Search Results (655)

Focal therapy is emerging as an organ-preserving strategy for selected patients with localized prostate cancer, aiming to preserve urinary and sexual function while maintaining acceptable cancer control. However, patient and technology selection remain complex because prostate cancer is often multifocal, clinically significant lesions may be missed by imaging or biopsy, and long-term comparative oncological data are limited. This narrative review summarizes current evidence and consensus recommendations on oncological suitability, histopathological risk features, tumor burden, imaging assessment, anatomical considerations, functional priorities, and follow-up. We also discuss the complementary roles of multiparametric magnetic resonance imaging, prostate-specific membrane antigen positron emission tomography, micro-ultrasound, and artificial intelligence-assisted planning. Finally, we review how tumor location and proximity to critical structures guide selection among high-intensity focused ultrasound, cryotherapy, irreversible electroporation, transurethral ultrasound ablation, laser ablation, and photodynamic therapy. Focal therapy remains promising but requires careful selection, shared decision-making, and structured follow-up. Full article

Building energy consumption accounts for 30% of global energy use, making building management pivotal to achieving global sustainability. Occupants have profound impacts on the building environment. Incorporating occupant-related factors into the environmental control process is essential for optimizing the efficiency of building management systems (BMSs), which thus gives rise to the concept of occupant-centric control (OCC). Conventional methods rely on simplified models and fixed schedules that fail to satisfy environmental control and occupant requirements, while constructing credible models places strict requirements on the dataset. In this paper, we propose a Model-Aware Predictive Control (MAPC) framework that can construct credible models with limited data and provide room-level control strategies to optimize the trade-off between occupant comfort and energy consumption. The technological innovations of this research are twofold. On the one hand, we design a model construction and fine-tuning method that combines data-driven subspace projection approach with physical priors that can construct credible thermal dynamic models with limited data. On the other hand, to balance the potential conflicts between enhancing occupant comfort and saving energy, we present a hierarchical decision-making mechanism that enables adaptive multi-objective room-level control considering dynamic occupant comfort requirements and energy usage. The experimental results obtained on an EnergyPlus-based simulation dataset and a publicly available dataset demonstrate that MAPC can provide room-level control strategies based on dynamic occupant requirements and user preferences and achieve superior trade-offs between occupant comfort and energy consumption. The ablation experiments also demonstrated the superiority of MAPC in constructing reliable models on limited datasets. MAPC provides pivotal support for the advancement of the intelligent buildings and sustainable indoor environment. Full article

Background/Objectives: Deep learning models for digital mammography sensor data are increasingly deployed across hospitals using different X-ray detector technologies and patient populations. Whether models trained on one sensor platform and population maintain accuracy when transferred to another has not been tested for the latest generation of mammography-specific foundation models under one controlled protocol. Methods: We fine-tuned five backbone architectures (ResNet-50, DINOv2-B14, Rad-DINO, Mammo-CLIP B5, and Mammo-FM) on CBIS-DDSM (film-digitized, USA, n = 714 validation) with three seeds, ablated a density-aware focal loss across three auxiliary weights, and evaluated transfer to three external sensor cohorts: CMMD (full-field digital, China, n = 1032), DMID (mixed digital, India, n = 509), and MIAS (film-digitized, UK, n = 322). Significance used paired DeLong z-tests with Benjamini–Hochberg FDR correction; temperature scaling tested post hoc recalibration at all transfer targets. Results: Within this single-source three-seed evaluation, ResNet-50 outperformed all four foundation models on CBIS-DDSM (AUC 0.867 vs. 0.847, 0.846, 0.813, and 0.703; all gaps p_adj < 0.05). The density-aware focal loss degraded both AUC and calibration at every weight tested. At transfer, every model lost 0.165 to 0.320 AUC points relative to in-distribution performance, with sensitivity at 95% specificity collapsing from 0.31 to 0.47 in-distribution to 0.11 to 0.22 across the three external targets. A per-seed Stouffer meta-analysis confirms that Mammo-CLIP B5 and Mammo-FM significantly outperformed ResNet-50 on DMID and Mammo-CLIP on CMMD, after BH-FDR; MIAS comparisons remained directional only. In the extremely dense subgroup (BI-RADS D4), Mammo-FM reached AUC 0.870 versus ResNet-50 at 0.842, a directional observation whose 95% CIs overlap heavily at the n = 140 sample size and which we do not interpret as a statistically supported advantage. Conclusions: In this single training-source, three-seed protocol, mammography-specific pretraining did not deliver the in-distribution AUC premium reported in the originating papers, and no architecture reached a level at which transfer deployment without local validation would be defensible. We frame these as observations specific to the present protocol rather than as broader conclusions about foundation models for mammography classification. The findings argue for sensor-stratified and population-stratified external validation and for local recalibration as practical prerequisites before clinical use. Code and weights are released under MIT license. Full article

