Prototypes
Qadi - Quantitative Assessment of Dermoscopic Images
A web application developed to collect clinical diagnosis of specialists on dermoscopic images of pigmented skin lesions through Abcd Rule and 7 Point Checklist clinical algorithms. Go to Qadi application here.
DerMAT - Dermoscopic images Manual Annotation Tool
DerMAT allows building up a ground truth database with the manual segmentations both of pigmented skin lesions and of other regions of interest to be used in the assessment and validation of automatic segmentation and classification methods. The main functionalities of DerMAT are as following (see more details here):
- Image upload and display;
- Manual segmentation (of the lesion or other regions of interest);
- Region labeling;
- Boundary reshaping;
- A posteriori boundary edition;
- Multi-user ground truth annotation and segmentation comparison;
- Storage of segmented images.
ADPN - Automatic Detection of Pigment Network
This tool allows an automatic detection of the pigment network in dermoscopy images
of skin lesions. First, the dark lines associated to the pigment network are extracted
by using a bank of directional filters. Then, a post-processing algorithm (confirmation)
is applied, taking into account the topological properties of the detected regions.
The details can be found in the paper C. Barata, J. S. Marques, J. Rozeira, Detecting
the Pigment Network in Dermoscopy Images, IEEE EMBC, 2011.
Next Events
Computer vision and Machine learning events:
Dermoscopy events:
Visitors
ADDI Work
- Ongoing Work- Completed Work