Medicinal Plants Analysis for Chronic Diseases using Decision Tree Based U-Net Classification

Abstract

Chronic diseases provide important difficulties to world health, and there is a growing interest in the use of medicinal plants as alternative or complementary medicines aimed at addressing these challenges. Nevertheless, the problem of precisely identifying and categorising medicinal plants in terms of their effectiveness in the management of chronic diseases continues to be a difficult one. An novel method that makes use of decision tree-based U-Net classification is proposed in this research study for the purpose of analysing medicinal plants for chronic disorders. The methodology entails preprocessing photos of medicinal plants, extracting features through the utilisation of U-Net architecture, and categorising the extracted features through the utilisation of decision trees. To training the decision tree model, a dataset that contains photographs of a variety of medicinal plants that are recognised for their potential in the management of chronic diseases is utilised. The combination of deep learning with decision tree-based categorization for the purpose of analysing medicinal plants is the fundamental contribution that this research makes. Our technique delivers enhanced accuracy and interpretability in finding plants that are useful against chronic diseases. This is accomplished by integrating the strengths of both approaches. The proposed methodology is shown to be effective in accurately identifying medicinal plants based on their potential for treating chronic diseases, as showd by the results of the experiments. The U-Net classification, which is based on decision trees, achieves a high level of accuracy, exceeding more conventional classification methods. Moreover, the fact that the decision tree model may be interpreted makes it easier to recognise the essential characteristics that are connected to the effectiveness of medicinal plants.