Faculty of Science, Hong Kong Baptist University, Hongkong, China.
Biotechnol Genet Eng Rev. 2024 Oct;40(2):1193-1201. doi: 10.1080/02648725.2023.2193057. Epub 2023 Mar 26.
Due to a series of problems in the diagnosis of liver disease, the mortality rate of liver disease patients is very high. Therefore, it is necessary for doctors and researchers to find a more effective non-invasive diagnostic method to meet clinical needs. We analyzed data from 416 patients with liver disease and 167 patients without liver disease from northeastern Andhra Pradesh, India. On the basis of considering age, gender and other basic data of patients, this paper uses total bilirubin and other clinical data as parameters to build a diagnostic model. In this paper, the accuracy of artificial intelligence method Random Forest (RF) and Support Vector Machine (SVM) model in the diagnosis of liver patients was compared. The results show that the support vector machine model based on Gaussian kernel function is more excellent in diagnostic accuracy, that is, SVM method is more suitable for the diagnosis of liver diseases.
由于肝脏疾病诊断中存在一系列问题，肝病患者的死亡率非常高。因此，医生和研究人员有必要寻找一种更有效的非侵入性诊断方法来满足临床需求。我们分析了来自印度安得拉邦东北部的 416 名肝病患者和 167 名非肝病患者的数据。在考虑患者的年龄、性别和其他基本数据的基础上，本文使用总胆红素和其他临床数据作为参数来构建诊断模型。本文比较了人工智能方法随机森林（RF）和支持向量机（SVM）模型在肝病患者诊断中的准确性。结果表明，基于高斯核函数的支持向量机模型在诊断准确性方面更为出色，即 SVM 方法更适合于肝脏疾病的诊断。
