Raman fiber-optic probe for rapid diagnosis of gastric and esophageal tumors with machine learning analysis or similarity assessments: a comparative study.

Gastric and esophageal cancers, the predominant forms of upper gastrointestinal malignancies, contribute significantly to global cancer mortality. Routine detection methods, including medical imaging, endoscopic examination, and pathological biopsy, often suffer from drawbacks such as low sensitivity and laborious and complex procedures. Raman spectroscopy is a non-invasive and label-free optical technique that provides highly sensitive biomolecular information to facilitate effective tumor identification. In this work, we report the use of fiber-optic Raman spectroscopy for the accurate and rapid diagnosis of gastric and esophageal cancers. Using a database of 14,000 spectra from 140 ex vivo tissue pieces of both tumor and normal tissue samples, we compare the random forest (RF) and our established Euclidean distance Raman spectroscopy (EDRS) model. The RF analysis achieves a sensitivity of 85.23% and an accuracy of 83.05% in diagnosing gastric tumors. The EDRS algorithm with improved diagnostic transparency further increases the sensitivity to 92.86% and accuracy to 89.29%. When these diagnostic protocols are extended to esophageal tumors, the RF and EDRS models achieve accuracies of 71.27% and 93.18%, respectively. Finally, we demonstrate that fewer than 20 spectra are sufficient to achieve good Raman diagnostic accuracy for both tumor tissues. This optimizes the balance between acquisition time and diagnostic performance. Our work, although conducted on ex vivo tissue models, offers valuable insights for in vivo in situ endoscopic Raman diagnosis of gastric and esophageal cancer lesions in the future. Our study provides a robust, rapid, and convenient method as a new paradigm in in vivo endoscopic medical diagnostics that integrates spectroscopic techniques and a Raman probe for detecting upper gastrointestinal malignancies.

Raman spectroscopy analysis combined with computed tomography imaging to identify microsatellite instability in gastric cancers.

Accurate determination of microsatellite instability (MSI) status is critical for tailoring treatment approaches for gastric cancer patients. Existing clinical techniques for MSI diagnosis are plagued by problems of suboptimal time efficiency, high cost, and burdensome experimental requirements. Here, we for the first time establish the classification model of gastric cancer MSI status based on Raman spectroscopy. To begin with, we reveal that tumor heterogeneity-induced signal variations pose a prominent impact on MSI classification. To eliminate this issue, we develop Euclidean distance-based Raman Spectroscopy (EDRS) algorithm, which establishes a standard spectrum to represent the "most microsatellite stable" status. The similarity between each spectrum of tissues with the standard spectrum is calculated to provide a direct assessment on the MSI status. Compared to machine learning-algorithms including k-Nearest Neighbors, Random Forest, and Extreme Learning Machine, the EDRS method shows the highest accuracy of 94.6 %. Finally, we integrate the EDRS method with the clinical diagnostic modality, computed tomography, to construct an innovative joint classification model with good classification performance (AUC = 0.914, Accuracy = 94.6 %). Our work demonstrates a robust, rapid, non-invasive, and convenient tool in identifying the MSI status, and opens new avenues for Raman techniques to fit into existing clinical workflow.

Label-Free Surface-Enhanced Raman Spectroscopy with Machine Learning for the Diagnosis of Thyroid Cancer by Using Fine-Needle Aspiration Liquid Samples.

The incidence of thyroid cancer is increasing worldwide. Fine-needle aspiration (FNA) cytology is widely applied with the use of extracted biological cell samples, but current FNA cytology is labor-intensive, time-consuming, and can lead to the risk of false-negative results. Surface-enhanced Raman spectroscopy (SERS) combined with machine learning algorithms holds promise for cancer diagnosis. In this study, we develop a label-free SERS liquid biopsy method with machine learning for the rapid and accurate diagnosis of thyroid cancer by using thyroid FNA washout fluids. These liquid supernatants are mixed with silver nanoparticle colloids, and dispersed in quartz capillary for SERS measurements to discriminate between healthy and malignant samples. We collect Raman spectra of 36 thyroid FNA samples (18 malignant and 18 benign) and compare four classification models: Principal Component Analysis-Linear Discriminant Analysis (PCA-LDA), Random Forest (RF), Support Vector Machine (SVM), and Convolutional Neural Network (CNN). The results show that the CNN algorithm is the most precise, with a high accuracy of 88.1%, sensitivity of 87.8%, and the area under the receiver operating characteristic curve of 0.953. Our approach is simple, convenient, and cost-effective. This study indicates that label-free SERS liquid biopsy assisted by deep learning models holds great promise for the early detection and screening of thyroid cancer.

Oppositely-charged silver nanoparticles enable selective SERS molecular enhancement through electrostatic interactions.

Label-free surface-enhanced Raman spectroscopy (SERS) has attracted extensive attention as an emerging technique for molecular phenotyping of biological samples. However, the selective enhancement property of SERS mediated by complicated interactions between substrates and analytes is unfavorable for molecular profiling. The electrostatic force is among the most dominating interactions that can cause selective adsorption of molecules to charged substrates. This means if only negatively- or positively-charged SERS substrates are applied, then considerable SERS information from a portion of analytes would be lost, hindering comprehensive SERS sensing. In this work, we utilize both negatively- and positively-charged colloidal silver (Ag) nanoparticles (NPs) to detect various charged molecules. The negatively-charged citrate-stabilized Ag and the positively-charged Ag prepared via a cetyltrimethyl-ammonium chloride-based charge reversal protocol have been adopted as SERS substrates. The Ag NPs are all relatively well-dispersed with good uniformity. After applying the oppositely-charged NPs to the detection of charged molecules, we find the SERS results explicitly demonstrate the electrostatically-driven SERS selective enhancement, which is further supported and clarified by molecular electrostatic potential calculations. Our work highlights the importance of developing SERS substrates modified with appropriate surface charges for various analytes, and enlightens us that potentially more molecular SERS information can be acquired from complex bio-samples using combinations of oppositely-charged substrates.

