LFSamba: High-efficiency light field integral imaging 3D salient object detection method based on Mamba

Intracardiac Echocardiography from the Left Heart Chambers to Guide Electrophysiology Procedures.

This review explores the potential of quantitative imaging biomarkers (QIBs) for guiding personalized breast cancer treatment through prognostic stratification and measurement of treatment response. Accumulating data from prospective clinical trials show that QIBs are more predictive of treatment outcome than conventional sizebased measures (e.g., change in tumor diameter), highlighting these markers' potential clinical impact. Although QIBs have shown utility in the trial setting, clinical adoption has remained limited. In this article, we present biomarkers derived from dynamic contrast-enhanced MRI, DWI, [18F]FDG PET, and [18F]FES PET, and highlight implementation challenges including variable acquisition protocols, lack of standardization, and uncertainty around optimal timing of imaging during treatment. We distinguish between integrated biomarkers-those used for correlative or exploratory purposes within a trial-and integral biomarkers-those that are essential to trial design and directly inform patient stratification or therapeutic decisions. Standardization efforts by national and international organizations are underway to ensure imaging marker reliability, and commercial tools supporting clinical translation have become available. Dedicated end-to-end solutions remain needed to integrate QIBs into clinical workflows and ultimately into standard care pathways, to promote these markers' role in patient stratification and adaptive treatment decisions.

Advances in systemic therapy have improved outcomes for metastatic non-small cell lung cancer (NSCLC), yet resistance and progression remain nearly universal. Local therapies such as radiotherapy, surgery, and image-guided ablation can extend disease control in selected patients, but existing classifications-including dynamic models of oligometastatic disease-assign a single state per patient and do not capture lesion-level heterogeneity. We introduce the concept of metastatic trajectories-the spatiotemporal dynamics of response and progression across lesions, organs, and patients-as a framework to characterize intrapatient heterogeneity and inform adaptive treatment strategies. Dimensions of metastatic trajectories include the magnitude and homogeneity of response, mechanisms of resistance, organotropism, and the pattern, site, extent, and pace of progression. This framework shifts the focus from overall disease states to individual lesion behavior over time, enabling reactive strategies based on observed trajectories and anticipatory strategies based on predicted ones. We review the biological foundations of intrapatient heterogeneity-including genomic diversification, nongenetic plasticity, and tumor-microenvironmental adaptation-that drive divergent lesion evolution and treatment response. Emerging biomarkers such as circulating tumor DNA and radiomic signatures, together with integrative genomic and functional imaging approaches, may allow tracking and prediction of trajectory evolution. Standardized reporting parameters are proposed to ensure consistent documentation and facilitate validation across studies. Integrating trajectory-based assessment into clinical practice could refine patient selection for local and systemic therapy, enable biology-informed adaptation of treatment timing and intensity, and provide a foundation for next-generation clinical trials aimed at precision management of metastatic NSCLC.

Design, analysis, and performance testing of a long-wave infrared cryogenic integrated optical imaging system with a long focal length

Mineral scaling poses severe operational and economic challenges in water treatment and industrial processes. Prevailing models primarily focus on how surface chemistry triggers heterogeneous nucleation. However, a quantitative understanding of how substrate properties govern the more prevalent homogeneous nucleation-bulk deposition pathway remains elusive. Here, we introduce a label-free integrated imaging platform combining bright-field microscopy and low-angle rotational interferometric scattering microscopy (LRISM) to track crystal populations and quantify the three-dimensional orientation during interfacial crystallization. Using calcium sulfate (CaSO4) and calcium carbonate (CaCO3) as model scales, we decouple the roles of supersaturation and substrate chemistry. CaSO4 consistently undergoes heterogeneous nucleation, with morphology transitioning from rod-like morphology at high supersaturation to cruciform morphology at low supersaturation. In contrast, CaCO3 invariably follows a homogeneous nucleation-bulk solution deposition pathway. Quantitative analysis of this homogeneous nucleation pathway reveals that hydrophilic surfaces capture a greater density of nascent nuclei, limiting their final size through spatial confinement and growth competition. LRISM-based three-dimensional orientation analysis further confirms that hydrophilic surfaces preferentially induce horizontal crystal alignment. These results establish quantitative observables and a general methodology for designing scale-mitigation surfaces by controlling postnucleation capture and orientation rather than nucleation inhibition alone.

