Search Results (1,516)

Real-time fatigue classification on resource-constrained edge devices faces three unresolved computational challenges: just-in-time compilation latency spikes that violate the 50 ms inference budget, statistical moment features insensitive to temporal complexity signatures of fatigue, and binary anomaly outputs insufficient for actionable coaching decisions. A synthetic IMU dataset (9 subjects, 540,000 samples, 6 channels at 100 Hz) was generated as a reproducible computational benchmark, with fatigue signatures calibrated to published biomechanical effect sizes (sample entropy $d=+0.77$; permutation entropy $d=+0.38$). We present Safari (Stochastic Adaptive Fitness-Aware Real-time Inference), an end-to-end computational pipeline integrating: a dual-pathway entropy triplet (SampEn, PermEn, SpEn) replacing statistical moments; 16 pre-compiled polyhedral anchor kernels eliminating JIT latency; $O\left({(\Delta W)}^2\right)$-bounded runtime interpolation; subject-specific MaxEnt free-energy anomaly scoring; and a Banister fitness–fatigue adaptive threshold. Safari achieves AUC-ROC = 0.9820 (Monte Carlo 95% CI: 0.9726–0.9886), F1 = 0.8835, four-state accuracy = 83.3%, and worst-case latency = 7.2 ms on a Raspberry Pi 4. Entropy features achieve 1.55× higher discriminability than statistical moments. Safari is a computational framework for real-time fatigue monitoring, contributing a reproducible algorithmic benchmark for edge AI in movement analysis, with real-athlete validation as the recommended next step. Full article

Background/Objectives: Treatment intensification with novel hormonal therapies is now standard in metastatic hormone-sensitive prostate cancer (mHSPC), but real-world patients are often more heterogeneous than those included in pivotal trials. This study aimed to describe clinical characteristics, safety, treatment discontinuation, disease progression, polypharmacy, and clinically documented drug–drug interactions in a real-world mHSPC cohort. Methods: We conducted a retrospective observational study including 109 patients with mHSPC who initiated abiraterone, enzalutamide, apalutamide, or docetaxel-based triplet regimens between January 2015 and November 2025. Outcomes included adverse events, discontinuation, PSA50 response at 3 months, time to progression and overall survival. Descriptive analyses and Kaplan–Meier estimates were performed. Results: Apalutamide was the most frequent first-line treatment (56.9%), followed by abiraterone (23.9%), enzalutamide (11.9%), darolutamide-based triplet therapy (3.7%), and abiraterone-based triplet therapy (3.7%). Median age was 73 years, and median baseline PSA was 16.1 ng/mL. De novo metastatic disease was present in 69.7% of patients, ISUP grade 4–5 disease in 58.7%, high-risk disease according to LATITUDE criteria in 45.9%, and high-volume disease according to CHAARTED criteria in 38.5%. The median Charlson Comorbidity Index was 4, and polypharmacy was observed in 68.8%. Adverse events occurred in 56.0%, and non-death treatment discontinuation occurred in 22.0%. No documented drug–drug interactions requiring treatment modification were recorded. PSA50 response was achieved by 97.2%. Thirteen patients (11.9%) progressed and 18 (16.5%) died. Median time to progression was not reached, and median overall survival was 53.2 months. Conclusions: Novel hormonal therapies were used in a clinically heterogeneous real-world mHSPC cohort. The findings support individualized treatment assessment and should be interpreted as descriptive and exploratory. Full article

