Search Results (54)

Background/Objectives: Cervical Intraepithelial Neoplasia (CIN) is a significant risk factor for the development of invasive cancer, and the histological detection of High-Grade CIN (CIN2+) during screening generally indicates the need for surgical removal of the lesion; cervical conization is the current gold standard of treatment. The recurrence risk for disease is reported to be up to 30%, based on data in the literature. Follow-up protocols mainly rely on High-Risk Human Papillomavirus (hrHPV) detection at six months post-treatment; if negative, this is considered the test of cure. This approach assumes that all patients have an equal risk of disease recurrence, regardless of individual characteristics. The objective of this study was to evaluate the individual recurrence risk using a mathematical model, analyzing the weight of various parameters and their associations in terms of recurrence development. Methods: We retrospectively examined 428 patients treated for CIN2+ at San Raffaele Hospital in Milan between January 2010 and April 2019. Clinical and pathological data were recorded and correlated with disease recurrence; three different variables, known to behave as significant prognostic factors, were analyzed: hrHPV persistence, the surgical margin status, Neutrophil–Lymphocyte Ratio (NLR), along with their relative associations. Data were used to engineer a mathematical model for the identification of different risk classes, allowing for the risk stratification of cases. Results: Surgical margins status, hrHPV persistence, and a high NLR index were demonstrated to act as independent and significant risk factors for disease recurrence, and their different associations significantly correlated with different recurrence rates. The mathematical model identified eight classes of recurrence probability, with Odds Ratios (ORs) ranging from 7.48% to 69.4%. Conclusions: The developed mathematical model may allow risk stratification for recurrence in a hierarchical fashion, potentially supporting the tailored management of follow-up, and improving the current protocols. This study represents the first attempt to integrate these factors into a mathematical model for post-treatment risk stratification. Full article

Pharmaceutical scientists have researched the potential of secondary metabolites biosynthesized by microorganisms as active pharmaceutical ingredients (APIs) for the treatment of cancer. Ideally, these APIs should possess anticancer bioactivity that specifically targets tumor cells while having little cytotoxic effect on healthy tissue. Biosurfactants are microbial secondary metabolites with surface-active properties and individual bioactivities that have the potential to either destroy cancer cells in a targeted fashion or prevent tumor cell formation. Currently, the best-studied class of microbial biosurfactants for the purpose of anticancer bioactivity is glycolipids, which contain a hydrophilic sugar moiety bonded to a hydrophobic fatty acid. Anticancer investigations are mainly carried out using in vitro models that show that compounds belonging to each of the four sub-classes of microbial glycolipid have significant anticancer bioactivity. The targeted action of this activity appears to be highly dependent on a specific congener molecular structure with nuanced alterations in structure leading to the killing of both tumor and healthy cells. This review compiles the current literature relating to glycolipid anticancer activity and provides a critical appraisal of exploiting the bioactivity of these compounds as novel anticancer agents. Finally, we propose several suggestions on how this research could be improved moving forward via method standardization. Full article

This work reports the design of a 1x2 photonic digital switch controlled by an electrically induced metasurface, configurated by a rectangular array of points where the refractive index is locally changed through the application of an external bias. The device is simulated using the Beam Propagation Method (BPM) and Finite Difference Time Domain (FDTD) algorithms and the structure under evaluation is an amorphous silicon 1x2 multimode interference (MMI), joined to an arrayed Metal Oxide Semiconductor (MOS) structure Al/SiNx/a-Si:H/ITO to be used in active-matrix pixel fashion to control the output of the switch. MMI couplers, based on self-imaging multimode waveguides, are very compact integrated optical components that can perform many different splitting and recombining functions. The input–output model has been defined using a machine learning approach; a high number of images have been generated through simulations, based on the beam propagation algorithm, obtaining a large dataset for an MMI structure under different activation maps of the MOS pixels. This dataset has been used for training and testing of a machine learning algorithm for the classification of the MMI configuration in terms of binary digital output for a 1x2 switch. Also, a statistical analysis has been produced, targeting the definition of the most incident-activated pixel for each switch operation. An optimal configuration is proposed and applied to demonstrate the operation of a digital cascaded switch. This proof of concept paves the way to a more complex device class, supporting the recent advances in programmable photonic integrated circuits. Full article

