Burkholderia gladioli pathovar cocovenenans produces the potent mitochondrial toxin bongkrekic acid (BA), causing severe foodborne outbreaks. Conventional detection methods are constrained by prolonged turnaround times and the need for sophisticated instruments, limiting their utility for on-site surveillance. In this study, two independent one-step RPA-CRISPR/Cas13a detection systems were established: one targeting a species-level marker for broad screening of B. gladioli and the other targeting the bonA gene cluster for specific identification of the toxigenic pathovar. The complete one-step reaction was performed at 37°C for 30 min, and results were visually interpreted using a portable blue-light transilluminator. The visual detection limit reached 1 × 10⁰ copy/μL for both assays, with no cross-reactivity observed against a panel of six non-target strains. Component validation experiments confirmed that all listed constituents were indispensable for fluorescence signal generation. The practical applicability of the platform was evaluated using 100 naturally contaminated black fungus and mushroom substrate samples. The species-level assay returned positive rates of 78% (39/50) for black fungus and 42% (21/50) for substrate samples, while the bonA-targeted assay yielded positive rates of 38% (19/50) and 8% (4/50), respectively. The visual readout results were in 100% concordance with conventional quantitative real-time PCR. This field-deployable, equipment-free visual detection platform holds considerable promise for on-site surveillance and early warning of BA-producing B. gladioli in food and environmental matrices.IMPORTANCEBongkrekic acid (BA) poisoning caused by Burkholderia gladioli pv. cocovenenans is a lethal food safety threat, particularly in China and Southeast Asia. Current detection methods depend heavily on laboratory-based instruments, limiting their utility for on-site surveillance. This study developed two independent one-step RPA-CRISPR/Cas13a detection systems: one for broad screening of B. gladioli at the species level and the other for specific identification of the toxigenic pathovar. The one-step reaction was completed at 37°C within 30 min, and results were visually interpreted using only a portable blue-light transilluminator. The visual detection limit reached 1 × 10⁰ copy/μL, and the method exhibited 100% concordance with quantitative real-time PCR when applied to 100 naturally contaminated samples. This field-deployable, equipment-free platform provides a practical tool for on-site surveillance and early warning of BA-producing B. gladioli in food and environmental matrices.

We report a case of delayed-onset hyphema with reflux bleeding adjacent to the PreserFlo MicroShunt (PFM) tube entry site. A 66-year-old Japanese man with bilateral primary open-angle glaucoma underwent PFM implantation in the left eye because of elevated intraocular pressure (IOP) and progressive visual field loss despite treatment with four topical anti-glaucoma agents. He had previously undergone cataract surgery and microhook ab interno trabeculotomy (μLOT) in the same eye. Because of insufficient IOP control and further visual field progression after PFM implantation, Ahmed glaucoma valve implantation with pars plana tube insertion was subsequently performed in combination with 25-gauge pars plana vitrectomy. At two years and three months after PFM implantation, a small number of red blood cells were observed floating in the anterior chamber. Gonioscopy showed that the PFM tube was positioned between the pigmented and non-pigmented trabecular meshwork, with reflux bleeding from both sides of the tube entry site. The previous μLOT clefts did not overlap with the PFM tube entry site. The hyphema resolved with observation alone, although possible recurrence was noted six months later. This case demonstrates that delayed-onset hyphema can occur more than two years after PFM implantation and that careful gonioscopic examination can help identify the source of bleeding.

Extremely low-frequency magnetic fields (ELF-MF; <300 Hz) can induce visual percepts known as magnetophosphenes through weak, subthreshold electric fields in neural tissue. Although magnetophosphene perception has been characterized in healthy observers, the anatomical locus of interaction (retinal vs. cortical) remains incompletely constrained, particularly under conditions of severe retinal dysfunction.

Decompressive craniectomy (DC) is a life-saving intervention for refractory intracranial hypertension across traumatic, ischemic, and hemorrhagic etiologies. While its immediate hemodynamic benefits are well established, DC fundamentally disrupts intracranial physiology by abolishing the closed-box condition of the cranial vault, thereby altering the interpretation of all monitored parameters. Neuromonitoring (NM) after DC remains a largely unaddressed domain in current consensus statements and evidence-based guidelines. This viewpoint synthesizes available evidence to address five critical clinical questions: the pathophysiological changes induced by DC; the rationale for continued invasive monitoring; the optimal sensor placement site; the contribution of noninvasive monitoring modalities; and whether conventional ICP and CPP thresholds remain applicable in the decompressed brain. Observational data indicate that intracranial hypertension persists or recurs in 36-76% of patients post-DC, underscoring the continued value of monitoring. Emerging evidence challenges the universal applicability of standard ICP thresholds, with post-DC data suggesting clinically relevant outcomes at substantially lower values. These findings highlight a fundamental gap in the field and underscore the urgent need for prospective, standardized research in the decompressed brain.

Remote sensing referring image segmentation faces critical challenges including arbitrary target rotation, drastic scale variation, cluttered complex backgrounds, and large visual-language semantic gaps. Existing mainstream segmentation models adopt fixed-receptive-field backbones, coarse unidirectional cross-modal interaction and static learnable object queries, which easily cause small-object missed detection, blurred boundary segmentation and fragmented predictions. This paper proposes a context-aware referring expression segmentation model for remote sensing to tackle the above limitations. We construct a dual-stream feature extraction backbone with InternImage and CLIP text encoder, and design a semantic prior localization map module to generate spatial heatmaps for spatial inductive bias and improve small-object localization recall. A cross-modal context aggregator performs multi-scale bidirectional visual-text alignment, while a dynamic query initialization strategy and a language-guided Transformer decoder progressively improve target localization and mask refinement. Experiments are conducted on two standard RRSIS benchmarks, RefSegRS and RRSIS-D. On RefSegRS, the proposed model achieves an mIoU of 70.43% and an oIoU of 77.34%, and obtains significant improvements on small vehicles, buildings and slender road markings. On the larger-scale RRSIS-D dataset, our model achieves an mIoU of 63.71% and an oIoU of 73.68%. The proposed model provides a solution for accurate multi-scale target segmentation guided by natural language descriptions in complex remote sensing scenes.