The increasing convergence of Operational Technology (OT) and Information Technology (IT) within the Industrial Internet of Things (IIoT) brings about remarkable improvements in monitoring and automation. However, it also exposes industrial systems to large-scale Distributed Denial of Service (DDoS) attacks. Edge-based defences are essential in satisfying low-latency demands and data sovereignty rules, yet they must function under severe resource limitations and adapt to shifting traffic characteristics without cloud assistance. In this work, we introduce a lightweight hybrid deep learning architecture that fuses a Convolutional Neural Network (CNN) with a Convolutional Block Attention Module (CBAM) and a Multi-Layer Perceptron (MLP) in a single detector. A sequential transfer learning scheme is adopted, including a feature projection layer that handles differences in input dimensionality. The model is pre-trained on the CIC-DDoS2019 dataset, then adapted to the more recent CICIoT23 dataset. Evaluations are performed on both datasets while preserving their natural class imbalance. We provide extensive ablation and variance analysis under identical experimental conditions. The proposed method achieves 99.52% accuracy on CICIoT23 while maintaining 99.65% recall, which is a crucial property for critical systems. Real-time measurements on a CPU-only testbed show an average inference latency of 0.013 ms, inference-only throughput exceeding 93,000 packets/s, and end-to-end batch throughput of approximately 38,000 packets/s. The solution demonstrates effective domain adaptation, sub-millisecond latency, and suitability for resource-constrained IIoT edge gateways. Full article

Automated defect detection in precision agriculture serves as a critical technology for enhancing the quality of agricultural products. Although supervised-only semantic segmentation has demonstrated remarkable performance in citrus surface defect detection, it relies heavily on training with large-scale labeled data, which results in prohibitive acquisition costs. Semi-supervised learning mitigates reliance on labeled data by generating pseudo-labels. However, existing semi-supervised segmentation methods still face challenges. On the one hand, the instability of pseudo-labels and the propagation of noise can mislead the training of semi-supervised models. On the other hand, due to the lack of semantic constraints in feature learning, models often suffer from insufficient feature discriminability when handling complex samples, such as citrus surface defects characterized by similar textures and blurred boundaries. Therefore, this study proposes UP-ETS, a dual-guided semi-supervised semantic segmentation model based on the Mean Teacher–Student framework, specifically designed for the segmentation of complex citrus surface defects. UP-ETS employs Uncertainty Estimation (UE) based on Kullback–Leibler (KL) divergence to quantify the prediction discrepancy between the teacher and student models on blurred and ambiguous pixels. This mechanism guides the model to dynamically adjust weights, thereby reducing noise propagation and enhancing pseudo-label stability under complex citrus surface textures. Prototype Contrastive Learning (PCL) is utilized to align pixel-level features of difficult samples with class prototypes, optimizing the feature discriminability for complex citrus surfaces. Experimental results demonstrate that the UP-ETS model exhibits superior semi-supervised segmentation performance. Notably, at a labeled data ratio of only 1/16, the dice improved from 85.57% to 87.76% compared to the supervised-only baseline. Furthermore, the model shows significant performance enhancements in segmenting difficult samples, such as small targets, complex boundaries, and blurred regions. The results of ablation studies and t-SNE visualization prove the effectiveness of the proposed UE and PCL. These two methods synergistically guide the model to construct a feature space that is better structured and highly discriminative. Furthermore, UP-ETS outperforms various representative semi-supervised segmentation models in terms of segmentation performance, parameters, and inference speed. In cross-dataset validation, the model exhibits robust generalization capabilities, achieving performance comparable to supervised-only methods trained on the full augmented dataset. Consequently, the framework introduced in this study effectively mitigates the heavy dependency on annotated datasets, providing significant practical value for agricultural deployment. Full article