Large language models (LLMs) have demonstrated remarkable capabilities in text generation, which also raise numerous concerns about their potential misuse, especially in educational exercises and academic writing. Accurately identifying and tracing the origins of LLM-generated content is crucial for accountability and transparency, ensuring the responsible use of LLMs in educational and academic environments. Previous methods utilize binary classifiers to discriminate whether a piece of text was written by a human or generated by a specific LLM or employ multi-class classifiers to trace the source LLM from a fixed set. These methods, however, are restricted to one or several pre-specified LLMs and cannot generalize to new LLMs, which are continually emerging. This study formulates source LLM tracing in a class-incremental learning (CIL) fashion, where new LLMs continually emerge, and a model incrementally learns to identify new LLMs without forgetting old ones. A training-free continual learning method is further devised for the task, the idea of which is to continually extract prototypes for emerging LLMs, using a frozen encoder, and then to perform origin tracing via prototype matching after a delicate decorrelation process. For evaluation, two datasets are constructed, one in English and one in Chinese. These datasets simulate a scenario where six LLMs emerge over time and are used to generate student essays, and an LLM detector has to incrementally expand its recognition scope as new LLMs appear. Experimental results show that the proposed method achieves an average accuracy of 97.04% on the English dataset and 91.23% on the Chinese dataset. These results validate the feasibility of continual origin tracing of LLM-generated text and verify its effectiveness in detecting cheating in student coursework. Full article

There are multiple uses for single-channel images, such as infrared imagery, depth maps, and others. To automatically classify objects in such images, an algorithm suited for single-channel image processing is required. This study explores the application of deep learning techniques for the recognition of wild animals using infrared images. Traditional methods of wildlife monitoring often rely on visible light imaging, which can be hindered by various environmental factors such as darkness, fog, and dense foliage. In contrast, infrared imaging captures the thermal signatures of animals, providing a robust alternative for wildlife detection and identification. We test a Convolutional Neural Network (CNN) model specifically designed to analyze infrared images, leveraging the unique thermal patterns emitted by different animal species. The model is trained and tested on a diverse dataset of infrared images, demonstrating high accuracy in distinguishing between multiple species. In this paper, we also present a comparison of several well-known artificial neural networks on this data. To ensure accurate testing, we introduce a new dataset containing infrared photos of Slovak wildlife, specifically including classes such as bear, deer, boar, and fox. To complement this dataset, the Fashion MNIST dataset was also used. Our results indicate that deep learning approaches significantly enhance the capability of infrared imaging for wildlife monitoring, offering a reliable and efficient tool for conservation efforts and ecological studies. Full article

Background: This study aimed to evaluate the anti-inflammatory effects of a Zephyranthes rosea in an ovalbumin-induced asthma model. Methods: Allergic asthma was induced in mice via intraperitoneal injection, followed by intranasal ovalbumin challenge. Methanolic extract of Z. rosea bulb was orally administered to asthmatic mice for 14 days. Hematological parameters for bronchoalveolar lavage fluid (BALF) and blood were analyzed. The mRNA expression levels of interleukins and transforming growth factor beta (TGF-β1) in lung tissues were determined using reverse transcriptase–polymerase chain reaction (RT–PCR). Network pharmacology analysis was used to find possible Z. rosea targets. After building a protein–protein interaction network to find hub genes, GO and KEGG enrichment analyses were carried out to determine the potential mechanism. In silico analysis was performed by Molecular Operating Environment. Results: GC-MS analysis of Z. rosea extract detected major classes of phytochemicals. Hematological parameters in blood and BALF from Z. rosea extract-treated animals were significantly reduced in a dose-dependent fashion. Histopathology revealed that Z. rosea bulb had an ameliorative effect on lung tissues. Moreover, treatment with Z. rosea bulb extract significantly restored the normal levels of IL-4, IL-6, IL-1β, IL-10, IL-13, and TGF-β1 in allergic asthmatic mice compared to the diseased group. In silico analysis, particularly of the binding affinities of Z. rosea bulb phytoconstituents for IL6, AKT1, and Src, supported in vivo results. Conclusions: These findings indicated that Z. rosea bulb extract significantly ameliorates cellular and molecular biomarkers of bronchial inflammation and could be a potential candidate for treating allergic asthma. Full article

