Research On Clinical Process Analyzing And Opti-mizing Technologies

Along with the development of medicine and technology, clinical process has become more and more specific and subtle. Clinical process analysis and optimization can help sort out the concept of medical practice that could be covered by scientific approaches, improve patient care services, and bring the best clinical prognosis through the scientific methods.The research on clinical process analyzing and optimizing technologies is becoming an important focus in clinical engineering and medical informatics domain. However, the characters, such as complexity and flexibility of clinical processes, result in the absence of real progress towards applying advances in information technology in order to analyze and optimize clinical processes. Especially, it lacks the scientific methods aimed at continuous optimization of clinical processes within clinical environmental changes. Considering the challenges above, this thesis introduces a detailed analysis of clinical processes, presents an in-depth study on the key technologies of clinical process analysis and optimization, i.e., process representation, execution control, medical knowledge retrieval and recommenda-tion, resource allocation and process analysis, and builds an initial technology framework of clinical processes analysis and optimization. The main work includes the following:An appropriate clinical process model is the key issue of the efficiency of clinical process practice. This thesis presents a hierarchical clinical process modeling mech-anism. It exploits the later binding modeling technique to support the content changes and the structure adjustments of clinical processes, and enhances the clinical process modeling flexibility.In order to guarantee the reliable clinical process execution, this thesis presents an adaptive clinical process execution mechanism, which can recommend appropriate medical interventions within the changes of patient state in clinical processes.Clinical practice heavily depends on the efficiency of knowledge retrieval and uti- lization. It requires to provide medical knowledge timely and accurately in clinical processes. This thesis proposes a clinical process oriented medical knowledge rec-ommendation mechanism, which can improve the efficiency of medical knowledge retrieval and assist medical staffs'work in clinical processes. The proposed solu-tion is easy to implement and can benefit medical staffs by facilitating collaboration work.Resource allocation is an integral part of clinical processes. It is widely acknowl-edged as being important for the efficiency of clinical processes. This thesis intro-duces a reinforcement learning based clinical resource allocation mechanism, which can optimize resource allocation and improve clinical process performance.In clinical practice, information about patient care are regularly recorded in vari-ous medical information systems and stored in clinical databases. Those information express the medical behaviors of clinical processes. This thesis proposes a process mining based clinical process analysis and evaluation mechanism in order to sup-port clinical process (re)design and optimization through the collection, mining and analysis of clinical process execution information.The efficiency and usability of the proposed approaches is validated by a set of experi-ments in specific clinical departments. The experimental results indicate that the proposed approaches can effectively solve the key challenges of clinical processes, optimize the per-formance of the executions of clinical processes, and improve the quality of patient care. By fusing the proposed research results, a clinical processes analyzing and optimizing tech-nology framework is designed and implemented. The methods and practice can favor the dissemination of clinical process analysis and optimization, and can be beneficial for med-ical staffs, patients and hospital managers.