Towards automated cancer screening: Label-free classification of fixed cell samples using wavelength modulated Raman spectroscopy

High-Precision Intelligent Diagnosis of Pancreatic Cancer: Flowing Diffuseness from Single to Whole

Raman spectroscopy, as a label-free optical technique, provides a unique solution for tissue diagnosis. However, due to the limitation of point-by-point acquisition mode and multivariate statistical analysis methods, conventional methods pose a major bottleneck toward achieving highly time-efficient, accurate, and holistic diagnosis. Here, the authors establish line-scan Raman spectrochemical holistic analysis (LRSHA), an intelligent diagnostic method for rapid Raman data collection and holistic tissue diagnosis. The line-scan technique is first used for rapid spectral acquisition (∼15 s) in a tissue block area (0.75 × 0.5 mm2), which is 2 orders of magnitude faster than conventional methods. Then, the one-dimensional (1D) Raman spectra are converted into two-dimensional (2D) Raman encoding figures by the spectral recurrence plot transformation. The 2D deep learning models achieve 96.0% accuracy, 7% higher than that of 1D deep learning models. Moreover, the neighborhood enhancement method is applied to correct the initial deep learning results, like ripple flowing diffuseness from single to whole, which ultimately greatly improves the diagnostic accuracy to 99.7%. We also demonstrated that our method can identify tissue neoplasia at resection margins that appear nearly normal to the naked eye. Together, the LRSHA method shows valuable potential for rapid, efficient, and accurate tissue diagnosis.

High-Precision Intelligent Cancer Diagnosis Method: 2D Raman Figures Combined with Deep Learning

Raman spectroscopy, as a label-free detection technology, has been widely used in tumor diagnosis. However, most tumor diagnosis procedures utilize multivariate statistical analysis methods for classification, which poses a major bottleneck toward achieving high accuracy. Here, we propose a concept called the two-dimensional (2D) Raman figure combined with convolutional neural network (CNN) to improve the accuracy. Two-dimensional Raman figures can be obtained from four transformation methods: spectral recurrence plot (SRP), spectral Gramian angular field (SGAF), spectral short-time Fourier transform (SSTFT), and spectral Markov transition field (SMTF). Two-dimensional CNN models all yield more than 95% accuracy, which is higher than the PCA-LDA method and the Raman-spectrum-CNN method, indicating that 2D Raman figure inputs combined with CNN may be one reason for gaining excellent performances. Among 2D-CNN models, the main difference is the conversion, where SRP is based on the structure of wavenumber series with the best performances (98.9% accuracy, 99.5% sensitivity, 98.3% specificity), followed by SGAF on the wavenumber series, SSTFT on wavenumber and intensity information, and SMTF on wavenumber position information. The inclusion of external information in the conversion may be another reason for improvement in the accuracy. The excellent capability shows huge potential for tumor diagnosis via 2D Raman figures and may be applied in other spectroscopy analytical fields.

Raman Spectroscopy of Individual Cervical Exfoliated Cells in Premalignant and Malignant Lesions

Cervical cancer is frequent neoplasia. Currently, the diagnostic approach includes cervical cytology, colposcopy, and histopathology studies; combining detection techniques increases the sensitivity and specificity of the tests. Raman spectroscopy is a high-resolution technique that supports the diagnosis of malignancies. This study aimed to evaluate the Raman spectroscopy technique discriminating between healthy and premalignant/malignant cervical cells. We included 81 exfoliative cytology samples, 29 in the “healthy group” (negative cytology), and 52 in the “CIN group” (premalignant/malignant lesions). We obtained the nucleus and cytoplasm Raman spectra of individual cells. We tested the spectral differences between groups using Permutational Multivariate Analysis of Variance (PERMANOVA) and Canonical Analysis of Principal Coordinates (CAP). We found that Raman spectra have increased intensity in premalignant/malignant cells compared with healthy cells. The characteristic Raman bands corresponded to proteins and nucleic acids, in concordance with the increased replication and translation processes in premalignant/malignant states. We found a classification efficiency of 76.5% and 82.7% for cytoplasmic and nuclear Raman spectra, respectively; cell nucleus Raman spectra showed a sensitivity of 84.6% in identifying cervical anomalies. The classification efficiency and sensitivity obtained for nuclear spectra suggest that Raman spectroscopy could be helpful in the screening and diagnosis of premalignant lesions and cervical cancer.

