Adaptive resonance theory (ARTMAP) for Analysis and Prediction of Survival rate of Patient after Liver Transplantion
Gaurav Soni · 2022 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT) · 2022
In both industrialized and developing nations, chronic liver disease affects a significant number of individuals (CLD). Excessive drinking, exposure to toxic gases, and the eating of polluted food all contribute to a rise in the number of liver disease sufferers. Doctors have several challenges in treating patients with liver disease since it affects so many vital bodily systems. A liver transplant is a gruelling procedure with a high likelihood of complications after the procedure. The donor and recipient's compatibility is critical to the transplant's success. If a huge patient and donor database could be utilised to precisely match a donor recipient pair, the post-transplant mortality rate might be considerably reduced. Automated medical diagnostic systems often employ classification algorithms. Artificial neural networks (ANNs), a powerful technology, may aid in the discovery of patterns. It ranges from medicine to the arts that they work in. ANN-based dermatological tools and software may be useful in medicine, according to the conclusions of this study. It was used to analyse and study the Artificial Neural Networks, Radial Basis Function and ARTMAP. ILPD was obtained from the UCI machine learning library. It was possible to evaluate the outcomes using these several methods in order to determine if one method was more accurate than another in terms of precision, accuracy, mean absolute error, and other metrics (root MAE). The best method was found to be a multilayer perceptron (MLP) artificial neural network, which had a success rate of 98.9708 percent after a 10-fold cross validation. Other nations, such as the United States, have done more study on this area than India has. According to previous US dataset analyses, MLP was the best. Finally, we'll go through the procedure used to find out whether someone has Chronic Liver Disease. Long-term therapy for cancer may be predicted using artificial neural networks (ANN). With neural networks like ARTMAP, we want to develop a viable way for estimating patient survival following liver transplantation. Accuracy in ARTMAP calculations based on the evaluation of the degree of accuracy among the three models used in this study was found to be 58.2 percent.