AI-based content generation has become increasingly important across various industries, en-abling users to create textual and visual content efficiently using artificial intelligence. This project presents the design and development of a unified AI-powered content generation platform implemented without relying on external APIs. The proposed system utilizes LLaMA-based large language models for text generation and diffusion models for image generation, deployed locally to ensure greater control, data privacy, and reduced dependency on third-party services. The system allows users to generate articles, blogs, sum-maries, and AI-generated images through a single web-based interface. User inputs are processed by the in-tegrated language and image generation models, and the generated outputs are delivered in real time. The system architecture is designed to efficiently manage model inference, content generation workflows, and user interactions. This project demonstrates the feasibility of building a standalone AI content generation system using open-source models and highlights its applicability in domains such as education, digital content creation, and media automation.

Integrative imaging and transcriptomics implicate neutrophils in microvascular no-reflow after stroke

Label-free digital protein sensing by integral plasmonic imaging

High resolution 3D model-based topographic assessment of scour around a hollow artificial reef by integration monocular vision and image processing techniques

Physical artificial intelligence has emerged as a pivotal component in next-generation humanoid technologies, including advanced optical sensors, as it enables autonomous acquisition of sensory information. This study reports a high-performance 0D/2D hybrid photodetector using a high-gain Ag2Te/MoS2 hybrid structure for visible to short-wave infrared (SWIR) photodetection, achieved by the absorption of Ag2Te quantum dots in the infrared region (∼1450nm). The Ag2Te/MoS2 photodetector exhibits a high photoresponsivity of around 7.5×105AW-1 and a specific detectivity of over 9.9×108 Jones at 1 µW/cm2 illumination power with a 0.2V drain bias voltage. Furthermore, depending on the gain of the photodetector, a fast response speed can also be achieved, with rise and decay times as short as 13 and 23ms. The 0D/2D hybrid devices were successfully implemented in a 32×32 array format for infrared imaging, with the results demonstrating spatially resolved pattern reconstruction and real-time photoresponse acquisition. By hybridizing quantum dots and 2D materials, the developed photodetector has broad potential applications, including use in highly integrated SWIR image sensors.

This prospective cohort study evaluated the utility of [¹⁸F]AlF-NOTA-FAPI-04 PET/CT for whole-body assessment of fibroblast activation in systemic sclerosis (SSc), correlating organ-specific uptake with clinical indicators of severity and multi-organ dysfunction. 31 patients with SSc and 21 patients from a non-SSc cohort underwent total-body [¹⁸F]AlF-NOTA-FAPI-04 PET/CT imaging. Quantitative uptake parameters were measured in lungs, heart, kidneys, skeletal muscles, esophagus, and skin. Clinical assessments were completed within ± 14 days. Inter-organ metabolic correlations were analyzed. Abnormally elevated FAPI uptake was prevalent in the lungs (96.8%), heart (35.5%), kidneys, and skeletal muscles (12.9%). Compared with the non-SSc cohort, patients with SSc exhibited significantly elevated whole lung SUVmean(wlSUVmean:overall adjusted P-value(padj) < 0.001) and cardiac SUVmax (cSUVmax: median1.37, IQR 1.17-1.87 vs.1.01 ± 0.15, padj <0.001). Reduced pulmonary function showed significant negative correlations with wlSUVmean (padj <0.01). SSc patients with renal impairment demonstrated markedly higher renal SUVmax(rSUVmax)and renal SUVmean (rSUVmean) compared to both SSc patients with preserved renal function and the non-SSc cohort(both overall padj < 0.001). Muscular uptake aligned with elevated creatine kinase and biopsy-proven myopathy. No significant tracer accumulation was observed in the esophagus or skin. Inter-organ correlation analysis revealed metabolic coupling across pulmonary, cardiac, and renal systems (wlSUVmean vs. rSUVmean: r = 0.722; cSUVmean vs. rSUVmean: r = 0.632). [¹⁸F]AlF-NOTA-FAPI-04 PET/CT provides a noninvasive, integrative imaging platform for mapping fibroblast activation across multiple organs in SSc. It demonstrated potential for detection of fibrotic burden and inter-organ metabolic profiling. Its sensitivity in gastrointestinal and cutaneous involvement remains limited, warranting further optimization.