Classifying imbalanced data is important due to the significant practical value of accurately categorizing minority class samples, garnering considerable interest in many scientific domains. This study primarily uses analysis of variance (ANOVA) to investigate the main and interaction effects of different parameters on imbalanced data, aiming to optimize convolutional neural network (CNN) parameters to improve minority class sample recognition. The CIFAR-10 and Fashion-MNIST datasets are used to extract samples with imbalance ratios of 25:1, 15:1, and 1:1. To thoroughly assess model performance on imbalanced data, we employ various evaluation metrics, such as accuracy, recall, F1 score, P-mean, and G-mean. In highly imbalanced datasets, optimizing the learning rate significantly affects all performance metrics. The interaction between the learning rate and kernel size significantly impacts minority class samples in moderately imbalanced datasets. Through parameter optimization, the accuracy of the CNN model on the 25:1 highly imbalanced CIFAR-10 and Fashion-MNIST datasets improves by 14.20% and 5.19% compared to the default model and by 8.21% and 3.87% compared to the undersampling model, respectively, while also enhancing other evaluation metrics for minority classes. Full article

Aiming at tracking sharply maneuvering targets, this paper develops novel variational adaptive state estimators for joint target state and process noise parameter estimation for a class of linear state-space models with abruptly changing parameters. By combining variational inference with change-point detection in an online Bayesian fashion, two adaptive estimators—a change-point-based adaptive Kalman filter (CPAKF) and a change-point-based adaptive Kalman smoother (CPAKS)—are proposed in a recursive detection and estimation process. In each iteration, the run-length probability of the current maneuver mode is first calculated, and then the joint posterior of the target state and process noise parameter conditioned on the run length is approximated by variational inference. Compared with existing variational noise-adaptive Kalman filters, the proposed methods are robust to initial iterative value settings, improving their capability of tracking sharply maneuvering targets. Meanwhile, the change-point detection divides the non-stationary time sequence into several stationary segments, allowing for an adaptive sliding length in the CPAKS method. The tracking performance of the proposed methods is investigated using both synthetic and real-world datasets of maneuvering targets. Full article

Deep learning based visual cognition has greatly improved the accuracy of defect detection, reducing processing times and increasing product throughput across a variety of manufacturing use cases. There is however a continuing need for rigorous procedures to dynamically update model-based detection methods that use sequential streaming during the training phase. This paper reviews how new process, training or validation information is rigorously incorporated in real time when detection exceptions arise during inspection. In particular, consideration is given to how new tasks, classes or decision pathways are added to existing models or datasets in a controlled fashion. An analysis of studies from the incremental learning literature is presented, where the emphasis is on the mitigation of process complexity challenges such as, catastrophic forgetting. Further, practical implementation issues that are known to affect the complexity of deep learning model architecture, including memory allocation for incoming sequential data or incremental learning accuracy, is considered. The paper highlights case study results and methods that have been used to successfully mitigate such real-time manufacturing challenges. Full article

