysis (LDA), PCA-eXtreme gradient boosting (XGB), PCA- support vector machine (SVM), and PCA- (LDA, XGB, SVM)-stacked Gradient Boosting Machine (GBM). Four machine learning algorithms were able to classifiy ESCC cell subtypes from normal esophageal cells. The PCA-XGB model achieved an overall predictive accuracy of 85% for classifying ESCC and adjacent tissues. Moreover, an overall predictive accuracy of 90.3% was achieved in distinguishing low differentiation and high differentiation ESCC tissues with the same stage when PCA-LDA, XGM, and SVM models were combined. This study illustrated the Raman spectral traits of ESCC cell lines and esophageal tissues related to clinical pathological diagnosis. Future studies should investigate the role of Raman spectral features in ESCC pathogenesis.

Machine Learning of Raman Spectroscopy Data for Classifying Cancers: A Review of the Recent Literature

Raman Spectroscopy has long been anticipated to augment clinical decision making, such as classifying oncological samples. Unfortunately, the complexity of Raman data has thus far inhibited their routine use in clinical settings. Traditional machine learning models have been used to help exploit this information, but recent advances in deep learning have the potential to improve the field. However, there are a number of potential pitfalls with both traditional and deep learning models. We conduct a literature review to ascertain the recent machine learning methods used to classify cancers using Raman spectral data. We find that while deep learning models are popular, and ostensibly outperform traditional learning models, there are many methodological considerations which may be leading to an over-estimation of performance; primarily, small sample sizes which compound sub-optimal choices regarding sampling and validation strategies. Amongst several recommendations is a call to collate large benchmark Raman datasets, similar to those that have helped transform digital pathology, which researchers can use to develop and refine deep learning models.

Rapid Discrimination of Clinically Important Pathogens Through Machine Learning Analysis of Surface Enhanced Raman Spectra

With its low-cost, label-free and non-destructive features, Raman spectroscopy is becoming an attractive technique with high potential to discriminate the causative agent of bacterial infections and bacterial infections per se. However, it is challenging to achieve consistency and accuracy of Raman spectra from numerous bacterial species and phenotypes, which significantly hinders the practical application of the technique. In this study, we analyzed surfaced enhanced Raman spectra (SERS) through machine learning algorithms in order to discriminate bacterial pathogens quickly and accurately. Two unsupervised machine learning methods, K-means Clustering (K-Means) and Agglomerative Nesting (AGNES) were performed for clustering analysis. In addition, eight supervised machine learning methods were compared in terms of bacterial predictions via Raman spectra, which showed that convolutional neural network (CNN) achieved the best prediction accuracy (99.86%) with the highest area (0.9996) under receiver operating characteristic curve (ROC). In sum, machine learning methods can be potentially applied to classify and predict bacterial pathogens via Raman spectra at general level.

Discrimination of glioma patient-derived cells from healthy astrocytes by exploiting Raman spectroscopy

Glioblastoma multiforme (GBM) is one of the most common and aggressive brain tumors. It presents a very bad prognosis with a patients' overall survival of 12-15 months; treatment failure is mainly ascribable to tumor recurrence. The development of new tools, that could help the precise detection of the tumor border, is thus an urgent need. During the last decades, different vibrational spectroscopy techniques have been developed to distinguish cancer tissue from heathy tissue; in the present work, we compared GBM cells deriving from four patients with healthy human astrocytes using Raman spectroscopy. We have shown that the region between 1000 and 1300 cm^-1 is enough informative for this discrimination, indeed highlighting that peaks related to DNA/RNA and cytochrome c are increased in cancer cells. Finally, our model has been able to discriminate cancer cells from healthy cells with an average accuracy of 92.5%. We believe that this study might help to further understand which are the essential Raman peaks exploitable in the detection of cancer cells, with important perspectives under a diagnostic point of view.

