A series of ML-based sketch recognition experiments is conducted, evaluating 10 modern off-the-shelf ML classifiers (i.e., SVMs, Deep Learning, etc.) on a sketch data set which we recorded with 40 subjects from a geriatrics daycare clinic. As part of our evaluation we examine how accurately different feature-based, supervised ML models can automatically score cognitive tests, with and without semantic content analysis. Second, the results of an automatic test analysis are presented to the physician in real-time, thereby reducing manual scoring effort and producing objective reports. ![]() The physician can, first, observe the sketching process in real-time on a mobile tablet, e.g., in telemedicine settings or to follow Covid-19 distancing regulations. We record these tests with a digital pen as part of a new interactive cognitive assessment tool with automatic analysis of pen input. Most cognitive assessments, for dementia screening for example, are conducted with a pen on normal paper. The feature set is evaluated in the use case of analyzing paper-pencil-based neurocognitive assessments in the medical domain. In this work, we use a state-of-the-art set of more than 170 digital pen features, which we implement and make publicly available. The multifaceted nature of the Rey–Osterrieth Complex Figure-containing configural elements, detached compound features, and texture-like components-coupled with the flexibility of the free-response drawing paradigm and the availability of standardized scoring systems, provides a promising method to probe peripheral perception and crowding.ĭigital pen features model characteristics of sketches and user behavior, and can be used for various supervised machine learning (ML) applications, such as multi-stroke sketch recognition and user modeling. Based on evaluation of the drawings, we also identified new error classes unique to peripheral presentation, including number errors for adjacent similar features and distinctive spatial distortions. Feature scores tended to decrease with increasing eccentricity, both within and between conditions, reflecting reduced resolution and increased crowding in peripheral vision. To assess the drawings, two raters used standard scoring systems that evaluated feature positions, spatial distortions, and omission errors. A foveal condition was used to measure copying performance in direct view. Participants were asked to draw the figure with free viewing, capturing its peripheral appearance. ![]() The figure was presented at 12° or 6° in the right visual field, with eye tracking to ensure that the figure was only presented when observers maintained stable fixation. ![]() We utilized a drawing technique to probe the phenomenology of peripheral vision, using the Rey–Osterrieth Complex Figure, a standard neuropsychological clinical instrument. Many quantitative aspects of this phenomenon have been characterized, but the specific nature of the perceptual degradation remains elusive. Peripheral vision is strongly limited by crowding, the deleterious influence of neighboring stimuli on target perception.
0 Comments
Leave a Reply. |