Search Results (3,173)

Background/Objectives: Magnetic nanoparticles have emerged as powerful tools for biomedical imaging, targeted drug delivery, and hyperthermia therapy. Magnetic particle imaging (MPI) is among the most promising technologies built around its properties: a radiation-free, quantitative tomographic modality that detects superparamagnetic iron oxide nanoparticles (SPIONs) directly against a biologically silent background. This review synthesizes MPI’s physical principles, nanoparticle design strategies, and preclinical applications within the broader landscape of magnetic material engineering for biomedical use. Methods: A systematic review was conducted covering MPI signal generation and image reconstruction, nanoparticle core synthesis and surface coating approaches, and preclinical applications, spanning cell tracking, oncological imaging, vascular perfusion, neuroimaging, and MPI-guided theranostics. Studies were selected to provide quantitative benchmarks and direct comparisons with competing modalities where available. Results: MPI delivers signal-to-background ratios above 1000:1, iron-mass linearity at R² ≥ 0.99, regardless of tissue depth, and acquisition rates up to 46 volumes per second. Tracer architecture—encompassing single-core particles, multicore nanoflowers, and stimuli-responsive cluster designs—is the primary determinant of sensitivity, environmental robustness, and theranostic capability. Preclinical results include detection of cell populations in the low thousands, earlier ischaemia identification than diffusion-weighted MRI, real-time drug release quantification, and spatially confined tumour hyperthermia. Three translational bottlenecks are identified: the absence of a clinically approved tracer with optimal relaxation dynamics, hardware performance losses when scaling to human-bore systems, and overestimation of passive tumour accumulation in murine models. Conclusions: MPI illustrates how progress in magnetic material design directly expands clinical imaging and theranostic possibilities. Successful translation will require indication-driven, interdisciplinary development that integrates materials science, scanner engineering, and regulatory strategy in parallel. Full article

Three-dimensional seed phenotyping requires imaging systems capable of achieving micron-level resolution across a centimeter-level field of view (FOV), a goal constrained by the resolution–FOV trade-off in conventional light field architectures. This paper presents a hardware–software co-optimized framework that integrates a reconfigurable optical system with computational imaging pipelines to address this limitation. At the hardware level, we develop a tunable-focus lens module that enables flexible adjustment of the effective focal length, combined with a custom-designed microlens array (MLA). A mathematical model is established to analyze the interdependencies among FOV, lateral resolution, depth of field (DOF), and system configuration, guiding the design of individual optical components. On the computational side, we propose a hybrid aberration correction strategy: first, a co-calibration of lens and MLA aberrations based on line-feature detection; second, a conditional generative adversarial network (cGAN) with attention-guided residual learning to enhance sub-aperture images, achieving a PSNR of 34.63 dB and an SSIM of 0.9570 on seed datasets. Experimentally, the system achieves a resolution of 6.2 lp/mm at MTF50 over a 2–3 cm FOV, representing a 307% improvement over the initial configuration (1.52 lp/mm). The reconstruction pipeline combines epipolar plane image (EPI) analysis with multi-view consistency constraints to generate dense 3D point clouds at a density of approximately 1.5 × 10⁴ points/cm² while preserving spectral and textural features. Validation on bitter melon and rice seeds demonstrates accurate 3D reconstruction and accurate extraction of morphological parameters across a large area. By integrating optical and computational design, this work establishes a reconfigurable imaging framework that overcomes the resolution–FOV limitations of conventional light field systems. The proposed architecture is also applicable to robotic vision and biomedical imaging. Full article

