| Agent-Based Modeling technology is often used to study complex multicellular biological phenomena.It plays an important role in biomedical research.In recent years,speculating on the timing of cancer cell metastasis by establishing simulation models to study the process of cancer development has become an important research direction.Existing studies have shown that metastatic spread of cancer is the leading cause of death in cancer patients.For colorectal cancer,which has a high incidence,the liver is a common target organ for metastasis.Therefore,this article uses ABM technology to carry out the research on the liver metastasis process of colon cancer.This research can help assess the progression of the disease and provide more targeted diagnosis and treatment recommendations,which is of great significance for improving the prognosis of colon cancer.This article raises the following three scientific questions based on the investigation of the current global research status:(1)How to build a three-dimensional model that simulates the proliferation of colon tumors and the process of liver metastasis to improve the accuracy of model prediction and expand the scope of application of the model?(2)Can the model of liver metastasis of colon cancer be optimized so that the model can not only obtain the spatial information of cancer cells,but also reduce the computational complexity?(3)Is it possible to build a visualization platform that can display the liver metastasis process of colon cancer and can perform interactive operations for colon cancer metastasis prediction and risk assessment?This article has carried out related research on the above three scientific questions,and the main work content is as follows:(1)The establishment of the model.First,construct a three-dimensional model ISCIMET that can simulate the process of cell proliferation and metastasis in colon tumors.Secondly,real case data and particle swarm algorithm are used to estimate unknown parameters in the model during model establishment,and parallel computing technology is combined when estimating parameters,which speeds up the calculation of the objective function of particle swarm algorithm and reduces the time consumed for parameter estimation.Finally,survival analysis and error analysis were used to prove the validity and accuracy of the model.(2)Optimization of the model.First,a continous model describing the proliferation of colon tumors was constructed through machine learning algorithms.Then,the model is optimized by combining the continous model and the discrete model.Finally,experimental comparisons proved that the optimized liver metastasis model of colon cancer can not only easily obtain the spatial information of cancer cells during simulation,but also reduce the computational complexity of the model,thereby shortening the response time of the model simulation.(3)Construction of a visual analysis platform.A visual analysis platform for colon cancer liver metastasis was developed based on Web technology.The platform has three functional modules: platform introduction,visualization of tumor development,disease analysis and prediction.This platform uses visualization technology to visually display the changes in the number and spatial distribution of cancer cells in the process of colon tumor proliferation,and the time when cancer cells metastasize to the liver.The platform achieves the function of predicting the size of the current possible liver tumor for patients who have only found colon cancer;and for patients with liver metastases from colon cancer,it can evaluate the invasion of colon tumors and the risk of potential metastasis. |
PDF Full Text Request