Assessment of Cognitive Impairment Syndromes using Automated, Dynamical Analyses of Physiologic Signals

点击翻译按钮获取中文摘要

Risk assessment for cognitive impairment syndromes, culminating in Alzheimer's disease and related dementias, is the major public health priority addressed by this special notice supplement to our ongoing NIBIB-sponsored PhysioNet project, “Resource for Complex Physiologic Signals.” A key motivating knowledge gap is the lack of reliable, quantitative and non-invasive methods that complement expensive brain imaging technologies and semi-quantitative functional tests, as well as emerging biochemical probes. An important insight into the pathophysiology of cognitive impairment syndromes derives from increasing evidence linking central nervous system (CNS) dysfunction to alterations in the function of other organ systems including the regulation of the heartbeat. A major focus of PhysioNet has been on the development and dissemination of open source algorithms quantifying information encoded in the dynamics of physiologic signals. We have developed widely used tools such as multiscale entropy, cardiopulmonary coupling and, more recently, heart rate fragmentation, along with the physiologic frameworks that support their applications. However, the potential value of these tools in assessing and predicting cognitive decline have yet to be established in large studies. We now propose to quantify the value of dynamical metrics of relevant physiologic signals in this context. We focus on heart rate, respiration and electroencephalographic (EEG) fluctuations and their interactions, using data from the Multi-Ethnic Study of Atherosclerosis (MESA, ongoing), the Sleep Heart Health Study (SHHS) and the Osteoporotic Fractures (MrOS). In preliminary MESA analyses, we find significant associations between altered measures of heart rate dynamics and cognitive decline. We also find that participants with the dynamical heart rate signature of central sleep apnea have a significantly higher risk of experiencing cognitive decline over the same period. Specifically, our working hypotheses are that higher heart rate fragmentation, lower multiscale heart rate complexity, higher amounts of low-frequency cardiopulmonary coupling and lower multiscale complexity of EEG are associated with decreased cognitive performance, greater cognitive decline and more incident dementia. We will pursue hypothesized links between neuroautonomic dysfunction and cognitive impairment by means of the following specific aims: 1) in cross-sectional analyses of MESA, quantify the associations of EEG- and ECG-derived dynamical indices with standard measures of cognitive function; 2) in prospective analysis of MESA, quantify the associations of EEG and ECG-derived dynamical indices with: i) standard measures of cognitive function; ii) changes in cognitive function measures, and iii) incident dementia; and 3) validate the findings in the SHHS and MrOS cohorts. The proposed analyses directly align with the NIH strategic objective to develop new, noninvasive ways of detecting, predicting, and monitoring cognitive impairment and dementia.