Micro- and nano-sensors lie at the heart of critical innovation in fields ranging from medical to environmental sciences. In recent years, there has been a significant improvement in sensor design along with the advances in micro- and nano-fabrication technology and the use of newly designed materials, leading to the development of high-performance gas sensors. Advanced micro- and nano-fabrication technology enables miniaturization of these sensors into micro-sized gas sensor arrays while maintaining the sensing performance. These capabilities facilitate the development of miniaturized integrated gas sensor arrays that enhance both sensor sensitivity and selectivity towards various analytes. In the past, several micro- and nano-gas sensors have been proposed and investigated where each type of sensor exhibits various advantages and limitations in sensing resolution, operating power, response, and recovery time. This paper presents an overview of the recent progress made in a wide range of gas-sensing technology. The sensing functionalizing materials, the advanced micro-machining fabrication methods, as well as their constraints on the sensor design, are discussed. The sensors’ working mechanisms and their structures and configurations are reviewed. Finally, the future development outlook and the potential applications made feasible by each category of the sensors are discussed.
Konstantinos Keremis, Eleni Vrochidou, George A. Papakostas
The ability of deep learning models to maintain consistent performance under image transformations-termed invariances, is critical for reliable deployment across diverse computer vision applications. This study presents a comprehensive empirical evaluation of modern convolutional neural networks (CNNs) and vision transformers (ViTs) concerning four fundamental types of image invariances: blur, noise, rotation, and scale. We analyze a curated selection of thirty models across three common vision tasks, object localization, recognition, and semantic segmentation, using benchmark datasets including COCO, ImageNet, and a custom segmentation dataset. Our experimental protocol introduces controlled perturbations to test model robustness and employs task-specific metrics such as mean Intersection over Union (mIoU), and classification accuracy (Acc) to quantify models’ performance degradation. Results indicate that while ViTs generally outperform CNNs under blur and noise corruption in recognition tasks, both model families exhibit significant vulnerabilities to rotation and extreme scale transformations. Notably, segmentation models demonstrate higher resilience to geometric variations, with SegFormer and Mask2Former emerging as the most robust architectures. These findings challenge prevailing assumptions regarding model robustness and provide actionable insights for designing vision systems capable of withstanding real-world input variability.
Photography, Computer applications to medicine. Medical informatics
Shery Jacob, Fathima Sheik Kather, Sai H. S. Boddu
et al.
Natural substances, especially those derived from plants, exhibit a diverse range of therapeutic benefits, such as antioxidant, anti-inflammatory, anticancer, and antimicrobial effects. Nevertheless, their use in clinical settings is frequently impeded by inadequate solubility, limited bioavailability, and instability. Nanovesicular carriers, such as liposomes, niosomes, ethosomes, transferosomes, transethosomes, and cubosomes, have emerged as innovative phytochemical delivery systems to address these limitations. This review highlights recent developments in vesicular nanocarriers for phytochemical delivery, emphasizing preparation techniques, composition, therapeutic applications, and the future potential of these systems. Phytosomes, along with their key advantages and various preparation techniques, are extensively described. Various in vitro and in vivo characterization techniques utilized for evaluating these nanovesicular carriers are summarized. Completed clinical trials and patents granted for nanovesicles encapsulating phytochemicals designed for systemic delivery are tabulated. Phytochemical delivery via vesicular carriers faces challenges such as low stability, limited active loading, scalability issues, and high production costs. Additionally, immune clearance and regulatory hurdles hinder clinical application, requiring improved carrier design and formulation techniques.
This paper explores the transfer of knowledge from general vision models pretrained on 2D natural images to improve 3D medical image segmentation. We focus on the semi-supervised setting, where only a few labeled 3D medical images are available, along with a large set of unlabeled images. To tackle this, we propose a model-agnostic framework that progressively distills knowledge from a 2D pretrained model to a 3D segmentation model trained from scratch. Our approach, M&N, involves iterative co-training of the two models using pseudo-masks generated by each other, along with our proposed learning rate guided sampling that adaptively adjusts the proportion of labeled and unlabeled data in each training batch to align with the models' prediction accuracy and stability, minimizing the adverse effect caused by inaccurate pseudo-masks. Extensive experiments on multiple publicly available datasets demonstrate that M&N achieves state-of-the-art performance, outperforming thirteen existing semi-supervised segmentation approaches under all different settings. Importantly, ablation studies show that M&N remains model-agnostic, allowing seamless integration with different architectures. This ensures its adaptability as more advanced models emerge. The code is available at https://github.com/pakheiyeung/M-N.