Indocyanine green (ICG)-based near-infrared (NIR) fluorescence imaging is widely used for vascular morphology visualization but lacks functional information, such as blood flow velocity and perfusion. Here, we present a prototype NIR ICG imaging system combined with a quantitative framework to simultaneously measure vascular structure and function. The system, incorporating a 785nm laser, high-sensitivity charge-coupled device camera, and motorized stage, was validated using tissue-mimicking phantoms with vessel diameters of 1 and 2mm under physiological flow rates. Blood-mimicking fluids circulated via a peristaltic pump simulated realistic hemodynamics. Image processing employed adaptive histogram equalization, bilateral filtering, and Otsu's segmentation within a semi-automated workflow to extract vessel diameter, flow velocity, and perfusion rate with reduced user intervention. The relative errors were below 7% for diameter and flow velocity measurements. By integrating morphological and functional quantification into a unified framework, the proposed approach improves reproducibility over conventional manual ICG analysis. Laser irradiance remained within safe exposure limits. This quantitative and integrated imaging strategy demonstrates promising translational potential for intraoperative vascular assessment.

This study investigates how Social-Media Affordance Quality (SMAQ) and Omnichannel Integration Quality (OCIQ) shape Bali’s MICE destination image (DI-MICE) and perceived MICE destination competitiveness (PMDC), and whether these effects differ between first-time and repeat international visitors. Data were collected through purposive intercept surveys at MICE-related hotels, venues, and attractions in Bali, targeting international visitors who had interacted with Bali’s official MICE digital channels. A quantitative cross-sectional design was applied to 300 valid responses and analyzed using PLS-SEM (SmartPLS 4). The results show that SMAQ and OCIQ significantly enhance destination image and perceived competitiveness, with destination image acting as a key mediator. Multi-group analysis reveals that first-time visitors rely more strongly on social-media affordances, whereas repeat visitors respond more to seamless omnichannel integration. Once an image is formed, both groups evaluate competitiveness similarly. These findings position digital affordances and omni-channel integration as strategic signals of professional capability, offering practical guidance for MICE destination managers in Bali and comparable emerging destinations.

Artificial intelligence (AI) and radiomics are increasingly applied in pediatric neuroradiology to enhance diagnostic precision. However, their clinical implementation remains limited due to methodological variability and lack of standardization. To systematically evaluate the diagnostic applications, performance, and methodological quality of artificial intelligence models, including both radiomics/machine learning and deep learning approaches, in pediatric brain tumor imaging. A systematic review was conducted in accordance with PRISMA 2020 guidelines, searching PubMed, Scopus, and Web of Science (2010-2025). Eligible studies included patients aged 0-18 years with brain tumors, applied AI to neuroimaging for diagnostic classification, molecular characterization, or integral image processing tasks, and reported performance metrics. Methodological quality was assessed using the Newcastle-Ottawa Scale (NOS) and the Checklist for Artificial Intelligence in Medical Imaging (CLAIM). From 638 records, 24 studies were included. Most studies used MRI (96%) and machine learning models based on radiomics (88%), with a smaller proportion employing deep learning (29%). The primary diagnostic tasks were tumor classification (50%) and molecular subtype prediction (33%). Reported AUCs for diagnostic tasks ranged from 0.73 to 0.98 (median: 0.91). Based on the NOS, 19 studies (79%) were rated as low risk of bias (scores 8-9), though only 3 studies (12.5%) performed external validation on independent cohorts. AI and radiomics demonstrate high diagnostic accuracy for pediatric brain tumor characterization. Nonetheless, the lack of prospective design and critically low rate of external validation limits generalizability. Standardized, multicenter studies are needed to support broader clinical adoption.

Chromatin condensation, accompanied by pronounced nuclear shrinkage, is a pivotal step in terminal erythroid differentiation and is essential for enucleation. However, the precise timing and molecular mechanisms initiating this process remain poorly understood, particularly as global transcriptional repression occurs only at the orthochromatic erythroblast stage. Here we perform a comprehensive analysis of three-dimensional chromatin architecture dynamics during erythropoiesis using integrative epigenomic and imaging approaches. We demonstrate that chromatin condensation begins earlier than previously recognized, initiating at the late basophilic erythroblast stage. Mechanistically, these intergenic regions were tethered to the nuclear lamina (predominantly heterochromatic) and enriched in H3K9me3, which drive large-scale chromatin compaction. Furthermore, we find that the redistribution of H3K9me3 and dynamic remodeling of Lamin B1 are critical for this structural transition, directly impacting erythroid maturation. These findings reveal a previously unrecognized regulatory axis linking H3K9me3 and chromatin-lamina interactions, providing novel insights into the spatial and temporal control of chromatin organization during erythroid development.

