Lubricating oil monitoring provides continuous health information for early fault warning and maintenance decision-making in industrial gas turbines. However, real-world multi-sensor monitoring streams exhibit pronounced nonstationary thermodynamic drifts that often obscure subtle high-frequency residuals containing critical incipient degradation signatures. Prevailing data-driven monitoring models typically struggle to separate these macroscopic trends from stochastic wear-related fluctuations, and their restrictive distributional assumptions are often inadequate for the heteroscedastic and heavy-tailed nature of industrial residuals. To address these challenges, this study proposes ResAD-Net, a framework for early fault warning in nonstationary multi-sensor oil monitoring that combines trend–residual decoupling, trend-conditioned residual modeling, and residual-domain dependency learning. Specifically, a signal trend–residual decoupling strategy is adopted to separate slowly varying operational trends from stochastic residual fluctuations captured by the sensors, thereby exposing residual information that is more sensitive to incipient degradation. On this basis, a trend-conditioned diffusion model is introduced to characterize state-dependent, skewed residual distributions and generate residual sample ensembles for nonstationary monitoring. Meanwhile, a graph-based variational autoencoder is employed to learn latent intersensor dependency structures from the residual domain, providing diagnostic cues for temporal risk evolution analysis and sensor-level inspection. Experiments on a real-world industrial oil-monitoring record show that the proposed framework achieves an average F1-score of 0.985 with no observed false positives in the predefined pre-alarm reference interval of the finite test set. In addition to accurate anomaly detection, ResAD-Net captures early residual distributional shifts before clear macroscopic deviations emerge and provides diagnostic association cues for interpreting oil-monitoring changes around the system-level alarm.

Intraoperative identification of early ectopic pregnancy tissue remains challenging because chorionic villi often exhibit low visual contrast from surrounding inflamed or hemorrhagic tissue under conventional white-light endoscopy. To address this limitation, we developed a compact dual-channel visible-light fluorescence endoscopic imaging system configured as a modular 10 mm rigid endoscopic platform compatible with standard 10 mm laparoscopic workflows for real-time dual-channel acquisition and fusion. The system integrates a narrow-band 490 nm LED excitation source with a customized long-pass dichroic beam-splitting module to simultaneously acquire targeted fluorescence at ∼515 nm and diffuse reflectance at 490 nm. The two image streams are synchronized and fused to provide real-time molecular–anatomical visualization. Using a FITC-labeled anti- β -hCG antibody as a model targeted probe, we evaluated the system on 14 ex vivo human ectopic pregnancy specimens. The system achieved high-contrast visualization of villous tissue, with a target-to-background ratio of 4.17 ± 0.59. Mean fluorescence intensity was significantly higher in villous tissue than in adjacent non-villous tissue (115.76 ± 21.31 vs. 28.29 ± 6.74, P < 0.001). High-fluorescence regions showed close spatial correspondence with the pathological ground truth confirmed by H&E and immunohistochemistry. These findings demonstrate the feasibility of combining a compact visible-light dual-channel endoscopic platform with a targeted fluorescent probe for ex vivo molecular imaging of ectopic pregnancy tissue, and support future studies on translational optimization, safety qualification, and in vivo validation.

Squeezed states tuned to alkali atomic transitions are essential for developing atom–light quantum interfaces. We experimentally demonstrate a frequency-tunable quadrature squeezed state across the hyperfine transition of F =4(6 ² S _1/2 )→ F ^′ =3(6 ² P _1/2 ) of a cesium ( ¹³³ Cs) D ₁ line at 894.6 nm. Using a compact semi-monolithic optical parametric oscillator (OPO) operating below threshold, we achieve up to −4.1 dB squeezing below the standard quantum limit. When reconfigured as an optical parametric amplifier (OPA), the system generates bright squeezed light that can be finely tuned over a 50 MHz range, which is verified using electromagnetically induced transparency (EIT) spectroscopy. This portable and tunable quantum light source provides a promising platform for cesium-based quantum interfaces in emerging quantum networks.