DDRugging glioblastoma: understanding and targeting the DNA damage response to improve future therapies

Glioblastoma is the most frequently diagnosed type of primary brain tumour in adults. These aggressive tumours are characterised by inherent treatment resistance and disease progression, contributing to ~ 190 000 brain tumour-related deaths globally each year. Current therapeutic interventions consist of surgical resection followed by radiotherapy and temozolomide chemotherapy, but average survival is typically around 1 year, with < 10% of patients surviving more than 5 years. Recently, a fourth treatment modality of intermediate-frequency low-intensity electric fields [called tumour-treating fields (TTFields)] was clinically approved for glioblastoma in some countries after it was found to increase median overall survival rates by ~ 5 months in a phase III randomised clinical trial. However, beyond these treatments, attempts to establish more effective therapies have yielded little improvement in survival for patients over the last 50 years. This is in contrast to many other types of cancer and highlights glioblastoma as a recognised tumour of unmet clinical need. Previous work has revealed that glioblastomas contain stem cell-like subpopulations that exhibit heightened expression of DNA damage response (DDR) factors, contributing to therapy resistance and disease relapse. Given that radiotherapy, chemotherapy and TTFields-based therapies all impact DDR mechanisms, this Review will focus on our current knowledge of the role of the DDR in glioblastoma biology and treatment. We also discuss the potential of effective multimodal targeting of the DDR combined with standard-of-care therapies, as well as emerging therapeutic targets, in providing much-needed improvements in survival rates for patients.

Integration of Raman spectra with transcriptome data in glioblastoma multiforme defines tumour subtypes and predicts patient outcome

Raman spectroscopy is an imaging technique that has been applied to assess molecular compositions of living cells to characterize cell types and states. However, owing to the diverse molecular species in cells and challenges of assigning peaks to specific molecules, it has not been clear how to interpret cellular Raman spectra. Here, we provide firm evidence that cellular Raman spectra (RS) and transcriptomic profiles of glioblastoma can be computationally connected and thus interpreted. We find that the dimensions of high-dimensional RS and transcriptomes can be reduced and connected linearly through a shared low-dimensional subspace. Accordingly, we were able to predict global gene expression profiles by applying the calculated transformation matrix to Raman spectra and vice versa. From these analyses, we extract a minimal gene expression signature associated with specific RS profiles and predictive of disease outcome.

Rapid identification of papillary thyroid carcinoma and papillary microcarcinoma based on serum Raman spectroscopy combined with machine learning models

Thyroid carcinoma is one kind of cancer with the highest diagnosis rate in the endocrine system, and its main histological subtype is papillary thyroid carcinoma (PTC) accounting for 80% of thyroid malignancies. In recent years, the incidence of thyroid cancer has increased exponentially, and its substantial increase was closely related to the overdiagnosis of papillary microcarcinoma (PMC). Therefore, early and accurate identification of PTC and PMC can prevent patients from being irreversibly damaged. This study aimed to identify PTC and PMC using Raman spectroscopy. We collected serum Raman spectra from 16 patients with PTC and 31 patients with PMC. Firstly, the collected imbalance data were preprocessed using the synthetic minority over-sampling technique (SMOTE). Then, the equalized data were dimensionality reduced by principal component analysis (PCA). Finally, the processed data were fed into the single decision tree (DT) classifier, as well as the random forest (RF) built on the idea of Boosting ensemble and the Adaptive Boosting (Adaboost) model built on the idea of Bagging ensemble for classification. The classification accuracy of the three models in the testing set were 75.38%, 81.54%, and 84.61%, respectively. Compared with the DT classifier, the accuracy of the models introducing the idea of ensemble learning was enhanced by 6.16% and 9.23%, respectively. The best model was the Adaboost. This result demonstrates that serum Raman spectroscopy combined with an ensemble learning algorithm was feasible in rapidly identifying PTC and PMC. At the same time, the method has great potential for application in the field of clinical diagnosis.

