Papillary (PTC) and follicular (FTC) thyroid carcinomas require different treatment strategies, but their accurate differentiation remains a challenge in conventional histopathology. Misclassification can lead to overtreatment of low-risk PTCs or inadequate treatment of FTCs, increasing the risk of recurrence and metastasis. Since the structure of the collagen capsule surrounding thyroid nodules provides diagnostically valuable information, label-free imaging with second harmonic generation (SHG) microscopy combined with machine learning (ML)-based analysis offers a promising approach for automated classification. In this study, we extracted intensity and texture features from SHG images of thyroid nodules scanned as a whole and optimized several ML classifiers, including logistic regression, support vector classification (C-SVC), multilayer perceptron, random forest, XGBoost, and LightGBM, using hyperparameters tuning with stratified 10-fold cross-validation. One of the major challenges in classification was label noise resulting from 1) mislabeling of adjacent tissue, 2) PTC calcifications mimicking FTC features, and 3) capsule heterogeneity. To address this issue, we applied unsupervised segmentation to exclude mislabeled regions and consider capsular heterogeneity as a diagnostic feature. Recursive feature elimination and mutual information selection further refined the feature set and improved classification accuracy. Among all models, C-SVC achieved the highest accuracy (84.73%) with robust generalization to unknown data, significantly outperforming standard ML approaches (60-70%). These results demonstrate the feasibility of SHG microscopy-based ML classification as a reliable adjunct to existing histopathologic methods, which could improve diagnostic accuracy and patient outcomes.

Principal component analysis and K-means clustering of Raman spectra of glioma cells exposed to single-walled carbon nanotubes revealed their specific particle distribution, interaction with cell compartments, metabolism in autolysosomes, and cell death.

Papillary thyroid carcinoma (PTC) is one of the most common, well-differentiated carcinomas of the thyroid gland. PTC nodules are often surrounded by a collagen capsule that prevents the spread of cancer cells. However, as the malignant tumor progresses, the integrity of this protective barrier is compromised, and cancer cells invade the surroundings. The detection of capsular invasion is, therefore, crucial for the diagnosis and the choice of treatment and the development of new approaches aimed at the increase of diagnostic performance are of great importance. In the present study, we exploited the wide-field second harmonic generation (SHG) microscopy in combination with texture analysis and unsupervised machine learning (ML) to explore the possibility of quantitative characterization of collagen structure in the capsule and designation of different capsule areas as either intact, disrupted by invasion, or apt to invasion. Two-step k-means clustering showed that the collagen capsules in all analyzed tissue sections were highly heterogeneous and exhibited distinct segments described by characteristic ML parameter sets. The latter allowed a structural interpretation of the collagen fibers at the sites of overt invasion as fragmented and curled fibers with rarely formed distributed networks. Clustering analysis also distinguished areas in the PTC capsule that were not categorized as invasion sites by the initial histopathological analysis but could be recognized as prospective micro-invasions after additional inspection. The characteristic features of suspicious and invasive sites identified by the proposed unsupervised ML approach can become a reliable complement to existing methods for diagnosing encapsulated PTC, increase the reliability of diagnosis, simplify decision making, and prevent human-related diagnostic errors. In addition, the proposed automated ML-based selection of collagen capsule images and exclusion of non-informative regions can greatly accelerate and simplify the development of reliable methods for fully automated ML diagnosis that can be integrated into clinical practice.

Black silicon (bSi) is a highly absorptive material in the UV-vis and NIR spectral range. Photon trapping ability makes noble metal plated bSi attractive for fabrication of surface enhanced Raman spectroscopy (SERS) substrates. By using a cost-effective room temperature reactive ion etching method, we designed and fabricated the bSi surface profile, which provides the maximum Raman signal enhancement under NIR excitation when a nanometrically-thin gold layer is deposited. The proposed bSi substrates are reliable, uniform, low cost and effective for SERS-based detection of analytes, making these materials essential for medicine, forensics and environmental monitoring. Numerical simulation revealed that painting bSi with a defected gold layer resulted in an increase in the plasmonic hot spots, and a substantial increase in the absorption cross-section in the NIR range.

