Currently, neuropsychologists typically use paper-pencil tests to assess individual neuropsychological functions and brain dysfunctions, including memory, attention, reasoning and problem-solving. In addition, the aging of our population, with the increasing prevalence of physical and cognitive disorders, poses a major burden on our society with an estimated economic cost of 2.5 trillion US$ per year ( Trautmann, Rehm, and Wittchen 2016). Neurological and psychiatric disorders are among the most common and debilitating illnesses across the lifespan. Our AI-powered scoring system provides healthcare institutions worldwide with a digital tool to assess objectively, reliably and time-efficiently the performance in the ROCF test from hand-drawn images. The neural network outperforms both online raters and clinicians.
Unbiased ground truth ROCF scores were obtained from crowdsourced human intelligence. For this, a multi-head convolutional neural network was trained on 20225 ROCF drawings. Here, we leverage novel deep learning architectures to automatize the rating of memory deficits. This manual procedure is time-consuming, slow and scores vary depending on the clinician’s experience, motivation and tiredness. To obtain a score, a trained clinician inspects a patient’s ROCF drawing and quantifies deviations from the original figure.
The Rey-Osterrieth complex figure (ROCF) is the state–of-the-art assessment tool for neuropsychologists across the globe to assess the degree of non-verbal visual memory deterioration. Memory deficits are a hallmark of many different neurological and psychiatric conditions.
0 kommentar(er)
