Sonification of medical images based on statistical descriptors
Loading...
Date
2015
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
We propose an image sonification approach based on statistical descriptors, with a special emphasis on medical images. These descriptors capture several characteristics of the image data and can be beneficial for a musically interesting sonification of the visual domain. We conducted an experiment where subjects were asked to evaluate the usefulness of an extra layer of sound in tasks of identification of malignant regions in mammograms. Our results show that the addition of a musical domain to the visual data can improve the efficiency and reliability of this task. We believe that this approach could yield interesting and promising results in fields such as the sonification of other kinds of medical images or multidimensional scientific data.