Ultrahigh lateral resolution (UHLR) optical coherence tomography (OCT) technology, also called optical coherence microscopy (OCM), has gained popularity, especially in the field of biomedical imaging. In these systems, high numerical aperture (NA) Microscope objectives (MO) are employed in OCM systems to offer better than 3 µm lateral resolution. However, in the implemented broadband OCM configuration, the use of complex multi-element microscope objectives can reduce the detected returned signal compared with a simpler imaging lens configuration. This reduction in detected returned signals can become an important practical limitation in many OCM applications, particularly for biomedical imaging when high imaging speed is crucial. This study investigates whether a single off-the-shelf lens can provide a practical alternative to conventional MOs, achieving higher throughput while maintaining reasonable spatial resolution. We systematically evaluated 14 commercial lenses using Zemax OpticStudio simulations, identifying an aspherized achromatic lens (Edmund Optics #85302) that best met these key criteria. To validate its feasibility for OCM, performance was tested in both Full-Field Time-Domain OCM (FF-TD-OCM) and Line-Field Spectral-Domain OCM (LF-SD-OCM) configurations. Using a broadband composite Superluminescent Diode (SLD) source (750–920 nm), we quantified the resolvable features, axial resolution, and overall light transmission. The validated system demonstrated near-diffraction-limited performance. In the LF-SD-OCM setup, it successfully resolved features as fine as Group 8, Element 6, corresponding to a 2.2 µm line pair pitch (~1.1 µm line width) and achieved a 2.86 µm axial resolution in air. A through-focus comparison further showed practically useful contrast retention around focus. Additional imaging of onion epidermal tissue and ex vivo porcine corneal tissue demonstrated that the proposed lens could provide interpretable structural images on representative biological samples. Under the tested LF-SD-OCM detection configuration, the selected lens delivered approximately 2.0 dB higher returned signal than the Mitutoyo MY10X-823 objective according to 1.59× larger received signal. Full article

Currently, silicone is a common material used in medical research and biomedical applications. This research aims to characterize extra-soft platinum silicone (shore A 00 50) and compare its mechanical behavior with that of the human thoracic aorta. By developing molds to get samples, for tensile testing according to ISO 37 and ASTM D412, and for compression testing according to ISO 7743 and ASTM D575, using a universal testing machine for tensile and compression tests, and applying digital image correlation (DIC) algorithms, the mechanical properties were characterized in a total of 10 tensile samples and 6 compression samples. The results show an ultimate tensile strength up to 1.77 ± 0.12 MPa in the ASTM samples and 2.10 ± 0.14 MPa in the ISO samples; alongside an incremental elastic module of 80.08 ± 7.94 kPa and 117.98 ± 11.39 kPa; finally, an elongation at break of 1114.49 ± 76.77% and 936.08 ± 63.56%, corresponding to the values of a healthy thoracic aorta. The replica of the thoracic aorta in this material was developed by a brush method, with a thickness of 1.82 mm, a length from the aortic arch to the descending aorta of 200.49 mm, and diameters of 20.45 and 16.05 mm for the ascending and descending aorta, respectively. Full article

The electrocardiogram (ECG) is a central tool in cardiovascular diagnostics, yet interpretation requires expertise and remains subject to variability. Multimodal large language models (MLLMs) have shown emerging capabilities in medical image analysis, but their performance in ECG interpretation remains insufficiently characterized. This study evaluated the diagnostic accuracy and inter-run reliability of five MLLMs across ECG interpretation tasks. Thirteen standard 12-lead ECGs were presented to five models (ChatGPT-5.3, Gemini 3.1 Pro, Claude Opus 4.6, Grok 4.1, and ERNIE 5.0) across five independent runs per case, yielding 2275 task-level assessments. Six categorical interpretation tasks (rhythm, electrical axis, PR/P-wave morphology, QRS duration, ST/T-wave morphology, and QTc interval) were compared with expert-consensus ground truth, while heart rate estimation was evaluated using mean absolute error (MAE). Overall categorical accuracy ranged from 52.3% to 64.9%. QRS duration classification achieved the highest accuracy (66.2–90.8%), whereas ST/T-wave assessment showed the lowest performance (20.0–41.5%). Heart rate MAE ranged from 14.8 to 46.7 bpm. A dissociation between diagnostic accuracy and inter-run reliability was observed across models. These findings indicate that current MLLMs do not achieve clinically reliable ECG interpretation performance and highlight the importance of assessing diagnostic accuracy and inter-run reliability when evaluating artificial intelligence systems in biomedical diagnostics. Full article

