The system structure and method for long-term continuous monitoring and control of patient's health state are considered. In order to build such systems and provide biomedical research outside of a medical institution the following problems have to be solved: to form a complex of diagnostically significant indicators for health state abnormality detection; to develop methods of biomedical signals acquisition at patient's vital activity as well as algorithms of signals processing and indicators assessment. It is shown that these systems should operate in the intelligent monitoring mode. It involves a change in the number of active channels for picking up and recording of biomedical signals, and their characteristics, the number of signs used to identify episodes of exacerbation of a chronic disease, when diagnostically significant indicators of the disease go beyond the patient's individual norm.