Organ chip database vectorization scheme for artificial intelligence algorithm

Abstract

The patent mainly describes a method suitable for organ chip database vectorization. The organ chip database comprises biological scaffold materials, reagents, cell lines, drugs, organ chip models and organ chip configuration parameters (concentrations of the reagents and the drugs, cell types and the like); a deep learning model weight matrix can be obtained through data training, and time information and experimental results (cell metabolite concentration, cell number and survival rate, PH, temperature, oxygen concentration, carbon dioxide concentration, TEER and air pressure of a microenvironment in a chip, whether drugs are added or not, and drug release rate and degradation rate) as label data; after the information is input, the model can automatically predict label data. The data in the database needs to be input into the model, data format conversion is firstly carried out, data types and formats stored in the organ chip database are not uniform, text information, digital information and even image information exist, and the information needs to be converted into vector information which can be recognized by a machine learning algorithm. In order to solve the problem, how to vectorize the data of the organ chip is designed.

Description

technical field [0001] The invention belongs to the field of biomedical engineering and computer science medical-engineering integration, and designs a method suitable for vectorization of an organ chip database. The organ-on-chip database contains bioscaffold materials, reagents, cell lines, drugs, organ-on-chip models, organ-on-chip configuration parameters (reagent and drug concentrations, cell types, etc.), as well as time information and experimental results (cellular metabolites) as label data. Concentration, number and survival rate of cells, PH, temperature, oxygen concentration, carbon dioxide concentration, TEER, air pressure, drug addition, drug release rate, degradation rate) of the microenvironment in the chip, deep learning can be obtained after training with these data Model weight matrix, when the above information is input, the model will automatically predict the label data. For the data in the database to be input into the model, the first thing to do is to...

AI Technical Summary

Problems solved by technology

[0003] During the process of organ chip culture and experimentation, a large amount of experimental data will be generated, but in previous studies, researchers did not carefully analyze the relationship between the data, especially the relationship between different organ chip experiments. , the data is not shared, so they can’t pay attention to the data correlation characteristics between different experiments. They just focus on the experimental results. The data in the experimental process, especially the dynamic data, is lost, and they only care about their own experimental data. There is no time and energy, and no tools to pay attention to the results of similar experiments done by others, and compare your own design parameters with others. Therefore, you need to develop appropriate data analysis methods to analyze and model these data