3results about How to "Good clinical application value" patented technology

The present invention discloses an intestinal microbiota Bacteroides fragilis toxin-neutralizing nanobody and its application. By deeply studying the enterotoxigenic Bacteroides fragilis and the toxin BFT secreted by it in the bodies of patients with colorectal cancer and breast cancer, nanobodies against BFT are expressed by phage display technology; nanobodies with high binding affinity to the antigen are screened out through biopanning technology; a nanobody with the ability to neutralize the toxicity of BFT, also known as Nb2.43, is obtained. The present invention expresses and optimizes to obtain a BFT nanobody protein with high affinity, stability and uniformity, opening up a new field for the treatment of colorectal cancer and breast cancer with nanobodies targeting BFT, and having good research value and application prospects.

This invention relates to the field of medical biopolymer materials technology, specifically to an injectable hydrogel for kidney stone removal, its preparation method, and its application. The preparation materials include at least a precursor solution 1 and a precursor solution 2. Precursor solution 1 contains at least a polyethylene glycol derivative and a thickener, while precursor solution 2 contains at least a polyamino polymer. By optimizing the formulation ratio of the two-component precursor solutions, a hydrogel with strong conformability, good toughness, and viscosity is formed in situ in the kidney stone area. This encapsulates the kidney stone, facilitating its removal from the body and avoiding the safety hazards associated with small kidney stone fragments or conventional drug treatment.

The invention discloses a high-risk coronary artery stenosis prediction method based on multi-level structure information collaboration, and belongs to the technical field of intelligent medical aid decision making. The method comprises the following steps: firstly, constructing a clinical pathological structure perception graph to describe a structure constraint relationship among variables, and constructing a patient group risk conduction graph to model risk association of patients in a multi-dimensional phenotypic space; then structure perception representation learning is carried out on the two images, and feature-level and patient-level embedded representations are obtained; and then realizing cross-level semantic alignment through linear projection, introducing a learnable weight to perform structure enhancement on the two types of representations, inputting the fused joint representation into a classifier, and outputting a risk prediction result of the high-risk coronary artery stenosis of the patient. According to the method, through double-graph collaborative modeling and multi-level structure information fusion, effective integration of heterogeneous structure characterization is realized, and the prediction accuracy and stability are improved.