Computational imaging shifts part of the aberration correction from optical hardware to algorithms, offering a viable path toward compact, simplified systems. However, overexposed regions-often caused by phenomena such as water-body reflections-can readily induce severe ringing artifacts in reconstructed images. To address this problem, we propose a Ringing-perceptive Cooperative Reconstruction Network (RPCR-Net). This network integrates a learned Wiener filter and a field-of-view shared kernel prediction network (FOV-KPN) for feature extraction and innovatively incorporates a combined regularization mechanism that leverages a Local Maximum Gradient Prior and a multi-scale ringing measurement model within its loss function to suppress artifacts while preserving details. Validated on a constructed overexposed image dataset, RPCR-Net improves the Peak Signal-to-Noise Ratio (PSNR) from 29.08 dB to 37.06 dB and the Structural Similarity Index Measure (SSIM) from 0.8795 to 0.9549. Experiments on real-world scenes further confirm its capability to suppress ringing artifacts while maintaining visual quality. The proposed method can generate high-quality images such as image reconstruction and robustness improvement in optical systems.

Background: Optical neuropathy is a severe complication in patients with type 2 diabetes mellitus (T2DM) undergoing vitreoretinal surgeries. Early diagnostic indicators are critical to preventing irreversible vision loss. This study aimed to evaluate the alterations in blood serum amino acid profiles, with a specific focus on serine levels, and their potential clinical association with the severity of neuroretinal impairment. Methods: The clinical cohort included 28 patients aged 32 to 77 years with T2DM following vitreoretinal interventions. Visual field functional status was assessed using Humphrey automated perimetry, with mean deviation (MD), pattern standard deviation (PSD), and visual field index (VFI). Structural changes were quantified using optical coherence tomography (OCT) to measure peripapillary retinal nerve fiber layer (RNFL) thickness. Serum amino acid profiles were determined using an ion-exchange chromatography amino acid analyzer, alongside a reference group (n = 12). Results: All patients presented with moderate-to-severe visual field impairment. A profound decline in serum serine levels was observed in diabetic patients compared to the reference group (p < 0.001). The reduction in serine content closely mirrored both the degree of retinal sensitivity loss (MD and PSD values) and the number of thinned peripapillary RNFL sectors. Notably, the compensatory synthesis of serine via glycine cleavage appeared insufficient, as evidenced by a concomitant statistical decrease in glycine content (p < 0.05). No significant differences in serine levels were found between the subgroups when stratified strictly by age or isolated severity stages. Conclusions: It was demonstrated that the decline in serum serine levels in patients with T2DM following vitreoretinal interventions is significantly associated with the severity of optic neuropathy as assessed by optical coherence tomography (OCT). Consequently, serum serine may serve as a valuable dual biomarker, acting as a prognostic indicator for retinopathy progression and a predictive marker for early-stage neuropathy development.

Semantic segmentation for autonomous driving is challenged by limited field-of-view (FoV), where occlusions and restricted viewpoints reduce available visual information. Existing approaches often incur considerable computational overhead, which is unsuitable for autonomous driving systems. To build robust and lightweight semantic segmentation models, attention-based knowledge distillation (KD) can be an efficient alternative. Nevertheless, KD remains primarily visual and provides limited information when visual cues are incomplete or ambiguous. In contrast, biological visual systems integrate visual cues with contextual and semantic knowledge to support robust scene perception. Motivated by this biomimetic principle, language-guided methods provide high-level semantic knowledge, offering another promising direction for enhancing visual understanding. However, the effectiveness of attention-based KD under limited FoV conditions and its interaction with semantic knowledge remain largely unexplored. To systematically investigate this problem, we devise Semantic-guided Attentive Feature Distillation (SAFD), a KD-based framework that combines attention-based feature distillation with semantic guidance derived from large language models (LLMs) without introducing additional inference-time computation. Through comprehensive analyses of multiple distillation strategies and textual representations using the framework, we observe that semantic guidance generally provides additional benefits when the gains from visual distillation are limited, while its effectiveness varies across different attention-based distillation strategies and textual representations. Furthermore, our analyses reveal that different attention mechanisms exploit semantic guidance in distinct ways, with consistent benefits across different pretrained text encoders and improved generalization. These findings provide practical insights into integrating linguistic semantics with visual knowledge transfer under limited FoV.

The visual system adapts to the statistics of everyday experience, yet these statistics depend on the observer's physical body. We combined large-scale egocentric video and behavioral testing to examine how stature shapes the "visual diet" of hands and faces and how this diet affects perceptual performance. In Experiment 1, automated analyses of over 200 hr of eye-tracked egocentric video revealed that taller observers encountered faces lower in the visual field and hands higher, reflecting biomechanical constraints on gaze and posture. In Experiment 2, observers showed greater categorization accuracy to faces in the upper field and to hands in the lower field, with stature modulating these effects-shorter participants were more attuned to faces above them. These findings demonstrate that perceptual performance mirrors the embodied statistics of visual experience: The visual system is tuned not only to our shared environment but also to the vantage point of our bodies. (PsycInfo Database Record (c) 2026 APA, all rights reserved).