EAPSI: Implementing and Analyzing 3-Dimensional Variants on Digital Neuropsychological Exams

Lara-Garduno Raniero A, College Station TX

Investigators

Abstract

The field of clinical neuropsychology has placed increasing emphasis in early detection of cognitively degenerative diseases such as Alzheimer's and Parkinson's. Existing research in digitizing the diagnosis process aims to help specialists bring their services to rural areas where access to neuropsychologists may be limited. This research intends to further explore new kinds of clinical neuropsychology exams designed to enrich collected data for computational analysis. The researcher will create a 3-dimensional variant of the existing popular Trail Making Test in neuropsychology. Existing normative data from paper-and-pencil Trail Making Tests show clear correlations between age and average test completion time among cognitively healthy users. The project will explore whether similar correlations can be found in this proposed variation among a wide distribution of age ranges. This project will be conducted at the University of Tokyo under the mentorship of Dr. Takeo Igarashi, one of the field's leading researchers in 3D computer graphics and Human-Computer Interaction. The test will be constructed for use in touch-enabled tablet devices, where patients will rotate a sphere comprised of labeled boxes and tap the next correct box in pre-determined sequences. The data points of most interest reside in the sphere manipulation behaviors as well as finger gesture data. Sphere manipulation would be captured by creating an invisible trail placed directly at the center of the screen's viewport. As the sphere rotates, the changing coordinates of the viewport as a function of time are stored, effectively recording where the patient was observing the sphere at any time. These coordinates will then be mapped to a 2D plane as treated as a digitized sketch, allowing machine learning techniques typically used in sketch recognition to be used in this context, with the intent to help classify the user's behavior in sphere manipulation per their cognitive function. This award, under the East Asia and Pacific Summer Institutes program, supports summer research by a U.S. graduate student and is jointly funded by NSF and the Japan Society for the Promotion of Science.