Neural architecture search (NAS) is a core technology in AutoML, but it faces challenges such as high computational costs and inefficient evaluation. Traditional NAS methods require fully training each candidate architecture, leading to substantial resource consumption and long evaluation times. This paper introduces a four-stage progressive multi-task learning framework that shifts from brute-force search to performance prediction. By progressively training from simple synthetic data to complex real data, the framework enables efficient architecture performance prediction. The main contributions are a unified progressive predictor paradigm, a deployment-aware multi-task prediction mechanism with dual lightweighting, and a benchmark-aware data-and-transfer framework based on NATS-Bench reconstruction and progressive knowledge distillation. Experiments on 15,000 NATS-Bench architectures with a fixed train–validation–test split (8:1:1) and consistent hyperparameters show a 95.56% correlation-based prediction score, computed as Pearson correlation expressed as a percentage (Pearson correlation 0.9556, R-squared 0.9134), 32-fold training efficiency improvement (37.87 s ± 2.1 s vs. 1247.6 s), and 95.7% convergence stability across five random seeds. Ablation studies, including a Direct Stage-3-Only comparison, quantify component contributions, and benchmarks compare against random forest, XGBoost, and MLP under identical data splits and feature spaces. Full article

With the rapid advancement of Earth observation technologies and the growing demand for intelligent remote sensing applications, high-resolution remote sensing imagery provides critical data support for a range of downstream applications, including land monitoring and disaster assessment. In this context, multi-label remote sensing image classification has become an important research task, because a single image may contain multiple ground-object categories with complex spatial distributions and semantic co-occurrence relationships. However, challenges such as the coexistence of multiscale objects, complex semantic dependencies, and long-tail category distributions impose significant limitations on existing methods in terms of feature representation capacity and class-balanced modeling. To address these challenges, a Multiscale Dynamic Reasoning Network (MSDR-Net) is proposed. Different from methods that focus on localized optimization for a single challenge, MSDR-Net establishes a task-driven modeling framework that jointly integrates multiscale feature extraction, label-aware semantic reasoning, and long-tail category optimization within an end-to-end architecture. The proposed network consists of three core modules. The Multiscale Feature Enhancement (MSFE) module incorporates a Feature Pyramid Network-based fusion mechanism, integrating deep semantic information with shallow, detailed features to effectively enhance the representation of multiscale objects. The Dynamic Semantic Reasoning (DSR) module introduces a Transformer-based global attention mechanism that models long-range dependencies among image features, enabling the capture of complex global semantic relationships. In the loss optimization stage, a Difficulty-Weighted Loss (DW-Loss) is introduced, which jointly incorporates category frequency weights and prior difficulty coefficients to dynamically regulate the contributions of rare classes and hard samples during training, thereby mitigating bias induced by class imbalance. Experiments conducted on the large-scale Detection in Optical Remote Sensing Images dataset demonstrate that the proposed method achieves superior performance. Ablation studies validate the effectiveness of each component, while comparative experiments indicate that MSDR-Net achieves a mean Average Precision of 95.88%, outperforming existing state-of-the-art methods. An improvement of approximately 1.74% is observed over the strongest baseline, MSCA, with consistent advantages demonstrated across Overall F1 and Class-wise F1 metrics. By unifying multiscale feature extraction, global semantic reasoning, and balanced loss optimization within a single framework, MSDR-Net provides a robust and efficient solution for multi-label classification in complex remote sensing scenarios. Full article