This short paper provides a means to classify augmentation technologies to reconceptualize them as sociotechnical, discursive and rhetorical phenomena, rather than only through technological classifications. It identifies a set of value systems that constitute augmentation technologies within discourses, namely, the intent to enhance, automate, and build efficiency. This short paper makes a contribution to digital literacy surrounding augmentation technology emergence, as well as the more specific area of AI literacy, which can help identify unintended consequences implied at the design stages of these technologies.
Medical image registration is a fundamental task in medical image analysis, aiming to establish spatial correspondences between paired images. However, existing unsupervised deformable registration methods rely solely on intensity-based similarity metrics, lacking explicit anatomical knowledge, which limits their accuracy and robustness. Vision foundation models, such as the Segment Anything Model (SAM), can generate high-quality segmentation masks that provide explicit anatomical structure knowledge, addressing the limitations of traditional methods that depend only on intensity similarity. Based on this, we propose a novel SAM-assisted registration framework incorporating prototype learning and contour awareness. The framework includes: (1) Explicit anatomical information injection, where SAM-generated segmentation masks are used as auxiliary inputs throughout training and testing to ensure the consistency of anatomical information; (2) Prototype learning, which leverages segmentation masks to extract prototype features and aligns prototypes to optimize semantic correspondences between images; and (3) Contour-aware loss, a contour-aware loss is designed that leverages the edges of segmentation masks to improve the model's performance in fine-grained deformation fields. Extensive experiments demonstrate that the proposed framework significantly outperforms existing methods across multiple datasets, particularly in challenging scenarios with complex anatomical structures and ambiguous boundaries. Our code is available at https://github.com/HaoXu0507/IPMI25-SAM-Assisted-Registration.
Artificial intelligence (AI) is expected to revolutionize the practice of medicine. Recent advancements in the field of deep learning have demonstrated success in a variety of clinical tasks: detecting diabetic retinopathy from images, predicting hospital readmissions, aiding in the discovery of new drugs, etc. AI's progress in medicine, however, has led to concerns regarding the potential effects of this technology upon relationships of trust in clinical practice. In this paper, I will argue that there is merit to these concerns, since AI systems can be relied upon, and are capable of reliability, but cannot be trusted, and are not capable of trustworthiness. Insofar as patients are required to rely upon AI systems for their medical decision-making, there is potential for this to produce a deficit of trust in relationships in clinical practice.
Vision foundation models have demonstrated strong generalization in medical image segmentation by leveraging large-scale, heterogeneous pretraining. However, they often struggle to generalize to specialized clinical tasks under limited annotations or rare pathological variations, due to a mismatch between general priors and task-specific requirements. To address this, we propose Uncertainty-informed Collaborative Learning (UnCoL), a dual-teacher framework that harmonizes generalization and specialization in semi-supervised medical image segmentation. Specifically, UnCoL distills both visual and semantic representations from a frozen foundation model to transfer general knowledge, while concurrently maintaining a progressively adapting teacher to capture fine-grained and task-specific representations. To balance guidance from both teachers, pseudo-label learning in UnCoL is adaptively regulated by predictive uncertainty, which selectively suppresses unreliable supervision and stabilizes learning in ambiguous regions. Experiments on diverse 2D and 3D segmentation benchmarks show that UnCoL consistently outperforms state-of-the-art semi-supervised methods and foundation model baselines. Moreover, our model delivers near fully supervised performance with markedly reduced annotation requirements.
Pedro M. Gordaliza, Nataliia Molchanova, Jaume Banus
et al.
Deep learning models for medical image segmentation suffer significant performance drops due to distribution shifts, but the causal mechanisms behind these drops remain poorly understood. We extend causal attribution frameworks to high-dimensional segmentation tasks, quantifying how acquisition protocols and annotation variability independently contribute to performance degradation. We model the data-generating process through a causal graph and employ Shapley values to fairly attribute performance changes to individual mechanisms. Our framework addresses unique challenges in medical imaging: high-dimensional outputs, limited samples, and complex mechanism interactions. Validation on multiple sclerosis (MS) lesion segmentation across 4 centers and 7 annotators reveals context-dependent failure modes: annotation protocol shifts dominate when crossing annotators (7.4% $\pm$ 8.9% DSC attribution), while acquisition shifts dominate when crossing imaging centers (6.5% $\pm$ 9.1%). This mechanism-specific quantification enables practitioners to prioritize targeted interventions based on deployment context.