There is increasing pressure on young consumers to practice sustainable consumption. With young women being key agents in fashion consumption, switching their purchasing to secondhand clothing over new is instrumental to reducing textile waste. This study applied the Theory of Planned Behaviour and Prototype Willingness Model to identify key drivers informing young women’s secondhand clothing purchasing decisions. Young Australian women (N = 48) completed qualitative surveys assessing their underlying attitudinal, normative, and control beliefs and perceived images of typical secondhand clothing shoppers. Thematic analysis indicated the main benefits of secondhand clothing purchasing to be the environmental impact and cost savings, with drawbacks being quality issues, reduced shopping experience, and greater effort required. Clothing diversity was both positive (‘unique finds’) and had a downside (limited sizes). Approvers of secondhand purchasing were mainly friends and family, with older relatives being less supportive. Key barriers were increased prices for quality items and the time required to locate them. Images of typical secondhand clothes shoppers were generally positive (‘cool’, ‘thrifty’, ‘unique’, ‘eco-friendly’), while ‘materialistic’, ‘upper-class’, and ‘ignorant’ but also ‘trendy’ indicated mixed perceptions about those who did not. Crucial in our findings was clarifying the intersections and contextual context of participants’ responses. Identifying the nuances in the underlying beliefs driving young women’s fashion choices assists in theory-informed strategies to encourage sustainable consumption of clothing. Full article

Ensuring road safety, structural stability and durability is of paramount importance, and detecting road cracks plays a critical role in achieving these goals. We propose a GM-ResNet-based method to enhance the precision and efficacy of crack detection. Leveraging ResNet-34 as the foundational network for crack image feature extraction, we consider the challenge of insufficient global and local information assimilation within the model. To overcome this, we incorporate the global attention mechanism into the architecture, facilitating comprehensive feature extraction across the channel and the spatial width and height dimensions. This dynamic interaction across these dimensions optimizes feature representation and generalization, resulting in a more precise crack detection outcome. Recognizing the limitations of ResNet-34 in managing intricate data relationships, we replace its fully connected layer with a multilayer fully connected neural network. We fashion a deep network structure by integrating multiple linear, batch normalization and activation function layers. This construction amplifies feature expression, stabilizes training convergence and elevates the performance of the model in complex detection tasks. Moreover, tackling class imbalance is imperative in road crack detection. Introducing the focal loss function as the training loss addresses this challenge head-on, effectively mitigating the adverse impact of class imbalance on model performance. The experimental outcomes on a publicly available crack dataset emphasize the advantages of the GM-ResNet in crack detection accuracy compared to other methods. It is worth noting that the proposed method has better evaluation indicators in the detection results compared with alternative methodologies, highlighting its effectiveness. This validates the potency of our method in achieving optimal crack detection outcomes. Full article

HPV-associated oropharynx carcinoma (HPVOPC) tumors have a relatively low mutational burden. Elucidating the relative contributions of other tumor alterations, such as DNA methylation alterations, alternative splicing events (ASE), and copy number variation (CNV), could provide a deeper understanding of carcinogenesis drivers in this disease. We applied network propagation analysis to multiple classes of tumor alterations in a discovery cohort of 46 primary HPVOPC tumors and 25 cancer-unaffected controls and validated our findings with TCGA data. We identified significant overlap between differential gene expression networks and all alteration classes, and this association was highest for methylation and lowest for CNV. Significant overlap was seen for gene clusters of G protein-coupled receptor (GPCR) pathways. HPV16–human protein interaction analysis identified an enriched cluster defined by an immune-mediated GPCR signal, including CXCR3 cytokines CXCL9, CXCL10, and CXCL11. CXCR3 was found to be expressed in primary HPVOPC, and scRNA-seq analysis demonstrated CXCR3 ligands to be highly expressed in M2 macrophages. In vivo models demonstrated decreased tumor growth with antagonism of the CXCR3 receptor in immunodeficient but not immunocompetent mice, suggesting that the CXCR3 axis can drive tumor proliferation in an autocrine fashion, but the effect is tempered by an intact immune system. In conclusion, methylation, ASE, and SNV alterations are highly associated with network gene expression changes in HPVOPC, suggesting that ASE and methylation alterations have an important role in driving the oncogenic phenotype. Network analysis identifies GPCR networks, specifically the CXCR3 chemokine axis, as modulators of tumor–immune interactions that may have proliferative effects on primary tumors as well as a role for immunosurveillance; however, CXCR3 inhibition should be used with caution, as these agents may both inhibit and stimulate tumor growth considering the competing effects of this cytokine axis. Further investigation is needed to explore opportunities for targeted therapy in this setting. Full article

