7. Conclusion and future development
Conclusion can be made that the rapidly advancing information technologies and emerging IoT technology have provided great op- portunities for developing smart healthcare information systems. Nevertheless, challenges still exist in achieving secure and effec- tive tele-healthcare applications. Some identified areas for future improvements are listed as follows:
(1) Self-learning and self-improvement. Facing the tremendous in- formation and great complexity, IoT itself cannot provide re- habilitation treatments or construct medical resources. Prompt and effective treatments must be made based on other two factors, quick diagnosis for patients, and creations of rehabil- itation treatments based on the diagnosis. Even with similar symptoms, the conditions of patients vary from one to an- other. All the factors have to be taken into account in or- der to generate an effective therapeutic regimen. A computer- aided tool relies merely on the data acquired by sensors and records of past similar cases, while self-learning methods can adaptively and intelligently diagnose and recommend the treat- ments. Some self-learning algorithms, such as Artificial Neu- ral Network (ANN), Genetic Algorithms (GA), Ant Colony Opti- mization (ACO), and Simulated Annealing (SA), can be applied to analyze data and mine knowledge. Besides, healthcare re- sources can be very dynamic due to reconfiguration, and pa- tients need to share the limited healthcare resources with the lowest cost and the highest efficiency. Topology- and ontology- based heuristic algorithms have demonstrated their power in finding optimal solutions for a large scale system.
