Seoul National University

• Addresses temporal data-resolution mismatch in greenhouses via diffusion framework • Generates biologically realistic data with environment–growth feedback model • Feedback guidance reduces errors by 69–90% across four crop growth variables • Guided feedback helps reduce the lab-to-field gap in agricultural AI applications Accurate crop growth forecasting is fundamental to precision agriculture, yet its application in real-world greenhouses faces challenges including mismatches in data acquisition frequencies. While automated environmental sensors generate high-density data streams containing millions of observations, manual growth measurements account for less than 1% of this temporal resolution. This data resolution mismatch often leads to “lab-to-field gap”, where models fail to generalise from benchmark datasets to operational settings. To address this challenge, this study introduces a feedback-guided temporal diffusion (FGTD) framework that generates synthetic growth data with data-driven causal consistency between environmental conditions and crop growth response. The FGTD framework integrates three core components: (1) a temporal diffusion model to learn from sparse growth patterns, (2) an environment-growth feedback network to enforce causal relationships, and (3) a guided reverse diffusion process to ensure biologically realistic outcomes. The framework was trained on a large-scale public dataset and validated through cross-domain testing on an independent operational tomato greenhouse dataset over the growing season with an even more severe data resolution mismatch. Ablation studies comparing FGTD against no augmentation, classical interpolation methods (linear, spline, Gaussian process regression), and standard diffusion revealed that standard diffusion improved predictions for continuous variables but failed on discrete variables like leaf count, increasing the MAE (Mean Absolute Error) by 37.3%. In contrast, the proposed feedback-guided approach achieved consistent improvements across all growth variables, reducing MAE by 74.3% for growth length, 69.2% for leaf length, and approximately 90% for both leaf width and leaf count relative to the no-augmentation baseline (leaf count bootstrap R² = 0.912 [95% CI: 0.760-0.996]; NRMSE = 0.019). Among the evaluated downstream forecasting models, XGBoost demonstrated optimal stability and accuracy when trained on the augmented dataset. These results demonstrate that environment-conditioned feedback guidance is essential for generating biologically faithful synthetic growth data, and that the FGTD framework provides a robust, scalable solution for bridging the data resolution gap toward practical deployment of AI-driven crop management systems.

Pathological platelet activation is central to cardiovascular disorders. Glycoprotein VI (GPVI), a pivotal collagen receptor, is a promising antithrombotic target. While earlier studies focused on the downstream effects of peroxisome proliferator-activated receptor γ (PPARγ) ligands, we recently identified the proximal PPARγ-Src family kinase (SFK) interaction as a key regulatory node. KR-62980 is a novel non-thiazolidinedione (non-TZD) PPARγ modulator designed to minimize traditional TZD-associated adverse effects. Here, we investigated whether KR-62980 operates through this "proximal mechanism." We found that KR-62980 dose-dependently inhibited collagen-induced human platelet aggregation in both washed platelets and platelet-rich plasma. Notably, KR-62980 exhibited a predominant sensitivity toward the collagen-GPVI axis compared to thrombin-induced protease-activated receptor signaling. Mechanistically, KR-62980 targeted the top of the GPVI signaling hierarchy by disrupting the physical and functional association between PPARγ and SFKs (Lyn and Fyn). This blockade suppressed SFK autophosphorylation and dismantled the LAT signalosome, preventing the recruitment of Gads, SLP-76, Btk, and PLCγ2. These findings confirm that the "proximal interaction" paradigm is a universal feature of PPARγ-mediated antiplatelet action, regardless of the agonist's chemical structure. In vivo, oral administration of KR-62980 significantly prolonged thrombotic occlusion time in a mouse carotid artery thrombosis model. Importantly, at therapeutic doses, KR-62980 did not significantly affect tail bleeding time, demonstrating a favorable safety profile with a wide therapeutic window. These findings confirm that the "proximal interaction" paradigm is a universal feature of PPARγ-mediated antiplatelet action and suggest that KR-62980 is a promising candidate for safe antithrombotic therapy.

The demand for portable real-time monitoring technologies for air quality and hazardous gas detection has significantly increased due to the growing interest in healthy living. In particular, nitrogen dioxide (NO 2 ) is a major contributor to air pollution in urban areas, and its concentration can fluctuate rapidly in a short period of time due to traffic emissions and industrial activities. Therefore, it is necessary to develop an ultra-sensitive sensor with fast response and recovery at room temperature. In this study, we demonstrate an NO 2 gas sensor operated at room temperature using Ag-decorated Sr 2 Nb 3 O 10 (ASNO) nanosheets, fabricated through a simple solution-based exfoliation process. This sensor exhibits excellent structural stability and an enhanced response of 935 % to NO 2 gas due to catalytic effect of the Ag decorated surface. It demonstrates fast response and recovery characteristics, which makes it a good candidate for reliable real-time monitoring and repeated use. Room-temperature NO 2 sensing using Ag-decorated 2D nanosheets supports sensitive and selective urban air pollution monitoring.