Bromodomains (BD) are epigenetic readers of histone acetylation involved in chromatin remodeling and transcriptional regulation of several genes including protooncogene cellular myelocytomatosis (c-Myc). c-Myc is difficult to target directly by agents due to its disordered alpha helical protein structure and predominant nuclear localization. The epigenetic targeting of c-Myc by BD inhibitors is an attractive therapeutic strategy for prostate cancer (PC) associated with increased c-Myc upregulation with advancing disease. MT-1 is a bivalent BD inhibitor that is 100-fold more potent than the first-in-class BD inhibitor JQ1. MT-1 decreased cell viability and causes cell cycle arrest in G0/G1 phase in castration-sensitive and resistant PC cell lines in a dose-dependent fashion. The inhibition of c-Myc function by MT-1 was molecularly corroborated by the de-repression of Protein Kinase D1 (PrKD) and increased phosphorylation of PrKD substrate proteins: threonine 120, serine 11, and serine 216 amino acid residues in β-Catenin, snail, and cell division cycle 25c (CDC25c) proteins, respectively. The treatment of 3D cell cultures derived from three unique clinically annotated heavily pretreated patient-derived PC xenografts (PDX) mice models with increasing doses of MT-1 demonstrated the lowest IC₅₀ in tumors with c-Myc amplification and clinically resistant to Docetaxel, Cabazitaxel, Abiraterone, and Enzalutamide. An intraperitoneal injection of either MT-1 or in combination with 3jc48-3, an inhibitor of obligate heterodimerization with MYC-associated protein X (MAX), in mice implanted with orthotopic PC PDX, decreased tumor growth. This is the first pre-clinical study demonstrating potential utility of MT-1 in the treatment of PC with c-Myc dysregulation. Full article

Gene expression data are usually known for having a large number of features. Usually, some of these features are irrelevant and redundant. However, in some cases, all features, despite being numerous, show high importance and contribute to the data analysis. In a similar fashion, gene expression data sometimes have limited instances with a high rate of imbalance among the classes. This can limit the exposure of a classification model to instances of different categories, thereby influencing the performance of the model. In this study, we proposed a cancer detection approach that utilized data preprocessing techniques such as oversampling, feature selection, and classification models. The study used SVMSMOTE for the oversampling of the six examined datasets. Further, we examined different techniques for feature selection using dimension reduction methods and classifier-based feature ranking and selection. We trained six machine learning algorithms, using repeated 5-fold cross-validation on different microarray datasets. The performance of the algorithms differed based on the data and feature reduction technique used. Full article

This paper presents a comprehensive study of Convolutional Neural Networks (CNN) and transfer learning in the context of medical imaging. Medical imaging plays a critical role in the diagnosis and treatment of diseases, and CNN-based models have demonstrated significant improvements in image analysis and classification tasks. Transfer learning, which involves reusing pre-trained CNN models, has also shown promise in addressing challenges related to small datasets and limited computational resources. This paper reviews the advantages of CNN and transfer learning in medical imaging, including improved accuracy, reduced time and resource requirements, and the ability to address class imbalances. It also discusses challenges, such as the need for large and diverse datasets, and the limited interpretability of deep learning models. What factors contribute to the success of these networks? How are they fashioned, exactly? What motivated them to build the structures that they did? Finally, the paper presents current and future research directions and opportunities, including the development of specialized architectures and the exploration of new modalities and applications for medical imaging using CNN and transfer learning techniques. Overall, the paper highlights the significant potential of CNN and transfer learning in the field of medical imaging, while also acknowledging the need for continued research and development to overcome existing challenges and limitations. Full article