"Integration-of-Neural-Networks-and-Kernel-Based-SVM-for-Accurate-Cirrhosis-Detection-via-VOCs"

"AbstractLiver cirrhosis is a progressive and potentially fatal disease that requires early and accurate diagnosis to improve patient outcomes. Traditional diagnostic methods are often invasive, costly, and not suitable for continuous monitoring. In this study, a non-invasive approach for cirrhosis detection is proposed based on the analysis of volatile organic compounds (VOCs) present in human breath. A hybrid machine learning model combining an artificial neural network (ANN) and a kernel-based support vector machine (SVM) is developed to enhance classification performance.The proposed methodology consists of preprocessing VOC and clinical data, followed by feature extraction using a multilayer neural network. The learned feature representations are then used as input to a support vector machine with a radial basis function (RBF) kernel, enabling effective separation of complex and non-linear patterns. The dataset includes multiple VOC biomarkers such as acetone, ammonia, dimethyl sulfide, ethanol, and methane, along with demographic and clinical variables.Experimental results demonstrate that the hybrid model outperforms individual models by achieving higher accuracy and improved generalization capability. Visualization techniques, including ROC curves and three-dimensional feature projections, confirm the robustness of the classification. The results suggest that integrating neural networks with kernel-based SVM provides a powerful framework for non-invasive cirrhosis detection, offering potential applications in early diagnosis and real-time health monitoring systems. "