Raman Imaging and Fluorescence Lifetime Imaging Microscopy for Diagnosis of Cancer State and Metabolic Monitoring

Hurdles for effective tumor therapy are delayed detection and limited effectiveness of systemic drug therapies by patient-specific multidrug resistance. Non-invasive bioimaging tools such as fluorescence lifetime imaging microscopy (FLIM) and Raman-microspectroscopy have evolved over the last decade, providing the potential to be translated into clinics for early-stage disease detection, in vitro drug screening, and drug efficacy studies in personalized medicine. Accessing tissue- and cell-specific spectral signatures, Raman microspectroscopy has emerged as a diagnostic tool to identify precancerous lesions, cancer stages, or cell malignancy. In vivo Raman measurements have been enabled by recent technological advances in Raman endoscopy and signal-enhancing setups such as coherent anti-stokes Raman spectroscopy or surface-enhanced Raman spectroscopy. FLIM enables in situ investigations of metabolic processes such as glycolysis, oxidative stress, or mitochondrial activity by using the autofluorescence of co-enzymes NADH and FAD, which are associated with intrinsic proteins as a direct measure of tumor metabolism, cell death stages and drug efficacy. The combination of non-invasive and molecular-sensitive in situ techniques and advanced 3D tumor models such as patient-derived organoids or microtumors allows the recapitulation of tumor physiology and metabolism in vitro and facilitates the screening for patient-individualized drug treatment options.

Raman Spectroscopy and Machine Learning for IDH Genotyping of Unprocessed Glioma Biopsies

Simple Summary

Isocitrate dehydrogenase (IDH) mutation is one of the most important prognostic markers in glioma tumors. Raman spectroscopy (RS) is an optical technique with great potential in intraoperative molecular diagnosis and surgical guidance. We analyzed RS’s ability to detect the IDH mutation onto unprocessed glioma biopsies. A total of 2073 Raman spectra were extracted from 38 tumor specimens. From the 103 Raman shifts screened, we identified 52 shifts (related to lipids, collagen, DNA and cholesterol/phospholipids) with the highest performance in the distinction of the two groups. We described 18 shifts never used before for IDH detection with RS in fresh or frozen samples. We were able to distinguish between IDH-mutated and IDH-wild-type tumors with an accuracy and precision of 87%. RS showed optimal accuracy and precision in discriminating IDH-mutated glioma from IDH-wild-type tumors ex-vivo onto fresh surgical specimens.

Abstract

Isocitrate dehydrogenase (IDH) mutational status is pivotal in the management of gliomas. Patients with IDH-mutated (IDH-MUT) tumors have a better prognosis and benefit more from extended surgical resection than IDH wild-type (IDH-WT). Raman spectroscopy (RS) is a minimally invasive optical technique with great potential for intraoperative diagnosis. We evaluated the RS’s ability to characterize the IDH mutational status onto unprocessed glioma biopsies. We extracted 2073 Raman spectra from thirty-eight unprocessed samples. The classification performance was assessed using the eXtreme Gradient Boosted trees (XGB) and Support Vector Machine with Radial Basis Function kernel (RBF-SVM). Measured Raman spectra displayed differences between IDH-MUT and IDH-WT tumor tissue. From the 103 Raman shifts screened as input features, the cross-validation loop identified 52 shifts with the highest performance in the distinction of the two groups. Raman analysis showed differences in spectral features of lipids, collagen, DNA and cholesterol/phospholipids. We were able to distinguish between IDH-MUT and IDH-WT tumors with an accuracy and precision of 87%. RS is a valuable and accurate tool for characterizing the mutational status of IDH mutation in unprocessed glioma samples. This study improves RS knowledge for future personalized surgical strategy or in situ target therapies for glioma tumors.

Classification of Diffuse Glioma Subtype from Clinical-Grade Pathological Images Using Deep Transfer Learning

Diffuse gliomas are the most common primary brain tumors and they vary considerably in their morphology, location, genetic alterations, and response to therapy. In 2016, the World Health Organization (WHO) provided new guidelines for making an integrated diagnosis that incorporates both morphologic and molecular features to diffuse gliomas. In this study, we demonstrate how deep learning approaches can be used for an automatic classification of glioma subtypes and grading using whole-slide images that were obtained from routine clinical practice. A deep transfer learning method using the ResNet50V2 model was trained to classify subtypes and grades of diffuse gliomas according to the WHO’s new 2016 classification. The balanced accuracy of the diffuse glioma subtype classification model with majority voting was 0.8727. These results highlight an emerging role of deep learning in the future practice of pathologic diagnosis.