Abstract Sparse scanning protocols in computed tomography enable rapid imaging of stacked battery cells. However, existing sparse scanning methods often overlook regions rich in internal structural information within the cells, resulting in reconstructed images with missing structural details. Furthermore, incomplete scanning introduces noise and artifacts, further degrading image quality. To address these issues, we propose an adaptive method of gradient-norm-based sparse scanning and feature-fusion-based reconstruction for stacked battery cells. First, a reference image is reconstructed from full-angle scan data to compute the gradient norm distribution corresponding to each scanning angle, which quantifies the contribution variation of different angles to structural feature information. A gradient norm distribution curve is subsequently plotted. Then, the cumulative gradient norm values are computed across all possible limited-angle ranges to identify the optimal range with the highest cumulative value. Within this range, a non-uniform angle sampling method is adopted: scanning angles with higher gradient norms are prioritized, while additional angles are adaptively supplemented using an angle spacing cost function. During the reconstruction process, an adaptive feature-fusion non-local means (AFF-NLM) regularization term is incorporated into the projections onto convex sets—total variation (TV) framework, alternating with TV minimization. The proposed algorithm dynamically adjusts the search range and filtering parameters of AFF-NLM based on local structural characteristics of the image. Moreover, integral image technology is introduced to accelerate the patch-matching process, enabling efficient evaluation of patch similarity through precomputation and weighted processing. Experimental results demonstrate that the proposed method more effectively utilizes projection data, excels in suppressing artifacts and noise, and significantly improves the quality of the reconstructed images.

Ghost imaging, which reconstructs object images through correlation calculations of light fields, offers several advantages such as overcoming diffraction limits and resisting turbulence or scattering interference. In this article, we demonstrate that random lasers are excellent light sources for high-quality ghost imaging due to their random light field intensity distribution. The peak of the random laser emission spectrum fluctuates randomly over time, and the resulting speckle distribution also varies randomly. This characteristic makes it particularly suitable for generating the dynamic speckle field required for ghost imaging. Results show that the ghost imaging exhibits the optimal signal-to-noise ratio when the pump energy slightly exceeds the random laser threshold, and the signal-to-noise ratio exceeds 9.5. The resolution of ghost imaging depends mainly on the average size of the speckles, and the resolution of random laser ghost imaging can reach 88.3 μm in our experimental conditions. This research not only demonstrates the application potential of random lasers in the field of ghost imaging but also offers insights for developing more integrated and stable ghost imaging systems.

Robinson Crusoe is a classic work in British and American literature. The protagonist, Robinson Crusoe, survived on the deserted island with his indomitable will and became the owner, ruler, and lawmaker of the island, presenting an ideal image of diligence and integrity. However, when he returned to the civilized society, his image was completely different from that of the “sincere laborer” on the island, revealing his greedy pursuit of wealth and status. This article analyzes the manifestations and causes of this hypocrisy at the interpersonal interaction and value level from multiple dimensions, combined with the historical background of the rise of British capitalism and colonial expansion in the 18th century, to explore the social roots of this phenomenon. Through this mirror of literature to examine reality, this article also provides a new perspective for understanding similar phenomena in contemporary society.

We evaluate a three-dimensional (3D) object detection system for operation in turbid water based on 3D profilometric Integral Imaging (InIm) and a deep neural network. While conventional InIm computational reconstruction provides the two-dimensional (2D) slices of the 3D scene at specific 2D depth planes, 3D profilometry allows visualization of the 3D surface from unique perspectives. In the proposed method, we develop a deep neural network-based red, green, blue-depth (RGB-D) object detection framework using passive 3D profilometry under turbid conditions. An image sensor on a moving platform captures multiple 2D perspective images of the 3D scene, from which a depth map is statistically estimated. The captured perspective image and the estimated depth map are then fused to generate a four-channel RGB-D image for 3D object detection in turbidity. Comparative experiments are conducted and demonstrate that the proposed 3D profilometry-based approach outperforms both 2D imaging and conventional 3D InIm-based reconstruction across various turbidity levels evaluated. To the best of our knowledge, this is the first report on InIm 3D profilometry for object detection in turbid water.