To address the critical challenges of edge chipping and poor processing quality in sapphire precision dicing, this paper proposes a femtosecond laser filamentation-guided dicing technology. By systematically investigating the influence of pulse overlap rate, energy, and scan counts on damage evolution, the physical differences between 343 nm UV and 515 nm visible lasers in suppressing plasma shielding and breaking through processing saturation limits are revealed. The results indicate that an extremely high pulse overlap rate (>98%) significantly inhibits lateral energy dissipation and drives the efficient propagation of the filament deep along the optical axis; furthermore, the 343 nm laser demonstrates superior removal rates and localisation compared to the 515 nm laser. Using super-resolution imaging, the precision cleavage cross-section is clearly categorised into four evolutionary stages: general ablation, filament ablation, transition, and mechanical cleavage. To mitigate morphological degradation induced by multiple scans, a sacrificial-layer-assisted strategy is innovatively proposed to achieve spatial damage transfer and in situ self-polishing, effectively eliminating longitudinal damage striations and residual stress-induced hackles. Finally, taper-free, high-precision separation of 1 mm × 450 μm micro-units is successfully achieved on a 220-μm-thick sapphire wafer. This technology not only achieves ultra-low-loss dicing but also establishes a highly efficient, contamination-free in situ characterisation paradigm for buried structures in hard and brittle materials. Full article

Safe and intelligent collision avoidance technology is essential for the autonomous underwater vehicle (AUV) to navigate in underwater environments. Most existing spike methods are constrained by a fixed static threshold and are unable to dynamically adjust to threshold changes reasonably, leading to difficulties in robustly adapting to external dynamic interference and thus resulting in insufficient homeostasis and generalization. To address these limitations, inspired by the dynamic threshold changes in biological neural systems, a bioinspired dynamic adaptive threshold (BDAT) is proposed. Combining the spiking neural network with deep reinforcement learning, a novel bioinspired dynamic adaptive threshold planner (BDAT-Planner) framework is constructed for underwater dynamic collision avoidance tasks performed by AUVs in complex, unknown environments. The proposed BDAT-Planner consists of the spiking dynamic adaptive actor network (SDAAN) and the deep critic normal network (DCNN). The BDAT is deployed to each spiking neuron in the SDAAN, dynamically adjusting the spike firing rate through threshold changes and avoiding excessive excitation or inhibition, thus maintaining homeostasis. The spiking encoder and spiking decoder are designed to convert continuous information and spiking sequences. Experimental results from both the training process and evaluation process (ablation studies, comparison experiments, and homeostasis experiments) demonstrate that the proposed BDAT-Planner has achieved superior performance in dynamic collision avoidance and model homeostasis compared to static threshold methods and existing comparison methods. The novel idea of bioinspired dynamic adaptive threshold can maintain model homeostasis and effectively enhance its adaptability to external dynamic interference, which offers significant development potential for promoting the efficient and stable operation of AUVs in marine environments. Full article

Large-scale electric vehicle (EV) integration provides considerable flexibility for integrated energy systems (IESs), but stronger incentives do not always yield proportional demand response benefits under physical operating limits. This paper proposes a techno-economic dispatch optimization framework for an IES incorporating a Kalina Cycle (KC), power-to-gas (P2G) technology, and hydrogen storage. A 24 h mixed-integer linear programming (MILP) model is developed to evaluate the interaction among the incentive intensity, multi-energy load response, and operating cost under different integrated demand response (IDR) scenarios. The results show a saturation-like response pattern: as the incentive intensity increases, the total operating cost rises monotonically, while the total demand response does not increase proportionally. In the studied case, the low-incentive strategy achieves the best economic performance because the additional compensation paid under high-incentive scenarios outweighs the marginal operating-cost savings. Load response analysis further indicates that EV-dominated electrical loads provide the main flexibility, whereas thermal and gas loads are strongly constrained by physical balance and equipment limits. The ablation results show that the saturation trajectory is affected by physical flexibility resources, especially the hydrogen subsystem. These findings suggest that incentive design should be coordinated with system physical constraints to avoid over-incentivization. Full article