NIR-II Surface-Enhanced Raman Scattering Nanoprobes in Biomedicine: Current Impact and Future Directions.

The field of second near-infrared (NIR-II) surface-enhanced Raman scattering (SERS) nanoprobes has made commendable progress in biomedicine. This article reviews recent advances and future development of NIR-II SERS nanoprobes. It introduces the fundamental principles of SERS nanoprobes and highlights key advances in the NIR-II window, including reduced tissue attenuation, deep penetration, maximized allowable exposure, and improved photostability. The discussion of future directions includes the refinement of nanoprobe substrates, emphasizing the tailoring of optical properties of metallic SERS-active nanoprobes, and exploring non-metallic alternatives. The intricacies of designing Raman reporters for the NIR-II resonance and the potential of these reporters to advance the field are also discussed. The integration of artificial intelligence (AI) into nanoprobe design represents a cutting-edge approach to overcome current challenges. This article also examines the emergence of deep Raman techniques for through-tissue SERS detection, toward NIR-II SERS tomography. It acknowledges instrumental advancements like improved charge-coupled device sensitivity and accelerated imaging speeds. The article concludes by addressing the critical aspects of biosafety, ease of functionalization, compatibility, and the path to clinical translation. With a comprehensive overview of current achievements and future prospects, this review aims to illuminate the path for NIR-II SERS nanoprobes to innovate diagnostic and therapeutic approaches in biomedicine.

Raman spectroscopy for esophageal tumor diagnosis and delineation using machine learning and the portable Raman spectrometer.

Esophageal cancer is one of the leading causes of cancer-related deaths worldwide. The identification of residual tumor tissues in the surgical margin of esophageal cancer is essential for the treatment and prognosis of cancer patients. But the current diagnostic methods, either pathological frozen section or paraffin section examination, are laborious, time-consuming, and inconvenient. Raman spectroscopy is a label-free and non-invasive analytical technique that provides molecular information with high specificity. Here, we report the use of a portable Raman system and machine learning algorithms to achieve accurate diagnosis of esophageal tumor tissue in surgically resected specimens. We tested five machine learning-based classification methods, including k-Nearest Neighbors, Adaptive Boosting, Random Forest, Principal Component Analysis-Linear Discriminant Analysis, and Support Vector Machine (SVM). Among them, SVM shows the highest accuracy (88.61 %) in classifying the esophageal tumor and normal tissues. The portable Raman system demonstrates robust measurements with an acceptable focal plane shift of up to 3 mm, which enables large-area Raman mapping on resected tissues. Based on this, we finally achieve successful Raman visualization of tumor boundaries on surgical margin specimens, and the Raman measurement time is less than 5 min. This work provides a robust, convenient, accurate, and cost-effective tool for the diagnosis of esophageal cancer tumors, advancing toward Raman-based clinical intraoperative applications.

SERSomes for metabolic phenotyping and prostate cancer diagnosis.

Molecular phenotypic variations in metabolites offer the promise of rapid profiling of physiological and pathological states for diagnosis, monitoring, and prognosis. Since present methods are expensive, time-consuming, and still not sensitive enough, there is an urgent need for approaches that can interrogate complex biological fluids at a system-wide level. Here, we introduce hyperspectral surface-enhanced Raman spectroscopy (SERS) to profile microliters of biofluidic metabolite extraction in 15 min with a spectral set, SERSome, that can be used to describe the structures and functions of various molecules produced in the biofluid at a specific time via SERS characteristics. The metabolite differences of various biofluids, including cell culture medium and human serum, are successfully profiled, showing a diagnosis accuracy of 80.8% on the internal test set and 73% on the external validation set for prostate cancer, discovering potential biomarkers, and predicting the tissue-level pathological aggressiveness. SERSomes offer a promising methodology for metabolic phenotyping.

Sentinel lymph node identification using NIR-II ultrabright Raman nanotags on preclinical models.

Surface-enhanced Raman spectroscopy (SERS) nanotags have garnered much attention as promising bioimaging contrast agent with ultrahigh sensitivity, but their clinical translation faces challenges including biological and laser safety. As breast sentinel lymph node (SLN) imaging agents, SERS nanotags used by local injection and only accumulation in SLNs, which were removed during surgery, greatly reduce biological safety concerns. But their clinical translation lacks pilot demonstration on large animals close to humans. The laser safety requires irradiance below the maximum permissible exposure threshold, which is currently not achievable in most SERS applications. Here we report the invention of the core-shell SERS nanotags with ultrahigh brightness (1 pM limit of detection) at the second near-infrared (NIR-II) window for SLN identification on pre-clinical animal models including rabbits and non-human primate. We for the first time realize the intraoperative SERS-guided SLN navigation under a clinically safe laser (1.73 J/cm2) and identify multiple axillary SLNs on a non-human primate. No evidence of biosafety issues was observed in systematic examinations of these nanotags. Our study unveils the potential of NIR-II SERS nanotags as appropriate SLN tracers, making significant advances toward the accurate positioning of lesions using the SERS-based tracer technique.