Jih-Kai Huang, Ping-Hsun Wu, Zhao-Feng Chen
et al.
Microbiota tryptophan metabolism and the biosynthesis of indole derivatives play an important role in homeostasis and pathogenesis in the human body and can be affected by the gut microbiota. However, studies on the interplay between gut microbiota and tryptophan metabolites in patients undergoing dialysis are lacking. This study aimed to identify the gut microbiota, the indole pathway in tryptophan metabolism, and significant functional differences in ESRD patients with regular hemodialysis. We performed the shotgun metagenome sequencing of stool samples from 85 hemodialysis patients. Using the linear discriminant analysis effect size (LEfSe), we examined the composition of the gut microbiota and metabolic features across varying concentrations of tryptophan and indole metabolites. Higher tryptophan levels promoted tyrosine degradation I and pectin degradation I metabolic modules; lower tryptophan levels were associated with glutamate degradation I, fructose degradation, and valine degradation modules. Higher 3-indoxyl sulfate concentrations were characterized by alanine degradation I, anaerobic fatty acid beta-oxidation, sulfate reduction, and acetyl-CoA to crotonyl-CoA. Contrarily, lower 3-indoxyl sulfate levels were related to propionate production III, arabinoxylan degradation, the Entner–Doudoroff pathway, and glutamate degradation II. The present study provides a better understanding of the interaction between tryptophan, indole metabolites, and the gut microbiota as well as their gut metabolic modules in ESRD patients with regular hemodialysis.
This study embarks on a comprehensive investigation of the effectiveness of repetitive transcranial direct current stimulation (tDCS)-based neuromodulation in augmenting steady-state visual evoked potential (SSVEP) brain-computer interfaces (BCIs), alongside exploring pertinent electroencephalography (EEG) biomarkers for assessing brain states and evaluating tDCS efficacy. EEG data were garnered across three distinct task modes (eyes open, eyes closed, and SSVEP stimulation) and two neuromodulation patterns (sham-tDCS and anodal-tDCS). Brain arousal and brain functional connectivity were measured by extracting features of fractal EEG and information flow gain, respectively. Anodal-tDCS led to diminished offsets and enhanced information flow gains, indicating improvements in both brain arousal and brain information transmission capacity. Additionally, anodal-tDCS markedly enhanced SSVEP-BCIs performance as evidenced by increased amplitudes and accuracies, whereas sham-tDCS exhibited lesser efficacy. This study proffers invaluable insights into the application of neuromodulation methods for bolstering BCI performance, and concurrently authenticates two potent electrophysiological markers for multifaceted characterization of brain states.
Abstract Killer cell lectin-like receptor B1 (KLRB1) is implicated in cancer progression and immunity. In this study, we aimed to evaluate the expression levels of KLRB1 in lung adenocarcinoma (LUAD) and analyze the relationship between KLRB1 expression levels, LUAD progression, and the tumor immune microenvironment. KLRB1 levels in LUAD were analyzed using data from the TCGA and XENA databases. Additionally, the diagnostic values of KLRB1 were analyzed in patients with LUAD. Survival and meta-analyses were employed to investigate the relationship between KLRB1 levels and other prognostic factors in patients with LUAD. Bioinformatics and cellular experiments were used to understand the functions and mechanisms of KLRB1. In addition, correlation analysis was used to investigate the relationship between KLRB1 levels and the immune microenvironment in LUAD. Reduced KLRB1 expression in LUAD was found to positively correlate with tumor size, distant metastasis, pathological stage, age, overall survival, diagnostic value, and disease-specific survival in patients with LUAD (P < 0.05). Conversely, increased KLRB1 expression was found to positively correlate with the overall survival and disease-specific survival in patients with LUAD (P < 0.05). We also found that the overexpression of KLRB1 can inhibit the proliferation, migration, and invasion of LUAD cells and promote apoptosis. KLRB1 was involved in immune cell differentiation, NF-kB, PD-L1, and PD-1 checkpoint pathways and others. Additionally, KLRB1 expression was linked to tumor purity, stromal, immune, and estimate scores, the levels of immune cells including B cells, CD8+ T cells, and CD4+ T cells, and immune cell markers in LUAD. Reduced KLRB1 expression has a significant positive correlation with diagnosis, poor prognosis, and immunity to cancer in patients with LUAD. KLRB1 inhibited cell proliferation and migration in patients with LUAD. These results suggest that KLRB1 may serve as a potential therapeutic target in patients with LUAD.
