Research project DMSS (Dementia Diagnosis and Management Support System) is a clinical decision support system, aimed at also supporting a continuing medical education.
DMSS main purpose is providing support for physicians in their diagnostic reasoning and choice of interventions when meeting new patients with a suspected dementia disease. Managing uncertain information and conflicting medical guidelines using state-of-the art artificial intelligence theories is one line of research, another is developing person-tailored support for reasoning and knowledge development in the medical professional.
Knut and Alice Wallenberg Foundation/Swedish Brainpower, 1.020.000 kr.
DMSS is a clinical decision support system that was developed in cooperation between the departments of Computing Science and Social Medicine and Rehabilitation, Geriatric division, Umeå University. The development was done in close cooperation with the Geriatric Centre at the University Hospital in Umeå, and health care centres in the region contribute in evaluations. The cooperation was extended through the research consortium Swedish Brain Power to include also Karolinska Institutet and Nordanstig health care centre among others in as a national effort to increase the quality of dementia care in Sweden. International cooperation include physicians, care providing organisations and ICT businesses in Japan, China and South Korea.
DMSS main functionality is support for physicians in their diagnostic reasoning when meeting new patients with a suspected dementia disease. Accurate and specific dementia diagnoses need to be assessed early in the disease progression in order to increase the quality of care. By providing the physician support in the gathering of data in the process of investigating a patient, and by providing the physician analyses of the evidence in a patient case, viewed using different contexts of interpretation (i.e. sets of evidence-based clinical guidelines), the diagnoses established by the physician can be supported and confirmed or questioned in an interactive process. The knowledge in the system is primarily based on the clinical guidelines DSM-IV, consensus guidelines for MCI, FTD and DLB. For providing support also in atypical patient cases, the guidelines NINCDS-AIRENS for VaD and NINCDS-ADRDA for AD is being integrated as well as the new criteria for AD.
Research focuses on the investigation and development of methods for representing the terminology and data structures, and interaction design for supporting reasoning processes. An ontology based on the clinical terminology SNOMED CT and the classification ICF was developed and integrated in DMSS. The modeling of the investigation process as whole is based on ACTclin, a semi-formal framework for assessing clinical knowledge to be integrated in decision support systems. The reasoning process as part of the investigation process was modelled using argumentation techniques for the purpose of improving the interactive clinical reasoning. This was done using ACKTUS.
DMSS-W is the web version built upon ACKTUS and is currently used in controlled clinical use for evaluation purposes in a number of clinics in Osaka prefecture, Japan (2016-2019). Based on user data, patterns of reasoning and decision making is extracted to form a base for person-tailored support.