A regression-based model for predicting the best mode of treatment for Egyptian liver cancer patients

In this study, we analyzed the data of 1427 Egyptian patients with liver cancer who were either treated by one of five different treatment methods or not treated. We proposed and compared between two pipelines, a Single-Model pipeline and a Multi-Model pipeline, for analyzing th e patient’s clinical and genetic data to recommend the best liver cancer treatment and, therefore, potentially improve their prognosis. We studied the performance of six regression methods in predicting the outcome of the treatment modalities for liver cancer patients. The best performing method w as used in building the models in the proposed pipelines. Our results show a difference in performance among different regression models, which proves the importance of choosing an appropriate one in decision-making, especially when dealing with important issues such as liver cancer treatment recomm endations. In our analysis, we also prove the crucial importance of genetic data and their effect on patients’ prognosis and response to treatment. Finally, this study signifies the great potential that data-mining methods could have in improving healthcare especially for serious diseases such as liver cancer.
Source: Network Modeling Analysis in Health Informatics and Bioinformatics - Category: Bioinformatics Source Type: research