Multimodal biomedical machine learning increasingly integrates heterogeneous data sources (including medical imaging, multi-omics profiles, electronic health records, and wearable sensor signals) to support clinical diagnosis, prognosis, and treatment response prediction. Despite strong empirical performance, most existing multimodal systems lack a principled theoretical foundation for understanding why fusion improves prediction, how information is distributed across modalities, and when models can be trusted under incomplete or shifting data. This paper develops a unified information-theoretic framework that formalizes multimodal biomedical learning as an information optimization problem. We formulate multimodal representation learning through the information bottleneck principle, deriving a variational objective that balances predictive sufficiency against informational compression in an architecture-agnostic manner. Building on this foundation, we introduce information-theoretic tools for decomposing modality contributions via conditional mutual information, quantifying redundancy and synergy, and diagnosing fusion collapse. We further show that robustness to missing modalities can be cast as an information consistency problem and extend the framework to longitudinal disease modeling through transfer entropy and sequential information bottleneck objectives. Applications to multimodal foundation models, uncertainty quantification, calibration, and out-of-distribution detection are developed. Empirical case studies across three biomedical datasets (TCGA breast cancer multi-omics, TCGA glioma clinical-plus-molecular data, and OASIS-2 longitudinal Alzheimer’s data) show that the framework’s key quantities are computable and interpretable on real data: MI decomposition identifies modality dominance and redundancy; the VMIB traces a compression–prediction tradeoff in the information plane; entropy-based selective prediction raises accuracy from 0.787 to 0.939 at 50% coverage; transfer entropy reveals stage-dependent modality influence in disease progression; and pretraining/adaptation diagnostics distinguish efficient from wasteful fine-tuning strategies. Together, these results develop entropy and mutual information as organizing principles for the design, analysis, and evaluation of multimodal biomedical AI systems. Full article

Wireless Capsule Endoscopy (WCE) is a minimally invasive technology for imaging the gastrointestinal (GI) tract, particularly the small intestine, where conventional endoscopy faces accessibility limitations. Traditional capsule endoscopes rely on passive motion driven by natural peristalsis, which limits controllability and may increase the risk of capsule retention. To address these challenges, this study presents the design and experimental validation of a compliant active capsule endoscope actuated by four Shape Memory Alloy (SMA) spring actuators. A key feature of the proposed system is a steering mechanism that reuses the same SMA actuators responsible for locomotion, enabling control of the camera orientation without increasing system complexity, size, or weight. The capsule architecture consists of rigid polylactic acid (PLA) links connected through thermoplastic polyurethane (TPU) flexure hinges, fabricated using dual-material 3D printing. Nonlinear finite element analysis (FEA) was employed to optimize the flexure hinge geometry for maximum displacement while maintaining safe stress levels. To validate the concept, a $3.5\times$ scaled prototype was fabricated and integrated with SMA actuators and an Arduino-based control system. The experimental results demonstrate effective locomotion and steering capabilities, achieving a maximum stroke of approximately 5.4 mm and a steering angle of 24° for the $3.5\times$ scaled prototype, corresponding to an estimated stroke of approximately 1.98 mm (Based on the FEA) at the intended clinical scale. Thermal characterization of the SMA actuators was also conducted to identify suitable operating current ranges for future biomedical deployment. The results demonstrate the feasibility of integrating locomotion and steering within a compact compliant capsule architecture, representing a step toward next-generation capsule endoscopy systems with improved navigation and diagnostic capability. Full article

This study introduces a physics-guided deep learning architecture designed for the simulation, reconstruction, and pattern recognition of biomedical images. By explicitly integrating physical priors into the learning model, the framework addresses the black-box nature of traditional artificial intelligence (AI). It provides an explainable AI pathway that enhances diagnostic accuracy, robustness, and clinical interpretation. The proposed framework was evaluated through systematic simulation studies. It involved complex geometric configurations, multimodal physical fields, and noise-corrupted synthetic three-dimensional brain volumes. Quantitative analysis demonstrates consistent improvements in reconstruction fidelity, with the peak signal-to-noise ratio (PSNR) reaching 47 dB and the structural similarity index exceeding 0.90 across all scenarios. Notably, at moderate noise levels (0.05), the framework maintains a PSNR greater than 32 dB, ensuring structural integrity essential for computer-aided diagnosis. Volumetric brain experiments further reveal a 38–44% reduction in activation localization errors, highlighting the framework’s utility in functional imaging and disease prognosis. By grounding deep learning in physical constraints, this study provides a transparent and robust solution for automated disease classification and advanced biomedical imaging tasks within clinical decision support systems. Full article