Melanin pigments protect human tissues from ultraviolet and visible radiation, yet their phototoxic potential increases with oxidative degradation. This computational study investigates how the oxidation state influences the first excited states of eu- and pheomelanin using molecular models of varying complexity (monomers to tetramers, both covalently and non-covalently bonded). First, vertical and adiabatic electronic transitions were computed, and supramolecular interactions were characterized with the ETS-NOCV method. In eumelanin, oxidation drastically lowers the first triplet-state (T₁) energies (from above 230 kJ/mol) to levels comparable to retinal carotenoids (≤66 kJ/mol), emphasizing its role in triplet quenching rather than singlet oxygen generation. Pheomelanin showed greater heterogeneity in the values of the first triplet state, staying mostly above the eumelanin T₁ energies. However, selected pheomelanin structures also exhibited relatively low triplet energies, particularly oxidized benzothiazole (BZox) and trichochromes, and although their T₁ energetics remained higher than those calculated for oxidized eumelanin, they were still sufficiently low to suggest a potential ability to quench singlet oxygen. Furthermore, supramolecular analysis reveals that eumelanin aggregates are moderately stabilized by both π-π stacking and hydrogen bonding, whereas pheomelanin aggregates are dominated by dense hydrogen-bond networks. Full article

Images captured by vision sensors provide visual evidence for scene understanding, including object appearances, pixel-level regions, and spatial relations among entities. Panoptic Scene Graph Generation (PSG) constructs structured scene representations by grounding visual entities with panoptic masks and predicting relationships among objects and regions. In pair-then-relation PSG pipelines, subject–object pair recall is critical to final triplet recall. However, existing pair proposal approaches mainly score candidate subject–object pairs based on object–query feature matching, while mask-derived spatial cues such as object locations, relative geometry, and local layouts remain underexplored. Consequently, ground-truth subject–object pairs may be excluded from the Top-$K_r$ proposals before relation decoding. To address this problem, this paper proposes a Spatially Aware Pair Proposal Model (SAPPM), which incorporates mask-derived soft centroids, relative geometry, and local-neighborhood context into pair scoring. SAPPM uses Grouped Vector Attention (GVA) to model local spatial interactions and introduces a spatially adaptive gating module to calibrate spatial-branch contributions. Experiments on the PSG dataset under the Scene Graph Detection (SGDet) protocol show that SAPPM achieves competitive performance, reaching 32.53 R@20 and 27.36 mR@20. These results indicate that SAPPM improves PSG performance by enhancing ground-truth pair coverage in the candidate proposal set. Full article

Background: Chest X-ray (CXR) foundation models can predict patient demographic categories (sex, age, race) from images alone by linear probing, but whether encoded attributes drive finding prediction has not been tested at scale. Methods: On MIMIC-CXR (230,697 images, 60,518 patients), we measured attribute dependence (AUROC drop after residualizing an attribute from a frozen embedding) across 24 patient attributes (four demographics and 20 ICD-coded comorbidities), 10 thoracic findings, and 6 overlap-free foundation models ($n=1440$ triplets), with 3 additional CXR-pretrained models (RAD-DINO, CheXzero, CheSS) for encoding and fairness analyses. Dependence was regressed on attribute-finding odds ratios (ORs), encoding strength, and model-level factors. Results: Encoding and dependence dissociated. Sex (AUROC 0.942) contributed $<0.001$; race (0.83) contributed $0.0015$ (rank 14/24); heart failure (0.774) showed the largest dependence ($0.018$). $\left|\log \right(\mathrm{OR}\left)\right|$ explained 50.6% of dependence variance ($\beta =0.029$, $p<{10}^{-15}$); model factors added no detectable contribution ($\Delta R^2=0.000$, $n=6$). Residualizing the top three high-OR attributes reduced AUROC by 0.026 without narrowing sex or age subgroup gaps (minimum detectable effect size (MDES) = 0.0019). Across 9 models, four-category race subgroup gaps (mean 0.069) were 30–75× larger than race residualization drops (mean 0.0015); CheXzero showed the same decoupling. Conclusions: Encoding, residualization-sensitive dependence, and subgroup bias are three separable quantities on the same model. Pre-deployment audits on inpatient-skewed cohorts can prioritize attributes by local OR; jointly residualizing race and its cardiac correlates does not narrow the race subgroup gap, which instead tracks group-wise finding base rates. Cross-institutional transfer remains open: no public CXR cohort currently links comorbidity electronic health records for external validation of the OR-dependence relationship. Full article

