The Proactive Screening and diagnosis of mild cognitive impairment and depression in patients aged 65 and over: An Implementation Study

点击翻译按钮获取中文摘要

Project Summary The proposed program aims to complete the work needed for Miro Health to finalize FDA approval for the automated diagnosis of amnestic MCI (aMCI) which often leads to Alzheimer’s and related dementias (ADRD), non-amnestic MCI (naMCI), and late life depression (LLD). The work involves the optimization of our data processing pipeline and digital signal processing methods and the refinement of our machine learning algorithms based on data types collected via real-world settings rather than through clinical research environments. Our objective is to improve patient outcomes and reduce healthcare costs by providing a universally available, self- administered mobile brain assessment and diagnostic platform. Prospective study participants will be recruited from clinics and the community. Participants will participate in novel mobile assessments and traditional cognitive and psychiatric assessments. The resulting data will be used to: (1) Assess usability of Miro for remote assessment; (2) Improve quantification of functional abilities and add RDoC metadata labels to Platform; (3) Refine A.I. models for the diagnosis of aMCI, naMCI, LLD, and comorbid MCI+LLD. Because late-life depression often mimics MCI and co-occurring depression may hasten the progression of MCI toward dementia, the identification and proper treatment of depression may resolve some apparent cases of MCI and slow the progression of others. Comorbid presentation of brain disorders is common and the increased availability of tools that can accurately and reliably identify complex comorbidities will improve care for our neediest patients. The work involved in this program will enhance Miro Health’s Mobile Research Platform. The introduction of RDoC metadata labels into our Mobile Research Platform’s metadata schema will help standardize clinical research, dramatically reduce the number of staff needed per study, and support massive scalability. Enhancing tools that yield precise, uniform data sets will improve the power, interpretability, and generalizability of studies examining disease etiology, genetic risk factors, and interventions.