In this study, for enhancing the resistance of probiotics to environmental factors, we designed a microgel beads delivery system loaded with synbiotics. Multiple droplets of W₁/O/W₂ emulsions stabilized with zein–apple pectin hybrid nanoparticles (ZAHPs) acted as the inner “egg,” whereas a three-dimensional network of poly-L-lysine (PLL)-alginate-CaCl₂ (Ca) crosslinked gel layers served as the outermost “box.” ZAHPs with a mass ratio of 2:1 zein-to-apple pectin showed excellent wettability (three-phase contact angle = 89.88°). The results of the ζ-potentials and Fourier transform infrared spectroscopy demonstrate that electrostatic interaction forces and hydrogen bonding were the main forces involved in the formation of ZAHPs. On this basis, we prepared W₁/O/W₂ emulsions with other preparation parameters and observed their microstructures by optical microscopy and confocal laser scanning microscope. The multi-chambered structures of W₁/O/W₂ emulsions were successfully visualized. Finally, the W₁/O/W₂ emulsions were coated with PLL-alginate-Ca using the solution extrusion method. The results of the in vitro colonic digestion stage reveal that the survival rate of probiotics in the microgel beads was about 75.11%, which was significantly higher than that of the free. Moreover, probiotics encapsulated in microgel beads also showed positive storage stability. Apple pectin would serve as both an emulsifier and a prebiotic. Thus, the results indicate that the “egg-box” shaped microgel beads, designed on the basis of pH-sensitive and enzyme-triggered mechanisms, can enhance the efficiency of probiotics translocation in the digestive tract and mediate spatiotemporal controlled release.

Background: The rapid advancement of radiomics and artificial intelligence (AI) technology has provided novel tools for the diagnosis of esophageal cancer. This study innovatively combines muscle imaging features with conventional esophageal imaging features to construct deep learning diagnostic models. Methods: This retrospective study included 1066 patients undergoing radical esophagectomy. Preoperative computed tomography (CT) images covering esophageal, stomach, and muscle (bilateral iliopsoas and erector spinae) regions were segmented automatically with manual adjustments. Diagnostic models were developed using deep learning (2D and 3D neural networks) and traditional machine learning (11 algorithms with PyRadiomics-derived features). Multimodal features underwent Principal Component Analysis (PCA) for dimension reduction and were fused for final analysis. Results: Comparative analysis of 1066 patients’ CT imaging revealed the muscle-based model outperformed the esophageal plus stomach model in predicting N2 staging (0.63 ± 0.11 vs. 0.52 ± 0.11, p = 0.03). Subsequently, multimodal fusion models were established for predicting pathological subtypes, T staging, and N staging. The logistic regression (LR) fusion model showed optimal performance in predicting pathological subtypes, achieving accuracy (ACC) of 0.919 in the training set and 0.884 in the validation set. For predicting T staging, the support vector machine (SVM) model demonstrated the highest accuracy, with training and validation accuracies of 0.909 and 0.907, respectively. The multilayer perceptron (MLP) fusion model achieved the best performance among all models tested for N staging prediction, although the accuracy remained moderate (ACC = 0.704 in the training set and 0.685 in the validation set), indicating potential for further optimization. Fusion models significantly outperformed single-modality models. Conclusions: Based on CT imaging data from 1066 patients, this study systematically constructed predictive models for pathological subtypes, T staging, and N staging of esophageal cancer. Comparative analysis of models using esophageal, esophageal plus stomach, and muscle modalities demonstrated that muscle imaging features contribute to diagnostic accuracy. Multimodal fusion models consistently showed superior performance.