The prediction of student engagement characteristics involves forecasting and analyzing student interaction with educational materials using engagement prediction models. This process encompasses the prediction of cognitive, behavioral, and emotional dimensions of engagement. The existing student engagement prediction models have some limitations, including poor convergence, less generalizability, complexity issues, overfitting, false errors, and limited resources. Hence, the research proposes the Multi-Instance Localization-based Gradient Boosted Long Short-Term Memory (MIL-GBLTM) model to tackle the challenge of predicting student engagement characteristics in online classes. The integration of effective MIL with a Triplet Attention mechanism focuses on the significant features that help with engagement prediction; LSTM layers capture intricate sequential patterns, and fractional gradient boosting is used for fine-tuning for accurate prediction, alongside ensemble-based learning. The LSTM layers with the Triplet Attention module refine temporal attention, and Fractional Gradient Boosting ensures the model’s adaptability and robustness. By combining these components, the proposed model is able to predict accurate student engagement with high convergence. This integrated approach enhances the capabilities of engagement prediction models in educational contexts, facilitating more effective interventions and personalized student support in online learning environments. Experimental results demonstrate that the proposed MIL-GBLTM model outperforms other existing models by achieving the highest accuracy of 96.55% with a k-fold of 10, utilizing the wacv2016 dataset. Full article

In this work, we report an example of tuning the photophysical properties of a polypyridine ruthenium(II) complex via the coordination of a second cation. A new ruthenium(II) complex contains a thiacrown-ether fragment that allows selective binding of additional metal cations (Ba²⁺, Cd²⁺, Pb²⁺), leading to pronounced changes in the optical and electronic properties of the bimetallic system. Spectroscopic and electrochemical studies reveal that the monoruthenium precursor displays dual excitation pathways involving either intraligand charge transfer (ILCT) or triplet metal-to-ligand charge transfer (³MLCT) excited states. Upon coordination of a second metal ion, the ILCT channel is suppressed, and only the ³MLCT state remains emissive, resulting in a significant increase in phosphorescence quantum yields (up to 22.6% in degassed solutions) for the bimetallic derivative. Time-resolved emission studies confirm the conversion from biexponential to monoexponential luminescence decay upon complexation. Electrochemical analysis and density functional theory (DFT) calculations support the hypothesis that cation binding alters the electron density distribution within the chromophore, stabilizing the MLCT pathway. These results demonstrate that incorporation of a second cation provides an effective strategy to control excited-state dynamics in ruthenium complexes, offering opportunities for the rational design of photosensitizers and photofunctional materials. Full article

Across all kingdoms of life, ribosomes are indispensable molecular machines that translate genetic information into the proteome of living cells. The fundamental catalytic centers of the ribosome, constructed primarily from ribosomal RNA (rRNA), exhibit remarkable conservation between the major domains of life. The ribosome’s A-site deciphers the mRNA’s triplet code, while the P-site synthesizes the growing protein chain and the E-site provides exit for deacylated tRNA; a distinct tunnel facilitates nascent polypeptide export. While the conservation of ribosomal proteins is less pronounced between bacteria and eukaryotes, striking homology exists from simple eukaryotes to humans. Ribosomal proteins were traditionally viewed mainly as scaffolding agents, steering rRNA folding during ribosome biogenesis and maintaining structural stability during translation. However, since the early 2000s, advances in structural and functional ribosome analysis have ushered in a more nuanced paradigm: ribosomes are no longer considered uniform machines. Instead, an array of rRNA and ribosomal protein modifications generates a spectrum of ribosome populations capable of specialized translation. RiboScreen^TM technology leverages this regulatory potential of individual ribosomal proteins, enabling deliberate modulation of target protein output and representing a promising tool for correcting dysregulated protein expression involved in rare and common diseases. This review will first introduce relevant aspects of ribosome biology and then showcase the tools of this new technology. Finally, we report examples for the delivery of small molecules to target ribosomal proteins for tailored restoration of protein production levels in rare and prevalent diseases. Full article

