In this paper, we propose a decentralized agent-based Autonomic Semantic Service Adaptation Controller and Reconfiguration (ASSACR) to solve the problem of context-aware service selection and deployment for distributed healthcare mobile applications, where functional and non-functional users' needs are expressed using domain-independent ontology. The main objective of this approach is to provide optimal personalized services as fast as possible. Indeed, a new ontology for semantic description and parallel management of the contextual services is presented. At first, a new dominance operator-based constraints violation degree is used to reduce the search space. Then, optimal personalized paths that meet user's needs, preferences and devices' availability are discovered and selected in a smaller search space more effectively and efficiently using dynamic services selection algorithm, based on new ASSACR multi-agent strategies. To validate the performance and efficiency of the proposed approach, we compared the proposed approach with existing multi-agent strategies of the diabetic follow-up use cases. Experimental results show that the proposed approach effectively addressed both the semantic service representation and agent aspects in terms of performance and quality service selection.