Microscale diamond protection for a ZnO coated fiber optic sensor

Fiber optic sensors are widely used in environmental, biological and chemical sensing. Due to the demanding environmental conditions in which they can be used, there is a risk of damaging the sensor measurement head placed in the measuring field. Sensors using nanolayers deposited upon the fiber structure are particularly vulnerable to damage. A thin film placed on the surface of the fiber end-face can be prone to mechanical damage or deteriorate due to unwanted chemical reactions with the surrounding agent. In this paper, we investigated a sensor structure formed with a Zinc Oxide (ZnO) coating, deposited by Atomic Layer Deposition (ALD) on the tip of a single-mode fiber. A nanocrystalline diamond sheet (NDS) attached over the ZnO is described. The diamond structure was synthesized in a Microwave Plasma Assisted Chemical Vapor Deposition System. The deposition processes of the nanomaterials, the procedure of attaching NDS to the fiber end-face covered with ZnO, and the results of optical measurements are presented.

Imaging the invisible-Bioorthogonal Raman probes for imaging of cells and tissues

A revolutionary avenue for vibrational imaging with super-multiplexing capability can be seen in the recent development of Raman-active bioortogonal tags or labels. These tags and isotopic labels represent groups of chemically inert and small modifications, which can be introduced to any biomolecule of interest and then supplied to single cells or entire organisms. Recent developments in the field of spontaneous Raman spectroscopy and stimulated Raman spectroscopy in combination with targeted imaging of biomolecules within living systems are the main focus of this review. After having introduced common strategies for bioorthogonal labeling, we present applications thereof for profiling of resistance patterns in bacterial cells, investigations of pharmaceutical drug-cell interactions in eukaryotic cells and cancer diagnosis in whole tissue samples. Ultimately, this approach proves to be a flexible and robust tool for in vivo imaging on several length scales and provides comparable information as fluorescence-based imaging without the need of bulky fluorescent tags.

Real-time monitoring of live mycobacteria with a microfluidic acoustic-Raman platform

Tuberculosis (TB) remains a leading cause of death worldwide. Lipid rich, phenotypically antibiotic tolerant, bacteria are more resistant to antibiotics and may be responsible for relapse and the need for long-term TB treatment. We present a microfluidic system that acoustically traps live mycobacteria, M. smegmatis, a model organism for M. tuberculosis. We then perform optical analysis in the form of wavelength modulated Raman spectroscopy (WMRS) on the trapped M. smegmatis for up to eight hours, and also in the presence of isoniazid (INH). The Raman fingerprints of M. smegmatis exposed to INH change substantially in comparison to the unstressed condition. Our work provides a real-time assessment of the impact of INH on the increase of lipids in these mycobacteria, which could render the cells more tolerant to antibiotics. This microfluidic platform may be used to study any microorganism and to dynamically monitor its response to different conditions and stimuli.

Multivariate discrimination of heat shock proteins using a fiber optic Raman setup for in situ analysis of human perilymph

Raman spectroscopy has proven to be an effective tool for molecular analysis in different applications. In clinical diagnostics, its application has enabled nondestructive investigation of biological tissues and liquids. The human perilymph, for example, is an inner ear liquid, essential for the hearing sensation. The composition of this liquid is correlated with pathophysiological parameters and was analyzed by extraction and mass spectrometry so far. In this work, we present a fiber optic probe setup for the Raman spectroscopic sampling of inner ear proteins in solution. Multivariate data analysis is applied for the discrimination of individual proteins (heat shock proteins) linked to a specific type of hearing impairment. This proof-of-principle is a first step toward a system for sensitive and continuous in vivo perilymph investigation in the future.

Volumetric analysis of breast cancer tissues using machine learning and swept-source optical coherence tomography

In breast cancer, 20%-30% of cases require a second surgery because of incomplete excision of malignant tissues. Therefore, to avoid the risk of recurrence, accurate detection of the cancer margin by the clinician or surgeons needs some assistance. In this paper, an automated volumetric analysis of normal and breast cancer tissue is done by a machine learning algorithm to separate them into two classes. The proposed method is based on a support-vector-machine-based classifier by dissociating 10 features extracted from the A-line, texture, and phase map by the swept-source optical coherence tomographic intensity and phase images. A set of 88 freshly excised breast tissue [44 normal and 44 cancers (invasive ductal carcinoma tissues)] samples from 22 patients was used in our study. The algorithm successfully classifies the cancerous tissue with sensitivity, specificity, and accuracy of 91.56%, 93.86%, and 92.71% respectively. The present computational technique is fast, simple, and sensitive, and extracts features from the whole volume of the tissue, which does not require any special tissue preparation nor an expert to analyze the breast cancer as required in histopathology. Diagnosis of breast cancer by extracting quantitative features from optical coherence tomographic images could be a potentially powerful method for cancer detection and would be a valuable tool for a fine-needle-guided biopsy.