Named entity recognition (NER) of ancient Chinese texts is the foundation for their development and utilization. Previous studies have focused on the data-driven methodology which tries to utilize the semantic features of ancient Chinese text. With the continuous accumulation of ancient Chinese linguistic resources and textual data, how to fully utilize the data resource and lexical knowledge related to ancient Chinese text with the help of new-generation information technology, so as to improve the ability of semantic comprehension and achieve good performance of NER, has become a great challenge to be solved. In view of this, this paper proposes a named entity recognition model for ancient Chinese text by fusing explicit feature and implicit feature (NERM), on the basis of extracting the explicit features and implicit features of ancient Chinese texts using a pre-trained model and a multi-head attention mechanism. In this model, the GuwenBERT model is introduced to extract the semantic features of ancient Chinese texts, namely the explicit features. The implicit features include relative positional relations, part-of-speech, and character radicals. The experimental results on the corpus GuNER 2023 show that the proposed model NERM achieves an F1 value of 90.67%, outperforming the existing models. The ablation experimental results show that implicit features provide a modest but meaningful improvement over explicit features, and implicit features can be arranged in order of importance as follows: character radicals, part-of-speech, and relative positional relations. Full article

Focused ultrasound (FUS)-mediated therapies have evolved with the advent of modern ultrasound-guided technology and MRI imaging, moving from their initial use as thermal ablation to a multifunctional platform for thermal and non-thermal ablation, immunomodulation, and targeted drug delivery. This narrative review explores the potential, limitations, and challenges of ablative high-intensity focused ultrasound (HIFU) therapies: HIFU thermal ablation and non-thermal ablation, histotripsy, as well as non-ablative low-intensity focused ultrasound (LIFU) applications in the management of hepatobiliary cancers. HIFU and histotripsy are reviewed as alternative or complementary treatment options in liver tumors, as well as their potential as bridging therapy. Histotripsy is addressed as a theranostic tool, not only by combining ablation with real-time ultrasound imaging guidance, but also by integrating it with sonobiopsy. It facilitates a liquid sonobiopsy of the ablated tumor by releasing intact tumor antigens and damage-associated molecular patterns, leading to potential molecular profiling. LIFU-induced targeted drug delivery (sono-chemotherapy), sonodynamic therapy, radiosensitization, immunomodulation of the immunosuppressive tumor microenvironment (sono-immunotherapy), and the potential to enhance the effect of immune checkpoint inhibitors in these malignancies are discussed. Since FUS-assisted procedures exhibit dual actions through therapeutic functionality associated with intra- and post-procedural ultrasound imaging guidance, they could have value as a theranostic tool in hepatobiliary interventional oncology. Although promising, the available clinical evidence for FUS-mediated therapies in hepatobiliary malignancies consists predominantly of early-stage feasibility studies, retrospective observational cohorts, and non-randomized comparative analyses. Further studies focused on standardized protocols, validation through large-scale, multicenter, prospective randomized clinical trials comparing FUS-based therapies with established treatments, and long-term follow-up of oncological efficacy could define their future role in multimodal oncological strategies. Full article

Lung cancer remains the leading cause of cancer-related mortality worldwide despite advances in screening, navigational bronchoscopy, and systemic therapies. Diagnostic and therapeutic limitations persist, including uncertainty regarding intraprocedural tissue adequacy during biopsy sampling and constraints of existing ablative modalities for tumors located near critical thoracic structures. This review examines two emerging technologies: Full-Field Optical Coherence Tomography-based Dynamic Cell Imaging (DCI) and monopolar biphasic Pulsed Electric Field (PEF) ablation as complementary emerging technologies that may address these gaps. The Van Gogh™ Microscopy System (CellTivity Scientific, Inc.) utilizes DCI to enable real-time visualization of cellular metabolic activity without tissue destruction, providing functional information regarding tissue viability and microstructural morphology. The Aliya^® PEF ablation system (Galvanize Therapeutics, Inc.) delivers biphasic high-voltage electrical pulses that induce non-thermal tumor cell death while preserving extracellular matrix architecture, potentially allowing treatment near sensitive thoracic structures such as airways, vasculature, and pleura. Early preclinical studies and initial clinical experience suggest that DCI can facilitate rapid intraprocedural assessment of biopsy adequacy, while PEF ablation may provide reproducible focal tumor destruction with a favorable safety profile near critical structures. Although the current evidence base remains limited to early-phase studies and feasibility trials, the convergence of real-time biologic tissue assessment with structurally preserving ablation technologies introduces the possibility of integrating diagnostic confirmation and local therapy within a single procedural workflow. This review summarizes the mechanistic rationale, emerging evidence, and potential clinical applications of these technologies and proposes a conceptual “See and Strike” framework within these two emerging technologies. The methodological limitations, workflow considerations, and future research directions required to validate this approach are also discussed. Prospective multicenter trials and long-term oncologic outcomes will be necessary before widespread clinical adoption. Full article