The current work aims to enhance the structural, optical, photocatalytic, and cytotoxic properties of CuO NPs at varied rGO concentrations of 5% and 10%. In the present work, a one-step hydrothermal method was successfully applied to prepare rGO/CuO NCs at different concentrations of RGO. The novelty of this work was to enhance the structural, optical, photocatalytic, and cytotoxic properties of CuO using the addition of rGO sheets. XRD, TEM, SEM-EDX, XPS, FTIR, UV-vis, PL, and DLS techniques were used to characterize the prepared samples. XRD data confirmed the formation of the monoclinic phase of CuO with a decrease in crystallite size, from 21.14 nm for CuO to 16.94 nm for the 10% rGO/CuO NCs nanocomposite. SEM and TEM images verified the uniform anchoring and excellent dispersion of CuO nanoparticles on the rGO sheets, and the EDX spectra showed the presence of Cu, O, and C elements in the obtained rGO/CuO NCs. DLS measurements showed that the hydrodynamic radius dropped from 69.98 ± 17.81 nm for CuO to 51.72 ± 10.48 nm for 10% rGO/CuO NCs. The zeta potential values remained negative for all samples, ranging from −20.50 ± 8.69 mV for CuO to −25.60 ± 9.08 mV for 10% rGO/CuO NCs, suggesting enhanced colloidal stability with rGO incorporation. Furthermore, FTIR and XPS analyses confirmed that Cu–O–C bonding formed between CuO and rGO. UV-Vis analysis revealed a redshift in the absorption edges as rGO content increased, reducing the band gap from 3.65 eV to 3.60 eV. Additionally, PL spectra showed a marked reduction in emission intensity due to a decrease in the recombination rate between electron (e⁻)–holes (h⁺) pairs. The CuO/(10%)rGO NCs showed the best photocatalytic performance with a 93.56% degradation of methylene blue (MB) after 120 min under UV irradiation, and followed pseudo-first-order kinetics with k = 0.0203 min⁻¹. Cytotoxicity studies on HT1080 cells showed a dose-dependent decrease in viability. 10% rGO/CuO NCs exhibited the highest cytotoxicity effect, resulting in 58% and 50% viability at 1.4 mg/mL, respectively. The presented results showed that the presence of rGO in CuO NPs played a role in enhancing the structural stability, charge mobility, and biological reactivity of Cu NPs. This study highlighted that the rGO/CuO NCs are a promising multi-functional material for environmental and biomedical applications. Full article

As critical regulators of gene expression, microRNAs (miRNAs) are key biomarkers and therapeutic targets in cancer. However, current methods for intracellular miRNA imaging are often limited by poor sensitivity and operational complexity. In this study, we identified a site-specifically modified DNAzyme variant, 11Bn, which exhibits up to 7-fold higher catalytic activity than the wild-type 8-17 through systematic screening. Using this variant, we constructed a DNAzyme-based sensor for miRNA-21 imaging in living cells. The sensor achieves a limit of detection of 7.89 nM, threefold lower than that of the wild-type sensor, and enables sensitive visualization of intracellular miRNA-21 without signal amplification. Moreover, it can capture dynamic changes in miRNA levels within cells, providing a versatile molecular tool for miRNA imaging and related biomedical applications. Full article

Understanding Alzheimer’s Disease (AD) progression using resting-state functional Magnetic Resonance Imaging (rs-fMRI) remains an open challenge. Variational Autoencoders (VAEs) provide compact representations of high-dimensional neuroimaging data but lack mechanisms to highlight disease-relevant features. We propose KS-VAE, a novel framework that integrates the Kolmogorov–Smirnov test into the latent space of VAEs to identify statistically significant variables discriminating healthy from pathological brain states. This integration enables measurement of latent space shifts associated with cognitive decline, offering a quantitative approach to neurodegenerative processes. By modifying the most relevant variables, KS-VAE generates synthetic samples that simulate transitions between clinical conditions while preserving anatomical plausibility. The method enhances the modeling of temporal and distributional dynamics underlying disease progression and provides interpretable analysis of class-relevant features. Applied to rs-fMRI scans of 220 subjects from the ADNI cohort, KS-VAE demonstrated robust class separation between cognitively normal and Alzheimer’s disease subjects, achieving a classification accuracy of 84.5% and an F1-score of 84.5%, and clinically consistent synthetic transitions. KS-VAE thus offers a statistically grounded and clinically interpretable framework for understanding Alzheimer’s disease progression. Full article

This study proposes a unified multimodal temporal motion state perception framework for optical imaging-oriented biomedical applications, integrating visual skeleton sequences, inertial measurement unit (IMU) signals, and surface electromyography (EMG) signals. The framework utilizes modality-specific encoders and a cross-modal temporal alignment attention mechanism to explicitly model temporal offsets from heterogeneous sensing streams. A multimodal temporal Transformer backbone is introduced to capture long-range motion dependencies and cross-modal interactions, while an uncertainty-aware fusion module dynamically allocates weights based on modality confidence. Experimental results demonstrate that the proposed approach achieves an accuracy of 94.37%, an F1-score of 93.95%, and a mean average precision of 96.02%, outperforming mainstream baseline models. Robustness evaluations further confirm stable performance under visual occlusion and sensor noise. These results indicate that the framework provides a highly accurate and robust solution for rehabilitation assessment, sports training monitoring, and wearable intelligent interaction systems. Full article