Background/Objectives: To derive and validate a simplified modification of the Haigis intraocular lens (IOL) power calculation formula by reducing the three-constant effective lens position (ELP) model to a single constant while introducing an optimized keratometer index and axial length correction. Methods: In this retrospective study, a large multicentric dataset (Dataset 1; 22,466 eyes, 113 IOL models) was used to optimize the Haigis constant triplet and keratometer index using nonlinear programming with Cooke’s axial length correction. A second independent dataset (Dataset 2; 3181 eyes, six IOL models) was used for cross-validation. Three approaches were compared: classical Haigis, modified triplet, and two single-constant models acting on IOL power (H1) or ELP (H2). Results: The optimized keratometer index (1.3296 ± 0.0003) was significantly lower than the classical value, indicating systematic overestimation of corneal power. Modified triplet and single-constant approaches achieved comparable or slightly lower prediction errors than the classical formula. The H1 approach showed marginally superior performance. Bootstrapping confirmed parameter stability. Conclusions: A single-constant modification of the Haigis formula incorporating an optimized keratometer index and axial length correction maintains prediction accuracy while simplifying clinical implementation. Full article

Instance segmentation of surface defects is one of the research hotspots in the field of image segmentation. Due to limitations such as restricted receptive fields or the loss of fine-grained details, traditional neural network models still struggle to achieve sufficiently high-segmentation accuracy for surface defects. To meet the demand for high precision segmentation of surface defects on copper strips and plates in industrial quality inspection, this paper proposes a feature fusion segmentation network, termed DSFFNet. First, a dual-branch structure is designed in DSFFNet to fuse spatial-domain features with discrete cosine transform (DCT)-domain features, thereby obtaining richer feature information. Second, a 2D-DCT frequency feature extraction module is developed to more effectively capture the edge information of targets. Third, a triplet attention mechanism is introduced into the backbone network to form an attention-centric network. Finally, a bidirectional fusion module and a multi-scale fusion network are designed to capture finer-grained feature information. Comparative experiments conducted on the KUST-SEG-Dataset demonstrate that DSFFNet achieves 94.66% ± 1.07% (mask)mAP₅₀ and 95.38% ± 0.06% (box)mAP₅₀, outperforming several classic image segmentation methods. Furthermore, generalization experiments on the public NEU-Seg dataset yield a (mask)mAP₅₀ of 86.27% ± 0.01%. The generalization results indicate that DSFFNet is robust to datasets with similar defect types. Full article

LiDAR-based place recognition is critical for long-term autonomous navigation in Global Navigation Satellite System (GNSS)-denied environments, yet existing methods struggle to balance accuracy and efficiency under substantial yaw rotations. This paper proposes a robust framework based on a multi-channel polar bird’s-eye-view (BEV) representation. Under yaw-dominated revisits, the polar BEV image transforms yaw rotation into cyclic column shifts, providing a useful structural prior for rotation-equivariant feature extraction. Raw point clouds are projected onto polar BEV grids encoding density, height, and intensity. A rotation-equivariant feature extractor comprising a Radial Compression Module and a rotation-equivariant Transformer module captures long-range azimuthal dependencies via Conditional Positional Encoding and Circular Relative-Position Bias. The equivariant features are aggregated by NetVLAD into a compact global descriptor, trained end-to-end with a hard-example mining triplet loss. Extensive experiments on the public KITTI and NCLT datasets, as well as our self-constructed LiDAR Place Recognition Revisit (LPRR) dataset, demonstrate competitive performance on KITTI and superior performance on NCLT and LPRR among the compared methods. The proposed framework achieves a favorable trade-off between performance and computational cost, and shows promising cross-dataset generalization on the evaluated NCLT and LPRR datasets without fine-tuning. Full article