This article reviews the state of the art of laser ablation and deposition techniques applied so far to more than 50 elements, including metals, metalloids and lanthanides, yielding a wide variety of compounds in the form of thin films. Laser deposition processes have been performed in high-vacuum (HV) reactors at pressure values ranging between 10⁻¹ and 10⁻⁵ Pa, namely pulsed laser deposition (PLD), or, under different reactive gas ambient (O₂, N₂, CH₄, NH₃ and many others), so-called reactive pulsed laser deposition (RPLD), with the aim to form thin films with desirable chemical compositions. While a few metals have not been deposited as pure metallic films because they have no immediate technological interest, others, like alkali and alkaline earth metals, cannot be deposited in pure metallic form due to their very strong reactivity with oxygen, water vapor and hydrogen molecules which are always present, even in ultra-high-vacuum (UHV) systems, at pressure values of 10⁻⁵–10⁻¹⁰ Pa. Furthermore, elements of the Mendeleev periodic table with an atomic number higher than 88, such as actinides and synthetic elements, are dangerous to handle and deposit in the form of thin films due to their high radioactivity; therefore, they are excluded from this review. The inclusion of the non-metal thin films of carbon (C) and related chemical compounds prepared by PLD and RPLD in the present review is justified by the extensive research and the numerous scientific articles reported in the field. All the results obtained by PLD and RPLD techniques so far are discussed and presented in tabular format to guide the reader. Full article

Soft tissue sarcomas (STS) are locally invasive and aggressive tumors that occur spontaneously in humans, dogs, and cats. High-intensity focused ultrasound (HIFU) is a non-invasive ablation technology that has been explored in canine but not feline STS. The objective of this pilot study was to determine the in vivo safety and feasibility of HIFU ablation for feline STS and to investigate the impact of HIFU on the acute immunological response. Client-owned cats diagnosed with spontaneous STS were recruited. Computed tomography (CT) scans of the chest, abdomen, and tumor were performed prior to treatment for staging and treatment planning. A commercially available HIFU unit (Echopulse, Theraclion, Malakoff, France) was used to target portions of solid tumors before standard-of-care surgical resection. Ablation efficacy and local immunological response were characterized using histopathological and immunohistochemical assessments. Acute safety was monitored with physical examinations, owner reports, and CBC/serum biochemistry. Multiplex serum cytokine levels were used to evaluate the systemic immune response. A total of three cats diagnosed with STS were recruited and treated. No significant adverse events attributed to HIFU treatment were noted in this pilot study. In treated areas, hemorrhage as well as coagulative and lytic necrosis were observed microscopically and were more extensive than in untreated tissues. There was a statistically significant difference in the level of serum MCP-1 after HIFU treatment, but no significant changes in any other analytes. No differences in the infiltration of CD3-, CD79a-, or IBA1-positive cells were noted between treated and untreated samples. Overall, findings suggested that HIFU may offer a viable alternative to conventional therapies for feline STS, with pilot results showing effective tumor ablation in cats with STS without significant adverse events. Some preliminary evidence of immunomodulation following treatment was observed, but HIFU as an immunotherapeutic treatment option needs to be further investigated. Full article