Treatment of postsurgical osteosarcoma remains one of the major challenges in orthopedic clinics. Conventional implants often fail to address complex pathological issues, including irregular bone defects, residual tumor cells, and delayed bone regeneration. Herein, this study reports a multifunctional shape-memory polyurethane (SMPU)/manganese dioxide (MnO₂) composite that provides adaptive support, antitumor activity, and osteogenic bioactivity. SMPU was synthesized by introducing 1,4-butanediol (BDO) and dimethylolpropionic acid (DMPA) as chain extenders at a specific ratio. Commercial MnO₂ nanoparticles were incorporated as both a photothermal agent and a bioactive component to achieve multifunctionality. As designed, a coordination system was formed between the polymer chains and MnO₂ nanoparticles within the composites. The influence of MnO₂ content was systematically investigated. Although increasing MnO₂ amounts improved photothermal and mechanical performance, excessive incorporation adversely affected the molecular structure and compromised the composite’s biocompatibility. By adjusting the MnO₂ content, the composites were demonstrated to possess robust mechanical performance, good shape-memory behavior, and controllable Mn²⁺ release. Additionally, the composites exhibited tunable photothermal performance under near-infrared (NIR) irradiation. Furthermore, in vitro studies confirmed that the composites containing 4 wt% MnO₂ could eliminate tumor cells via photothermal effects and promote the osteogenic differentiation of human bone marrow-derived mesenchymal stem cells (hBMSCs). Overall, the SMPU/MnO₂ composites had superior multifunction for treating irregular bone defects following bone tumor surgery. Full article

Multifrequency operation in micromachined ultrasonic transducers, enabled by targeted excitation of specific vibrational modes, has emerged as an attractive approach for achieving tunable performance and configurability, well-suited for advanced ultrasound imaging and therapeutic applications. This paper presents a dual-electrode rectangular piezoelectric micromachined ultrasonic transducer (PMUT) designed for efficient dual-frequency operation through mode-selective actuation. The proposed architecture employs segmented electrodes that are spatially aligned with the strain distributions of two distinct flexural modes, enabling selective excitation of Mode 1 (fundamental) and Mode 3 (higher order) through appropriate electrode actuation. Finite element simulations and impedance analysis were used to guide the electrode configuration and validate the mode-selective behavior. The dual-mode PMUT was fabricated alongside a conventional single-electrode PMUT using identical membrane dimensions and material stack for direct comparison. Comprehensive electrical and underwater acoustic characterization confirmed that the conventional PMUT is limited to single-frequency operation at the fundamental resonance. In contrast, the proposed design achieved a substantial improvement in higher-order performance, with a threefold increase in acoustic pressure at Mode 3 compared to the conventional device. These results demonstrate that mode-aligned electrode segmentation enables efficient dual-mode operation without added fabrication complexity, making the design highly suitable for multifrequency ultrasonic applications such as biomedical imaging and sensing. Full article

Many patients with primary or metastatic lung cancer are not candidates for surgery, additional radiation, or further systemic therapy due to advanced age or comorbidities; this creates a need for minimally invasive locoregional options. Image-guided thermal ablation (IGTA) is being applied across a broader spectrum of lesions, while bronchial artery chemoembolization (BACE) is emerging as a therapy option for treatment-refractory advanced disease. Recent studies in thermal ablation have focused on optimizing energy delivery and protocols, as well as improving ablation zone predictability and analysis. Advances in lesion targeting, including cone beam CT fusion, electromagnetic guidance, and robotic-assisted ablation, allow for treatment of subcentimeter and ground-glass lesions in anatomically challenging locations. Growing clinical experience supports IGTA for intrathoracic oligoprogression and as salvage therapy after recurrence. In the endovascular space, improved imaging, microcatheters, and drug-eluting microspheres have expanded the use of BACE for disease and symptom control in advanced lung cancer. Multimodal strategies combining minimally invasive locoregional treatments with systemic therapies and radiation are being explored, with early data showing improvements in survival without increased toxicity. This narrative review synthesizes emerging techniques, clinical data, and indications for percutaneous and endovascular lung cancer treatments and underscores the need for prospective and randomized trials to refine patient selection, treatment sequencing, and long-term outcomes. Full article