Thiopurines are efficient triplet-state photosensitisers; however, the practical application of canonical derivatives such as 6-thioguanine (6TG) and 6-thioguanosine (6TGuo) is limited by competing deactivation pathways that reduce the fraction of triplet states available for productive interaction with molecular oxygen. In this work, we investigated how structural modification of the thiopurine scaffold through introducing of an additional five-membered etheno ring affects triplet-state energetics, deactivation pathways, and singlet oxygen sensitisation. The photophysical properties of four tricyclic thiopurine analogues—9-thio-1,N²-ethenoguanine (TEGua), 9-thio-1,N²-ethenoguanosine (TEGuo), 6-methyl-9-thio-1,N²-ethenoguanine (6MeTEGua), and 6-methyl-9-thio-1,N²-ethenoguanosine (6MeTEGuo)—were investigated using steady-state spectroscopy, low-temperature phosphorescence, nanosecond transient absorption spectroscopy, and direct detection of singlet oxygen phosphorescence. All investigated compounds exhibited efficient intersystem crossing and microsecond-lived triplet states. Compared with canonical thiopurines, the tricyclic analogues displayed lower triplet-state energies and significantly enhanced singlet oxygen generation. Quantum yields of singlet oxygen sensitisation reached ~0.56 in acetonitrile, approximately twofold higher than those observed for 6TG and 6TGuo under identical conditions. Analysis of triplet-state deactivation pathways showed that the enhanced photosensitising efficiency does not result from increased triplet formation, but from more effective use of the triplet-state population for energy transfer to molecular oxygen leading to singlet oxygen formation. These findings demonstrate that structural modification of the thiopurine scaffold enables control over triplet-state reactivity and provides a strategy for designing improved thiopurine-based photosensitisers for photodynamic therapy applications (PDT). Full article

Systemic therapy for metastatic prostate cancer is increasingly defined by upfront intensification and biomarker-guided mechanism switching. This narrative review synthesizes pivotal randomized trials, guideline recommendations, and implementation-focused literature across metastatic castration-sensitive prostate cancer (mCSPC) and metastatic castration-resistant prostate cancer (mCRPC). Evidence was organized around decision points at mCSPC diagnosis and at mCRPC transition, while incorporating biological mechanisms of resistance, including AR-axis reactivation, AR splice variants, lineage plasticity, DNA repair–hormone signaling interactions, and PSMA expression heterogeneity. In mCSPC, androgen deprivation therapy plus docetaxel and/or androgen receptor signaling inhibitors (ARSIs) improves survival, with triplet regimens favored for selected chemotherapy-fit patients with aggressive de novo disease. In mCRPC, cross-resistance limits routine ARSI-to-ARSI switching, and randomized data support mechanism-distinct options, including taxanes, PARP-based therapy in homologous recombination repair (HRR)-altered disease, and PSMA-targeted radioligand therapy (RLT) in selected PSMA-positive patients. As RLT moves earlier, PSMA heterogeneity, renal function, bone marrow reserve, and emerging dosimetry-based optimization should inform practical implementation. Ongoing trials are evaluating earlier theranostics, alpha-emitting radioligands, and biomarker-enriched combinations. An implementation-first approach that intensifies treatment when appropriate, tests early and acts on HRR results, uses PSMA PET to guide RLT, and preserves hematologic reserve may maximize access to multiple life-prolonging mechanisms. Full article

Aircraft cargo compartment fires constitute a significant type of aviation fire, posing a grave threat to aviation safety. To guard against and respond to such fires, existing aircraft cargo compartments are equipped with smoke detection fire detectors, which rely on perceiving changes in smoke transmittance to determine the onset of a fire. However, these detectors offer relatively low recognition accuracy and cannot provide a direct visual representation of the fire. In this work, we introduce a fire recognition method built on image sensors and a deep learning model. In light of the irregular shapes of flames and smoke, an improved interactive triplet attention mechanism (ITAM) is integrated into the You Only Look Once version 5 (YOLOv5) model, enhancing the model’s recognition accuracy. Furthermore, the original Neck structure is replaced with an Asymptotic Feature Pyramid Network (AFPN), improving the model’s ability to recognize small targets, which is particularly useful for detecting flames and smoke early in a fire. This paper further improves the model’s recognition accuracy by introducing the Focaler-IoU loss function, which balances the feature learning of hard and easy samples. Therefore, the network model in this paper is named IATF-YOLO. Ablation experiments demonstrate that our algorithm improves accuracy by 2%, while comparative experiments with several mainstream baseline models show that our algorithm achieves a 0.7% accuracy improvement, with a final peak accuracy of 93.6%. Full article