Deep Isotonic Embedding Network: A flexible Monotonic Neural Network.

Guaranteeing the monotonicity of a learned model is crucial to address concerns such as fairness, interpretability, and generalization. This paper develops a new monotonic neural network named Deep Isotonic Embedding Network (DIEN), which uses different modules to deal with monotonic and non-monotonic features respectively, and then combine outputs of these modules linearly to obtain the prediction result. A new embedding tool called Isotonic Embedding Unit is developed to process monotonic features and turn each one into an isotonic embedding vector. By converting non-monotonic features into a series of non-negative weight vectors and then combining them with isotonic embedding vectors that have special properties, we enable DIEN to guarantee monotonicity. Besides, we also introduce a module named Monotonic Feature Learning Network to capture complex dependencies between monotonic features. This module is a monotonic feedforward neural network with non-negative weights and can handle scenarios where there are few non-monotonic features or only monotonic features. In comparison to existing methods, DIEN does not require intricate structures like lattices or the use of additional verification techniques to ensure monotonicity. Additionally, the relationship between DIEN's inputs and outputs is obvious and intuitive. Results from experiments on both synthetic and real-world datasets demonstrate DIEN's superiority over existing methodologies.

Hyaluronic acid encapsulated silver metal organic framework for the construction of a slow-controlled bifunctional nanostructure: Antibacterial and anti-inflammatory in intrauterine adhesion repair.

Intrauterine adhesion (IUA) is a common gynecological disease caused by endometrial injury, which might result in abnormal menstruation, miscarriage, and even fetal deaths. Nevertheless, existing treatment strategies such as intrauterine device and uterine cavity balloons only provide a physical barrier, and not circumvent inflammation of endometrial microenvironment and retrograde infection. In this study, a slow-controlled bifunctional nanostructure was developed via encapsulating hyaluronic acid (HA) on surface of silver-metal organic framework (Ag-MOF), and then loaded in poly lactic-co-glycolic acid scaffold to prevent IUA. In therapy, macro-molecule of HA provided anti-inflammatory function by the adjustment of signal transduction pathways of macrophage surface receptors, whereas Ag-MOF inactivated bacteria by destroying bacterial membrane and producing reactive oxygen. Significantly, the coated HA effectively avoided burst release of Ag+, thus achieving long-term antibacterial property and good biocompatibility. Antibacterial results showed antibacterial rate of the scaffold reached 87.8 % against staphylococcus aureus. Anti-inflammatory assays showed that the scaffold inhibited the release of inflammatory cytokines and promoted the release of anti-inflammatory cytokines. Moreover, in vitro cell tests revealed that the scaffold effectively inhibited fibroblast growth, indicating its good ability to prevent IUA. Taken together, the scaffold may be a promising candidate for IUA treatment.