The article concerns the development of an intelligent information system for forecasting components of medical projects. The purpose of the study is to propose an intelligent information system for forecasting component medical projects, which is based on the use of neural network models, as well as statistical and expert methods, which, unlike existing ones, ensures the accuracy of forecasting component medical projects, adaptability to changes in their project environment, as well as accessibility for users. The task of the research is to substantiate the architecture and develop an algorithm for the operation of an intelligent information system for forecasting component medical projects, as well as to develop the user interface of this system and to carry out forecasting of component medical projects. The object of research is decision support processes in medical project management. The subject of the study is the architecture of an intelligent information system for forecasting the components of medical projects, which defines the model, structure, functions, and relationships between its components. The scientific novelty is to substantiate the algorithm and architecture of an intelligent information system for forecasting the components of medical projects based on the use of dynamic data (modern medical records, medical information system server, and other external data), which ensures the formation of a large database underlying the training of neural network models and ensuring high accuracy of forecasting the components of medical projects. The proposed intelligent information system is an effective tool that can be used to increase the accuracy of management decisions and the effectiveness of medical projects. The system involves the use of a medical information system for the formation of a historical database, which ensures the formation of a knowledge base and the development of a set of systematically interconnected blocks. The developed algorithm of the proposed intelligent information system involves the implementation of 17 steps, which reflect the intellectual approach, which involves the use of neural networks for forecasting the components of medical projects. A user interface of an intelligent information system for evaluating component medical projects has been developed, which involves the use of 6 tabs. The use of dialog boxes for forecasting components of 5 groups of projects is foreseen. These include projects for the creation of hospital districts, projects of highly specialized medicine, projects of specialized medicine, projects of primary medicine, and advisory and diagnostic projects. A proposed neural network for predicting the duration of diabetes treatment projects in children. It is a deep neural network with two levels, which provides forecast accuracy at the level of 95.4 %, which indicates its sufficient efficiency and feasibility of use in an intelligent information system. Established trends in the duration of diabetes treatment in children due to changes in the main factors that cause them. The obtained results depend based on improving the quality and accuracy of decision support for assessing the duration of diabetes treatment in children with different states of their disease