Main content

Home

Menu

Loading wiki pages...

View
Wiki Version:
Citation: Klein, Arno. (2013). “fundus_evaluation.” Open Science Framework. Accessed June 17, 2016. https://osf.io/r95wb/ This evaluation was run by Mindboggle's evaluate_features.py (https://github.com/nipy/mindboggle/blob/master/mindboggle/evaluate/evaluate_features.py). Since there exists no ground truth for fundus curves, we must resort to other means of evaluation. We leave it to future work to determine their utility for practical applications such as diagnosis and prediction of disorders. Since the DKT labeling protocol defines many of its anatomical label boundaries along approximations of fundus curves, we used the manually edited anatomical label boundaries in the Mindboggle-101 data set as gold standard data to evaluate the positions of fundi extracted by four different algorithms in 2013. Specifically, for each of the 48 fundi/sulci defined by the DKT protocol, we computed the mean of the minimum distances from the label boundary vertices in the sulcus to the fundus vertices in the sulcus, as well as from the fundus vertices in the sulcus to the label boundary vertices in the sulcus. The algorithms included Mindboggle’s default connect_points_erosion function described above, Forrest Bao’s minimum spanning tree algorithm (Bao et al., 2011), Gang Li’s algorithm (Li et al., 2010), and an algorithm in the BrainVisa software (Cointepas et al., 2001). The final algorithm was omitted from the results because too few fundi were extracted to make an adequate comparison. While there was no clear winner, we can summarize our comparison by computing the mean distance between fundi and label boundaries across all sulci for the three methods and by tallying how many sulci had the smallest mean distance among the methods. When measured from label boundaries to fundi, Gang Li’s and Mindboggle’s fundi were closer than were Forrest Bao’s (mean distances of 2.09mm and 2.38mm vs. 3.65mm, respectively; 25 and 21 vs. 2 closest sulci), whereas when measured from fundi to label boundaries, Forrest Bao’s fundi were closer than were Mindboggle’s or Gang Li’s (mean distances of 3.33mm vs. 4.06mm and 4.65mm, respectively; 41 vs. 5 and 2 closest sulci). When measuring from either direction, the maximum distances averaged across all sulci were higher for Forrest Bao’s fundi (11.65mm and 11.61mm) than for Mindboggle’s (10.84mm and 9.75mm) or Gang Li’s (11.12mm and 6.87mm).
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
Accept
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.
Accept
×

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.