We propose a simple, fast, and low-cost method for producing Au-coated black Si-based SERS-active substrates with a proven enhancement factor of 106. Room temperature reactive ion etching of silicon wafer followed by nanometer-thin gold sputtering allows the formation of a highly developed lace-type Si surface covered with homogeneously distributed gold islands. The mosaic structure of deposited gold allows the use of Au-uncovered Si domains for Raman peak intensity normalization. The fabricated SERS substrates have prominent uniformity (with less than 6% SERS signal variations over large areas, 100 × 100 μm2). It has been found that the storage of SERS-active substrates in an ambient environment reduces the SERS signal by less than 3% in 1 month and not more than 40% in 20 months. We showed that Au-coated black Si-based SERS-active substrates can be reused after oxygen plasma cleaning and developed relevant protocols for removing covalently bonded and electrostatically attached molecules. Experiments revealed that the Raman signal of 4-MBA molecules covalently bonded to the Au coating measured after the 10th cycle was just 4 times lower than that observed for the virgin substrate. A case study of the reusability of the black Si-based substrate was conducted for the subsequent detection of 10-5 M doxorubicin, a widely used anticancer drug, after the reuse cycle. The obtained SERS spectra of doxorubicin were highly reproducible. We demonstrated that the fabricated substrate permits not only qualitative but also quantitative monitoring of analytes and is suitable for the determination of concentrations of doxorubicin in the range of 10-9-10-4 M. Reusable, stable, reliable, durable, low-cost Au-coated black Si-based SERS-active substrates are promising tools for routine laboratory research in different areas of science and healthcare.

Carbon nanotubes (CNTs) have a high potential for biomedical applications such as theranostics and image-guided drug delivery. Raman spectroscopy and microscopy are powerful tools to visualize and monitor the quality of CNTs and evaluate their distribution inside cells, providing valuable information for the development of effective carbon nanotube-based therapies. However, the features obtained from Raman spectra are quite dependent on each other, and their simultaneous presence is often redundant. In this study, Raman spectral mapping of C6 glioma cells accumulating two types of functionalized CNTs over time is performed, and principal component analysis is used to extract valuable information from big datasets of Raman spectra about specificity of CNT distribution, barely noticeable structural property modification, and intracellular degradation of CNTs.

Monitoring of tiny intracell temperature variations is of high importance to understand the mechanisms of exothermic/endothermic processes inside the living cells. Small shifts in thermal balance may drastically influence cell functioning and induce pathological conditions. By using biocompatible diamond single‐crystal microneedles enriched with nitrogen‐vacancy (NV)/silicon‐vacancy (SiV) color centers, this study demonstrates all‐optical in vitro temperature monitoring in the physiologically significant range (25–55 °C). Zero‐phonon line (ZPL) of SiV centers belonging to the “therapeutic window” is used to improve measurement precision via suppression of the tissue autofluorescence. The simultaneous detection of the NV and SiV fluorescence enables two‐band visualization of the living cells combined with the temperature sensing. This study demonstrates experimentally that temperature can be measured by lifetime, full‐width at half maximum, and peak position of SiV ZPL, while accuracy can be further improved by normalizing the photoluminescence (PL) ZPL peak intensity on the PL signal measured at the wavelength where it is temperature independent. According to performed numerical simulations diamond microneedles enable real‐time temperature measurements because their characteristic heating time is less than 10 ns. The results open a way toward accurate, noninvasive, precise, and real‐time monitoring of temperature variations accompanying intracellular biochemical reactions and processes on the single‐cell level.

The paper describes methods for theoretical and numerical simulation of the photoacoustic effect that occurs in one-dimensional carbon micro- and nanostructures under an action of pulsed laser radiation. The proposed numerical modeling technique is based on solving the equations of motion of continuous media in the Lagrange form for spatially inhomogeneous media. This model makes it possible to calculate fields of temperature, pressure, density, and velocity of the medium depending on the parameters of laser pulses and characteristics of micro- and nanostructures.

Single-walled carbon nanotubes (SWCNTs) demonstrate a strong potential as an optically activated theranostic nano-agent. However, using SWCNTs in theranostics still requires revealing mechanisms of the SWCNT-mediated effects on cellular functions. Even though rapid and delayed cellular responses can differ significantly and may lead to undesirable consequences, understanding of these mechanisms is still incomplete. We demonstrate that introducing short (150-250 nm) SWCNTs into C6 rat glioma cells leads to SWCNT-driven effects that show pronounced time dependence. Accumulation of SWCNTs is carried out due to endocytosis with modification of the actin cytoskeleton but not accompanied with autophagy. Its initial stage launches a rapid cellular response via significantly heightened mitochondrial membrane potential and superoxide anion radical production, satisfying the cell demand of energy for SWCNT transfer inside the cytoplasm. In the long term, SWCNTs agglomerate to micron-sized structures surrounded by highly active mitochondria having parameters return to control values. SWCNTs postponed effects are also manifested themselves in the suppression of the cell proliferative activity with further restoration after five passages. These results demonstrate relative cellular inertness and safety of SWCNTs eliminating possible side effects caused by optically activated theranostic applications.