The clinical application of cancer immunotherapy is unsatisfied due to low response rates and systemic immune-related adverse events. Microwave hyperthermia can be used as a synergistic immunotherapy to amplify the antitumor effect. Herein, we designed a Gd-based metal-organic framework (Gd-MOF) nanosystem for MRI-guided thermotherapy and synergistic immunotherapy, which featured high performance in drug loading and tumor tissue penetration. The PD-1 inhibitor (aPD-1) was initially loaded in the porous Gd-MOF (Gd/M) nanosystem. Then, the phase change material (PCM) and the cancer cell membrane were further sequentially modified on the surface of Gd/MP to obtain Gd-MOF@aPD-1@CM (Gd/MPC). When entering the tumor microenvironment (TME), Gd/MPC induces immunogenic death of tumor cells through microwave thermal responsiveness, improves tumor suppressive immune microenvironment and further enhances anti-tumor ability of T cells by releasing aPD-1. Meanwhile, Gd/MPC can be used for contrast-enhanced MRI. Transcriptomics data revealed that the downregulation of MSK2 in cancer cells leads to the downregulation of c-fos and c-jun, and ultimately leads to the apoptosis of cancer cells after treatment. In general, Gd/MPC nanosystem not only solves the problem of system side effect, but also achieves the controlled drug release via PCM, providing a promising theranostic nanoplatform for development of cancer combination immunotherapy.
Materials of engineering and construction. Mechanics of materials, Biology (General)
Predictions on stock market prices are a noble task owing to huge complex, dynamic, and chaotic surroundings. Fast ups and downs arise in the stock market due to influences from foreign merchandise, such as sensitive political, stockholder, economic, and emotional behaviour. In the stock market, incessant unsettlement is the main reason why financiers give away at the wrong time and frequently fail to get a profit. While financing in the stock market, the stakeholders should not disremember the gamble of payment rule and reveal their assets to greater dangers. Discovering economic time series data and exhibiting the relationship between the stock trend and past data is the main method to resolve the issue. Machine learning (ML), a conventional technique, has also been considered for its ability to predict financial markets. This manuscript proposes a new Predicting Stock Price Movements with Combined Deep Learning Models and Two-Tier Metaheuristic Optimization (PSPMCDL-TTMO) method. The PSPMCDL-TTMO methodology employs an optimal deep learning model to forecast stock price movements, determining whether prices will rise or fall. At the primary stage, the PSPMCDL-TTMO model utilizes data pre-processing using Z-score normalization to ensure that the input features are standardized for consistent performance. For feature selection (FS), the dingo optimizer algorithm (DOA) is employed to optimize the most relevant and impactful features from historical stock data. In addition, the multi-head attention bi-directional gated recurrent unit (MHA-BiGRU) model is used for stock price movement prediction. Finally, the hyperparameter range of the MHA-BiGRU model is implemented by the design of the equilibrium optimizer (EO) model. The experimentation outcome analysis of the PSPMCDL-TTMO approach takes place, and the results are inspected using various features. The investigational validation of the PSPMCDL-TTMO technique attained a superior CORR value of 0.9999 over existing models.
Medical physics. Medical radiology. Nuclear medicine, Nuclear engineering. Atomic power
Deep learning based methods often suffer from performance degradation caused by domain shift. In recent years, many sophisticated network structures have been designed to tackle this problem. However, the advent of large model trained on massive data, with its exceptional segmentation capability, introduces a new perspective for solving medical segmentation problems. In this paper, we propose a novel Domain-Adaptive Prompt framework for fine-tuning the Segment Anything Model (termed as DAPSAM) to address single-source domain generalization (SDG) in segmenting medical images. DAPSAM not only utilizes a more generalization-friendly adapter to fine-tune the large model, but also introduces a self-learning prototype-based prompt generator to enhance model's generalization ability. Specifically, we first merge the important low-level features into intermediate features before feeding to each adapter, followed by an attention filter to remove redundant information. This yields more robust image embeddings. Then, we propose using a learnable memory bank to construct domain-adaptive prototypes for prompt generation, helping to achieve generalizable medical image segmentation. Extensive experimental results demonstrate that our DAPSAM achieves state-of-the-art performance on two SDG medical image segmentation tasks with different modalities. The code is available at https://github.com/wkklavis/DAPSAM.
