Due to the high cost of replacing a power transformer, more and more attention is being paid to its monitoring and preventive maintenance. The first chapter provides an introduction to the issue of transformer insulation condition analysis, and offers a solution in the form of a unique value that describes this condition. We call this value the health index. This chapter also provides the main causes of transformer failures, and the components that are most affected by them. Finally, the fault diagnosis tests and their frequency are listed. The second chapter of this paper describes the parameters that most affect the health index of a power transformer. This group of parameters includes the analysis of dissolved gases, furan compounds, oil breakdown voltage, dielectric loss factor, water content in transformer oil and total acidity of transformer oil. The basics of artificial neural networks, together with the theory of predictive modeling, are described in the third chapter. Special attention is paid to the backpropagation algorithm, and classification and regression predictive modeling. The fourth chapter presents the analysis of the health index using the MATLAB software package and its built-in tool for creating artificial neural networks NNTOOL. The analysis was performed on a group of 50 transformers, whose input parameters and health indices were taken from the literature. After that, the results of the artificial neural network were compared with those taken from the literature, and the accuracy of the neural network in predicting the new value of the health index for transformers that were not used during its training was confirmed. The fifth chapter presents a conclusion on the results of the application of artificial neural networks in the analysis of the energy transformer health index.