Segment Anything Model (SAM) has gained significant attention because of its ability to segment various objects in images given a prompt. The recently developed SAM 2 has extended this ability to video inputs. This opens an opportunity to apply SAM to 3D images, one of the fundamental tasks in the medical imaging field. In this paper, we extensively evaluate SAM 2's ability to segment both 2D and 3D medical images by first collecting 21 medical imaging datasets, including surgical videos, common 3D modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) as well as 2D modalities such as X-ray and ultrasound. Two evaluation settings of SAM 2 are considered: (1) multi-frame 3D segmentation, where prompts are provided to one or multiple slice(s) selected from the volume, and (2) single-frame 2D segmentation, where prompts are provided to each slice. The former only applies to videos and 3D modalities, while the latter applies to all datasets. Our results show that SAM 2 exhibits similar performance as SAM under single-frame 2D segmentation, and has variable performance under multi-frame 3D segmentation depending on the choices of slices to annotate, the direction of the propagation, the predictions utilized during the propagation, etc. We believe our work enhances the understanding of SAM 2's behavior in the medical field and provides directions for future work in adapting SAM 2 to this domain. Our code is available at: https://github.com/mazurowski-lab/segment-anything2-medical-evaluation.
This paper introduces Med-Bot, an AI-powered chatbot designed to provide users with accurate and reliable medical information. Utilizing advanced libraries and frameworks such as PyTorch, Chromadb, Langchain and Autogptq, Med-Bot is built to handle the complexities of natural language understanding in a healthcare context. The integration of llamaassisted data processing and AutoGPT-Q provides enhanced performance in processing and responding to queries based on PDFs of medical literature, ensuring that users receive precise and trustworthy information. This research details the methodologies employed in developing Med-Bot and evaluates its effectiveness in disseminating healthcare information.
Data scarcity is a major limiting factor for applying modern machine learning techniques to clinical tasks. Although sufficient data exists for some well-studied medical tasks, there remains a long tail of clinically relevant tasks with poor data availability. Recently, numerous foundation models have demonstrated high suitability for few-shot learning (FSL) and zero-shot learning (ZSL), potentially making them more accessible to practitioners. However, it remains unclear which foundation model performs best on FSL medical image analysis tasks and what the optimal methods are for learning from limited data. We conducted a comprehensive benchmark study of ZSL and FSL using 16 pretrained foundation models on 19 diverse medical imaging datasets. Our results indicate that BiomedCLIP, a model pretrained exclusively on medical data, performs best on average for very small training set sizes, while very large CLIP models pretrained on LAION-2B perform best with slightly more training samples. However, simply fine-tuning a ResNet-18 pretrained on ImageNet performs similarly with more than five training examples per class. Our findings also highlight the need for further research on foundation models specifically tailored for medical applications and the collection of more datasets to train these models.
Gurucharan Marthi Krishna Kumar, Aman Chadha, Janine Mendola
et al.
Large Language Models (LLMs), known for their versatility in textual data, are increasingly being explored for their potential to enhance medical image segmentation, a crucial task for accurate diagnostic imaging. This study explores enhancing Vision Transformers (ViTs) for medical image segmentation by integrating pre-trained LLM transformer blocks. Our approach, which incorporates a frozen LLM transformer block into the encoder of a ViT-based model, leads to substantial improvements in segmentation performance across various medical imaging modalities. We propose a Hybrid Attention Mechanism that combines global and local feature learning with a Multi-Scale Fusion Block for aggregating features across different scales. The enhanced model shows significant performance gains, including an average Dice score increase from 0.74 to 0.79 and improvements in accuracy, precision, and the Jaccard Index. These results demonstrate the effectiveness of LLM-based transformers in refining medical image segmentation, highlighting their potential to significantly boost model accuracy and robustness. The source code and our implementation are available at: https://github.com/AS-Lab/Marthi-et-al-2025-MedVisionLlama-Pre-Trained-LLM-Layers-to-Enhance-Medical-Image-Segmentation